diff --git a/.gitattributes b/.gitattributes
index a6344aac8c09253b3b630fb776ae94478aa0275b..c90db0db984ef4d57c5b8a4f15c24b655e8a0591 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+checkpoint-141839/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-283678/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+checkpoint-354597/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+generated_predictions.txt filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
diff --git a/README.md b/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..95089a843583a5df2129a37658fc45d355855552
--- /dev/null
+++ b/README.md
@@ -0,0 +1,74 @@
+---
+language:
+- zh
+- ko
+base_model: facebook/mbart-large-50-many-to-many-mmt
+tags:
+- generated_from_trainer
+metrics:
+- bleu
+model-index:
+- name: zhko_mbartLarge_100p_run1
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# zhko_mbartLarge_100p_run1
+
+This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.8988
+- Bleu: 42.7907
+- Gen Len: 13.7941
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 2
+- eval_batch_size: 2
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- gradient_accumulation_steps: 2
+- total_train_batch_size: 16
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.01
+- num_epochs: 15
+- mixed_precision_training: Native AMP
+
+### Training results
+
+| Training Loss | Epoch | Step   | Validation Loss | Bleu    | Gen Len |
+|:-------------:|:-----:|:------:|:---------------:|:-------:|:-------:|
+| 0.9719        | 1.0   | 70919  | 0.9435          | 40.8492 | 13.8028 |
+| 0.7537        | 2.0   | 141839 | 0.8988          | 42.7907 | 13.7941 |
+| 0.5973        | 3.0   | 212758 | 0.9143          | 43.4697 | 13.6556 |
+| 0.4873        | 4.0   | 283678 | 0.9758          | 43.6153 | 13.6544 |
+| 0.388         | 5.0   | 354597 | 1.0400          | 43.3763 | 13.6067 |
+
+
+### Framework versions
+
+- Transformers 4.35.2
+- Pytorch 2.1.1+cu121
+- Datasets 2.15.0
+- Tokenizers 0.15.0
diff --git a/all_results.json b/all_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..7786032532063877e28671b4fc654f91c19d7001
--- /dev/null
+++ b/all_results.json
@@ -0,0 +1,22 @@
+{
+    "epoch": 5.0,
+    "eval_bleu": 42.7907,
+    "eval_gen_len": 13.7941,
+    "eval_loss": 0.8987888693809509,
+    "eval_runtime": 10014.5277,
+    "eval_samples": 141838,
+    "eval_samples_per_second": 14.163,
+    "eval_steps_per_second": 1.77,
+    "predict_bleu": 42.8094,
+    "predict_gen_len": 13.7545,
+    "predict_loss": 0.89827561378479,
+    "predict_runtime": 9974.8511,
+    "predict_samples": 141838,
+    "predict_samples_per_second": 14.22,
+    "predict_steps_per_second": 1.777,
+    "train_loss": 0.7189551201696451,
+    "train_runtime": 324742.1955,
+    "train_samples": 1134707,
+    "train_samples_per_second": 52.413,
+    "train_steps_per_second": 3.276
+}
\ No newline at end of file
diff --git a/checkpoint-141839/config.json b/checkpoint-141839/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..c7d123d026fa1d1679aeb1220d9ae316d7c13a0b
--- /dev/null
+++ b/checkpoint-141839/config.json
@@ -0,0 +1,59 @@
+{
+  "_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "early_stopping": true,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 250014,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_length": 200,
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_beams": 5,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tokenizer_class": "MBart50Tokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.35.2",
+  "use_cache": true,
+  "vocab_size": 250054
+}
diff --git a/checkpoint-141839/generation_config.json b/checkpoint-141839/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..7e47c12e3900189593d4b56d0d776b58a7a55627
--- /dev/null
+++ b/checkpoint-141839/generation_config.json
@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "early_stopping": true,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 250014,
+  "forced_eos_token_id": 2,
+  "max_length": 200,
+  "num_beams": 5,
+  "pad_token_id": 1,
+  "transformers_version": "4.35.2"
+}
diff --git a/checkpoint-141839/model.safetensors b/checkpoint-141839/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ce0d737acda6b9afef4df04505c044a395af9684
--- /dev/null
+++ b/checkpoint-141839/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:787d8f1a526506bcb42c60d3888520b5032131b63fc207871e65cb6cc98e53c8
+size 2444578688
diff --git a/checkpoint-141839/optimizer.pt b/checkpoint-141839/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..b25ed36cf8c40ca61e19e8be8d4d8cb621eb08fd
--- /dev/null
+++ b/checkpoint-141839/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acd0b7fd0dd2d035ce40a3e4569014e0cac6ac862d41f13fbbe3243a62be4981
+size 4887473903
diff --git a/checkpoint-141839/rng_state_0.pth b/checkpoint-141839/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..204477ae85f0992ec82d4ad5905a76babfec95ae
--- /dev/null
+++ b/checkpoint-141839/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:782426fa0f34829286ec0354111c62b1d522d5db2338b2e87085a36ca3c56bec
+size 15024
diff --git a/checkpoint-141839/rng_state_1.pth b/checkpoint-141839/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..dcd364648d99319d57b18aa4071ec93adeae340b
--- /dev/null
+++ b/checkpoint-141839/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b986360a2ded5f01640e0960b8f3114511afd7844eafe799f78c27c8eba55519
+size 15024
diff --git a/checkpoint-141839/rng_state_2.pth b/checkpoint-141839/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..f1f393aa12bfac8c7ffaed58b08d54403e869c5b
--- /dev/null
+++ b/checkpoint-141839/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae6979ec5dbb54570b20cc28ed2208b830ea173aabdd1eaeef42bb39383f31c9
+size 15024
diff --git a/checkpoint-141839/rng_state_3.pth b/checkpoint-141839/rng_state_3.pth
new file mode 100644
index 0000000000000000000000000000000000000000..18be524148338cc43fbe43bb664356bdee65a3fd
--- /dev/null
+++ b/checkpoint-141839/rng_state_3.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:856f6436a49b7119ba14c7616ecebe61b710758d1d9577e9f7d53fd4cdf36922
+size 15024
diff --git a/checkpoint-141839/scheduler.pt b/checkpoint-141839/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..2e01068bad4b034108db1d9e24b96b31b045c985
--- /dev/null
+++ b/checkpoint-141839/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c862e91d8df536f771de8881de2888162e0ee38ed437f74d8b767f5dcffb6d7
+size 1064
diff --git a/checkpoint-141839/sentencepiece.bpe.model b/checkpoint-141839/sentencepiece.bpe.model
new file mode 100644
index 0000000000000000000000000000000000000000..7a3f40a75f870bc1f21700cd414dc2acc431583c
--- /dev/null
+++ b/checkpoint-141839/sentencepiece.bpe.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051
diff --git a/checkpoint-141839/special_tokens_map.json b/checkpoint-141839/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92619141640d5fcbb4429807de2248352b0dca79
--- /dev/null
+++ b/checkpoint-141839/special_tokens_map.json
@@ -0,0 +1,69 @@
+{
+  "additional_special_tokens": [
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}
diff --git a/checkpoint-141839/tokenizer.json b/checkpoint-141839/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..d7c3ce71bb70639c3fb46702de9c8356f8e2f956
--- /dev/null
+++ b/checkpoint-141839/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d91c41f87c8dbce15b820b41232d0dcd26ba285c22362400d3dd771a711417d
+size 17110107
diff --git a/checkpoint-141839/tokenizer_config.json b/checkpoint-141839/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..733cf8031772d40c50da15c3fe56fe63f05c2a13
--- /dev/null
+++ b/checkpoint-141839/tokenizer_config.json
@@ -0,0 +1,528 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "ar_AR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250002": {
+      "content": "cs_CZ",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250003": {
+      "content": "de_DE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250004": {
+      "content": "en_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250005": {
+      "content": "es_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250006": {
+      "content": "et_EE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250007": {
+      "content": "fi_FI",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250008": {
+      "content": "fr_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250009": {
+      "content": "gu_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250010": {
+      "content": "hi_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250011": {
+      "content": "it_IT",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250012": {
+      "content": "ja_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250013": {
+      "content": "kk_KZ",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250014": {
+      "content": "ko_KR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250015": {
+      "content": "lt_LT",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250016": {
+      "content": "lv_LV",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250017": {
+      "content": "my_MM",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250018": {
+      "content": "ne_NP",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250019": {
+      "content": "nl_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250020": {
+      "content": "ro_RO",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250021": {
+      "content": "ru_RU",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250022": {
+      "content": "si_LK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250023": {
+      "content": "tr_TR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250024": {
+      "content": "vi_VN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250025": {
+      "content": "zh_CN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250026": {
+      "content": "af_ZA",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250027": {
+      "content": "az_AZ",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250028": {
+      "content": "bn_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250029": {
+      "content": "fa_IR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250030": {
+      "content": "he_IL",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250031": {
+      "content": "hr_HR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250032": {
+      "content": "id_ID",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250033": {
+      "content": "ka_GE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250034": {
+      "content": "km_KH",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250035": {
+      "content": "mk_MK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250036": {
+      "content": "ml_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250037": {
+      "content": "mn_MN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250038": {
+      "content": "mr_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250039": {
+      "content": "pl_PL",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250040": {
+      "content": "ps_AF",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250041": {
+      "content": "pt_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250042": {
+      "content": "sv_SE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250043": {
+      "content": "sw_KE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250044": {
+      "content": "ta_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250045": {
+      "content": "te_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250046": {
+      "content": "th_TH",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250047": {
+      "content": "tl_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250048": {
+      "content": "uk_UA",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250049": {
+      "content": "ur_PK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250050": {
+      "content": "xh_ZA",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250051": {
+      "content": "gl_ES",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250052": {
+      "content": "sl_SI",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250053": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "sp_model_kwargs": {},
+  "src_lang": "zh_CN",
+  "tgt_lang": "ko_KR",
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_token": "<unk>"
+}
diff --git a/checkpoint-141839/trainer_state.json b/checkpoint-141839/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..7033525463d778576b09db914b7deca2eb7dd98b
--- /dev/null
+++ b/checkpoint-141839/trainer_state.json
@@ -0,0 +1,1737 @@
+{
+  "best_metric": 0.8987888693809509,
+  "best_model_checkpoint": "./zhko_mbartLarge_100p_run1/checkpoint-141839",
+  "epoch": 2.0,
+  "eval_steps": 500,
+  "global_step": 141839,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.335965407031397e-06,
+      "loss": 2.879,
+      "step": 500
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6860312088738486e-06,
+      "loss": 2.0241,
+      "step": 1000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.0360970107162996e-06,
+      "loss": 1.8819,
+      "step": 1500
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.386162812558751e-06,
+      "loss": 1.8009,
+      "step": 2000
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.1736228614401204e-05,
+      "loss": 1.7072,
+      "step": 2500
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.4086294416243657e-05,
+      "loss": 1.682,
+      "step": 3000
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.643636021808611e-05,
+      "loss": 1.6188,
+      "step": 3500
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.878642601992856e-05,
+      "loss": 1.5802,
+      "step": 4000
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 2.1136491821771008e-05,
+      "loss": 1.5547,
+      "step": 4500
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 2.348655762361346e-05,
+      "loss": 1.5422,
+      "step": 5000
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2.5831923293852228e-05,
+      "loss": 1.5194,
+      "step": 5500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2.8181989095694684e-05,
+      "loss": 1.4815,
+      "step": 6000
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 3.053205489753713e-05,
+      "loss": 1.4509,
+      "step": 6500
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 3.288212069937958e-05,
+      "loss": 1.458,
+      "step": 7000
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 3.5227486369618354e-05,
+      "loss": 1.4415,
+      "step": 7500
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 3.757285203985712e-05,
+      "loss": 1.4284,
+      "step": 8000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 3.991821771009588e-05,
+      "loss": 1.4297,
+      "step": 8500
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.2268283511938334e-05,
+      "loss": 1.4307,
+      "step": 9000
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.461834931378079e-05,
+      "loss": 1.3971,
+      "step": 9500
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.696371498401955e-05,
+      "loss": 1.4106,
+      "step": 10000
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.9313780785862004e-05,
+      "loss": 1.3999,
+      "step": 10500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.998319322943521e-05,
+      "loss": 1.3937,
+      "step": 11000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.995945485293127e-05,
+      "loss": 1.3773,
+      "step": 11500
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.9935716476427316e-05,
+      "loss": 1.3621,
+      "step": 12000
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.991197809992337e-05,
+      "loss": 1.3552,
+      "step": 12500
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.988823972341943e-05,
+      "loss": 1.3312,
+      "step": 13000
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.986450134691549e-05,
+      "loss": 1.3331,
+      "step": 13500
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.984081044716455e-05,
+      "loss": 1.3259,
+      "step": 14000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.9817072070660604e-05,
+      "loss": 1.2983,
+      "step": 14500
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.979333369415666e-05,
+      "loss": 1.2828,
+      "step": 15000
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.976959531765271e-05,
+      "loss": 1.2793,
+      "step": 15500
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.9745856941148764e-05,
+      "loss": 1.2968,
+      "step": 16000
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.9722118564644826e-05,
+      "loss": 1.2774,
+      "step": 16500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.969838018814088e-05,
+      "loss": 1.2507,
+      "step": 17000
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.967464181163694e-05,
+      "loss": 1.2521,
+      "step": 17500
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.965090343513299e-05,
+      "loss": 1.2337,
+      "step": 18000
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.962716505862904e-05,
+      "loss": 1.2238,
+      "step": 18500
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.96034266821251e-05,
+      "loss": 1.2111,
+      "step": 19000
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.957968830562115e-05,
+      "loss": 1.2336,
+      "step": 19500
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.955599740587022e-05,
+      "loss": 1.1975,
+      "step": 20000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.9532259029366274e-05,
+      "loss": 1.2092,
+      "step": 20500
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.950852065286233e-05,
+      "loss": 1.2079,
+      "step": 21000
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.948482975311139e-05,
+      "loss": 1.1831,
+      "step": 21500
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.946109137660745e-05,
+      "loss": 1.1965,
+      "step": 22000
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.94373530001035e-05,
+      "loss": 1.1928,
+      "step": 22500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.9413614623599556e-05,
+      "loss": 1.1779,
+      "step": 23000
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.938987624709561e-05,
+      "loss": 1.1759,
+      "step": 23500
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.936613787059167e-05,
+      "loss": 1.1871,
+      "step": 24000
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.934239949408772e-05,
+      "loss": 1.1782,
+      "step": 24500
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.931866111758378e-05,
+      "loss": 1.1619,
+      "step": 25000
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.9294922741079834e-05,
+      "loss": 1.1615,
+      "step": 25500
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.927118436457589e-05,
+      "loss": 1.1476,
+      "step": 26000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.924749346482495e-05,
+      "loss": 1.1507,
+      "step": 26500
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.922380256507401e-05,
+      "loss": 1.1489,
+      "step": 27000
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.9200064188570064e-05,
+      "loss": 1.144,
+      "step": 27500
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.917632581206613e-05,
+      "loss": 1.1223,
+      "step": 28000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.915258743556218e-05,
+      "loss": 1.1319,
+      "step": 28500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.912884905905824e-05,
+      "loss": 1.1235,
+      "step": 29000
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.9105110682554286e-05,
+      "loss": 1.1408,
+      "step": 29500
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.908141978280335e-05,
+      "loss": 1.1299,
+      "step": 30000
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.905772888305242e-05,
+      "loss": 1.1172,
+      "step": 30500
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.903399050654847e-05,
+      "loss": 1.1316,
+      "step": 31000
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.9010299606797535e-05,
+      "loss": 1.1154,
+      "step": 31500
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.8986561230293584e-05,
+      "loss": 1.1162,
+      "step": 32000
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.896282285378964e-05,
+      "loss": 1.101,
+      "step": 32500
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.89390844772857e-05,
+      "loss": 1.1153,
+      "step": 33000
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.891534610078176e-05,
+      "loss": 1.1023,
+      "step": 33500
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.889160772427781e-05,
+      "loss": 1.0935,
+      "step": 34000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.886786934777387e-05,
+      "loss": 1.0898,
+      "step": 34500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.884413097126992e-05,
+      "loss": 1.0846,
+      "step": 35000
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.882039259476597e-05,
+      "loss": 1.0963,
+      "step": 35500
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.879665421826203e-05,
+      "loss": 1.0764,
+      "step": 36000
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.877291584175809e-05,
+      "loss": 1.0649,
+      "step": 36500
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.8749177465254145e-05,
+      "loss": 1.0935,
+      "step": 37000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.8725439088750194e-05,
+      "loss": 1.0739,
+      "step": 37500
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.870170071224625e-05,
+      "loss": 1.0779,
+      "step": 38000
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.8677962335742305e-05,
+      "loss": 1.083,
+      "step": 38500
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.865422395923836e-05,
+      "loss": 1.0729,
+      "step": 39000
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.8630485582734416e-05,
+      "loss": 1.0587,
+      "step": 39500
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.860679468298348e-05,
+      "loss": 1.078,
+      "step": 40000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.858305630647954e-05,
+      "loss": 1.0802,
+      "step": 40500
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.855931792997559e-05,
+      "loss": 1.0538,
+      "step": 41000
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.853557955347164e-05,
+      "loss": 1.0562,
+      "step": 41500
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.85118411769677e-05,
+      "loss": 1.0516,
+      "step": 42000
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.848810280046375e-05,
+      "loss": 1.0624,
+      "step": 42500
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.8464364423959815e-05,
+      "loss": 1.0587,
+      "step": 43000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.8440626047455864e-05,
+      "loss": 1.0544,
+      "step": 43500
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.841688767095192e-05,
+      "loss": 1.0467,
+      "step": 44000
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.8393291724707e-05,
+      "loss": 1.0613,
+      "step": 44500
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.836955334820306e-05,
+      "loss": 1.0365,
+      "step": 45000
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.834581497169911e-05,
+      "loss": 1.0424,
+      "step": 45500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.832207659519516e-05,
+      "loss": 1.0442,
+      "step": 46000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.829833821869122e-05,
+      "loss": 1.0354,
+      "step": 46500
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.8274789749199304e-05,
+      "loss": 1.0543,
+      "step": 47000
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.825105137269536e-05,
+      "loss": 1.0476,
+      "step": 47500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.8227312996191415e-05,
+      "loss": 1.0427,
+      "step": 48000
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.820357461968748e-05,
+      "loss": 1.0174,
+      "step": 48500
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.8179836243183526e-05,
+      "loss": 1.0325,
+      "step": 49000
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.815609786667958e-05,
+      "loss": 1.0126,
+      "step": 49500
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.813240696692865e-05,
+      "loss": 1.0335,
+      "step": 50000
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.81086685904247e-05,
+      "loss": 1.0219,
+      "step": 50500
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.808493021392075e-05,
+      "loss": 1.0291,
+      "step": 51000
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.8061191837416815e-05,
+      "loss": 1.0251,
+      "step": 51500
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.803745346091287e-05,
+      "loss": 1.01,
+      "step": 52000
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.8013715084408925e-05,
+      "loss": 1.0224,
+      "step": 52500
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.7989976707904974e-05,
+      "loss": 1.0163,
+      "step": 53000
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.796623833140103e-05,
+      "loss": 1.0214,
+      "step": 53500
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.7942499954897085e-05,
+      "loss": 1.0187,
+      "step": 54000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.791876157839314e-05,
+      "loss": 1.0091,
+      "step": 54500
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.78950232018892e-05,
+      "loss": 1.0256,
+      "step": 55000
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.787128482538525e-05,
+      "loss": 1.0025,
+      "step": 55500
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.784754644888131e-05,
+      "loss": 1.0013,
+      "step": 56000
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.782380807237736e-05,
+      "loss": 1.0098,
+      "step": 56500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.780006969587342e-05,
+      "loss": 1.017,
+      "step": 57000
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.777637879612248e-05,
+      "loss": 1.0114,
+      "step": 57500
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.775264041961854e-05,
+      "loss": 0.9977,
+      "step": 58000
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.7728902043114596e-05,
+      "loss": 1.0107,
+      "step": 58500
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.7705163666610644e-05,
+      "loss": 0.9869,
+      "step": 59000
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.76814252901067e-05,
+      "loss": 1.008,
+      "step": 59500
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.7657686913602755e-05,
+      "loss": 1.0025,
+      "step": 60000
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.763394853709881e-05,
+      "loss": 0.9964,
+      "step": 60500
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.7610210160594866e-05,
+      "loss": 0.9921,
+      "step": 61000
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.758647178409092e-05,
+      "loss": 0.9938,
+      "step": 61500
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.756273340758698e-05,
+      "loss": 0.9934,
+      "step": 62000
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.753899503108303e-05,
+      "loss": 1.0249,
+      "step": 62500
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.751530413133209e-05,
+      "loss": 0.9832,
+      "step": 63000
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.749156575482815e-05,
+      "loss": 1.001,
+      "step": 63500
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.74678273783242e-05,
+      "loss": 0.9755,
+      "step": 64000
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4.7444089001820266e-05,
+      "loss": 0.995,
+      "step": 64500
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.742035062531632e-05,
+      "loss": 0.9824,
+      "step": 65000
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.739661224881237e-05,
+      "loss": 0.9759,
+      "step": 65500
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.7372873872308425e-05,
+      "loss": 0.98,
+      "step": 66000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.734913549580448e-05,
+      "loss": 0.9905,
+      "step": 66500
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.7325397119300536e-05,
+      "loss": 0.9811,
+      "step": 67000
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.730165874279659e-05,
+      "loss": 0.9873,
+      "step": 67500
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.727792036629265e-05,
+      "loss": 0.9804,
+      "step": 68000
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.7254229466541714e-05,
+      "loss": 0.9695,
+      "step": 68500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.723049109003776e-05,
+      "loss": 0.9751,
+      "step": 69000
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.720675271353382e-05,
+      "loss": 0.9735,
+      "step": 69500
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.7183061813782885e-05,
+      "loss": 0.9832,
+      "step": 70000
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.7159370914031945e-05,
+      "loss": 0.9719,
+      "step": 70500
+    },
+    {
+      "epoch": 1.0,
+      "eval_bleu": 40.8492,
+      "eval_gen_len": 13.8028,
+      "eval_loss": 0.9435123801231384,
+      "eval_runtime": 10127.6159,
+      "eval_samples_per_second": 14.005,
+      "eval_steps_per_second": 1.751,
+      "step": 70919
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.7135632537528e-05,
+      "loss": 0.9687,
+      "step": 71000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.7111894161024056e-05,
+      "loss": 0.9639,
+      "step": 71500
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.708815578452011e-05,
+      "loss": 0.9669,
+      "step": 72000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7064417408016166e-05,
+      "loss": 0.956,
+      "step": 72500
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 4.704067903151222e-05,
+      "loss": 0.9563,
+      "step": 73000
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 4.701694065500828e-05,
+      "loss": 0.9564,
+      "step": 73500
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 4.699320227850433e-05,
+      "loss": 0.9555,
+      "step": 74000
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 4.696946390200039e-05,
+      "loss": 0.9408,
+      "step": 74500
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.6945725525496444e-05,
+      "loss": 0.9347,
+      "step": 75000
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.69219871489925e-05,
+      "loss": 0.9368,
+      "step": 75500
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 4.6898296249241566e-05,
+      "loss": 0.9401,
+      "step": 76000
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 4.6874557872737615e-05,
+      "loss": 0.9383,
+      "step": 76500
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.685081949623367e-05,
+      "loss": 0.9232,
+      "step": 77000
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.6827081119729726e-05,
+      "loss": 0.9046,
+      "step": 77500
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 4.680334274322578e-05,
+      "loss": 0.9236,
+      "step": 78000
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.6779604366721837e-05,
+      "loss": 0.9091,
+      "step": 78500
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.675586599021789e-05,
+      "loss": 0.9036,
+      "step": 79000
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 4.673212761371395e-05,
+      "loss": 0.909,
+      "step": 79500
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 4.670838923721e-05,
+      "loss": 0.9003,
+      "step": 80000
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.668469833745906e-05,
+      "loss": 0.8957,
+      "step": 80500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.666095996095512e-05,
+      "loss": 0.8928,
+      "step": 81000
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 4.66373640147102e-05,
+      "loss": 0.8842,
+      "step": 81500
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.6613625638206256e-05,
+      "loss": 0.885,
+      "step": 82000
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.658988726170231e-05,
+      "loss": 0.8821,
+      "step": 82500
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 4.656614888519836e-05,
+      "loss": 0.8769,
+      "step": 83000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 4.6542410508694416e-05,
+      "loss": 0.8681,
+      "step": 83500
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 4.651867213219048e-05,
+      "loss": 0.8767,
+      "step": 84000
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 4.6494933755686534e-05,
+      "loss": 0.8702,
+      "step": 84500
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 4.647119537918259e-05,
+      "loss": 0.8753,
+      "step": 85000
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 4.644745700267864e-05,
+      "loss": 0.8587,
+      "step": 85500
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 4.642371862617469e-05,
+      "loss": 0.8459,
+      "step": 86000
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 4.639998024967075e-05,
+      "loss": 0.853,
+      "step": 86500
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 4.6376241873166804e-05,
+      "loss": 0.8811,
+      "step": 87000
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 4.635255097341587e-05,
+      "loss": 0.8618,
+      "step": 87500
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 4.6328812596911926e-05,
+      "loss": 0.8493,
+      "step": 88000
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 4.630507422040798e-05,
+      "loss": 0.8603,
+      "step": 88500
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 4.628133584390403e-05,
+      "loss": 0.8333,
+      "step": 89000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 4.6257597467400086e-05,
+      "loss": 0.8386,
+      "step": 89500
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 4.623385909089614e-05,
+      "loss": 0.8424,
+      "step": 90000
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 4.6210120714392204e-05,
+      "loss": 0.8519,
+      "step": 90500
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 4.618638233788826e-05,
+      "loss": 0.8297,
+      "step": 91000
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 4.616264396138431e-05,
+      "loss": 0.8406,
+      "step": 91500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 4.613890558488036e-05,
+      "loss": 0.8392,
+      "step": 92000
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 4.611516720837642e-05,
+      "loss": 0.8269,
+      "step": 92500
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 4.6091476308625486e-05,
+      "loss": 0.8392,
+      "step": 93000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 4.6067785408874545e-05,
+      "loss": 0.8474,
+      "step": 93500
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 4.60440470323706e-05,
+      "loss": 0.8329,
+      "step": 94000
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 4.6020308655866656e-05,
+      "loss": 0.8327,
+      "step": 94500
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 4.599657027936271e-05,
+      "loss": 0.8435,
+      "step": 95000
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 4.597283190285877e-05,
+      "loss": 0.8343,
+      "step": 95500
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 4.5949141003107834e-05,
+      "loss": 0.8316,
+      "step": 96000
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 4.592540262660389e-05,
+      "loss": 0.8339,
+      "step": 96500
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 4.590166425009994e-05,
+      "loss": 0.8199,
+      "step": 97000
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 4.5877925873595994e-05,
+      "loss": 0.8162,
+      "step": 97500
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 4.585418749709205e-05,
+      "loss": 0.8239,
+      "step": 98000
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 4.5830496597341116e-05,
+      "loss": 0.8197,
+      "step": 98500
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 4.5806805697590176e-05,
+      "loss": 0.8164,
+      "step": 99000
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 4.578306732108623e-05,
+      "loss": 0.818,
+      "step": 99500
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 4.575932894458229e-05,
+      "loss": 0.8189,
+      "step": 100000
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 4.573559056807834e-05,
+      "loss": 0.8303,
+      "step": 100500
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 4.57118521915744e-05,
+      "loss": 0.8174,
+      "step": 101000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 4.568811381507045e-05,
+      "loss": 0.814,
+      "step": 101500
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 4.566437543856651e-05,
+      "loss": 0.8226,
+      "step": 102000
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 4.5640637062062564e-05,
+      "loss": 0.818,
+      "step": 102500
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 4.561689868555862e-05,
+      "loss": 0.8054,
+      "step": 103000
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 4.5593160309054675e-05,
+      "loss": 0.8191,
+      "step": 103500
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 4.556946940930374e-05,
+      "loss": 0.8134,
+      "step": 104000
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 4.554573103279979e-05,
+      "loss": 0.8077,
+      "step": 104500
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 4.5521992656295846e-05,
+      "loss": 0.8081,
+      "step": 105000
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 4.54982542797919e-05,
+      "loss": 0.7985,
+      "step": 105500
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 4.547451590328796e-05,
+      "loss": 0.8064,
+      "step": 106000
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 4.545077752678401e-05,
+      "loss": 0.8008,
+      "step": 106500
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 4.542703915028007e-05,
+      "loss": 0.7934,
+      "step": 107000
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 4.540330077377612e-05,
+      "loss": 0.7915,
+      "step": 107500
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 4.537956239727218e-05,
+      "loss": 0.8087,
+      "step": 108000
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 4.5355824020768234e-05,
+      "loss": 0.7941,
+      "step": 108500
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 4.533208564426429e-05,
+      "loss": 0.8032,
+      "step": 109000
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 4.5308347267760345e-05,
+      "loss": 0.8061,
+      "step": 109500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 4.5284608891256394e-05,
+      "loss": 0.7974,
+      "step": 110000
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 4.526091799150546e-05,
+      "loss": 0.7915,
+      "step": 110500
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 4.523722709175453e-05,
+      "loss": 0.7998,
+      "step": 111000
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 4.5213488715250576e-05,
+      "loss": 0.8014,
+      "step": 111500
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 4.518975033874663e-05,
+      "loss": 0.7848,
+      "step": 112000
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.516601196224269e-05,
+      "loss": 0.7914,
+      "step": 112500
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.5142321062491754e-05,
+      "loss": 0.795,
+      "step": 113000
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.5118630162740814e-05,
+      "loss": 0.7913,
+      "step": 113500
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.509489178623687e-05,
+      "loss": 0.7936,
+      "step": 114000
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.5071153409732924e-05,
+      "loss": 0.7932,
+      "step": 114500
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 4.504741503322898e-05,
+      "loss": 0.7835,
+      "step": 115000
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 4.502367665672504e-05,
+      "loss": 0.7942,
+      "step": 115500
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 4.499993828022109e-05,
+      "loss": 0.7771,
+      "step": 116000
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 4.4976199903717146e-05,
+      "loss": 0.7805,
+      "step": 116500
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 4.4952509003966206e-05,
+      "loss": 0.7952,
+      "step": 117000
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 4.492877062746226e-05,
+      "loss": 0.7799,
+      "step": 117500
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 4.490503225095832e-05,
+      "loss": 0.7932,
+      "step": 118000
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 4.488129387445438e-05,
+      "loss": 0.7919,
+      "step": 118500
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 4.4857555497950435e-05,
+      "loss": 0.7776,
+      "step": 119000
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 4.4833864598199495e-05,
+      "loss": 0.7731,
+      "step": 119500
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 4.481012622169555e-05,
+      "loss": 0.7778,
+      "step": 120000
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 4.47863878451916e-05,
+      "loss": 0.7638,
+      "step": 120500
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 4.476264946868766e-05,
+      "loss": 0.7819,
+      "step": 121000
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 4.473891109218372e-05,
+      "loss": 0.7717,
+      "step": 121500
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 4.471517271567977e-05,
+      "loss": 0.7803,
+      "step": 122000
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 4.469143433917583e-05,
+      "loss": 0.7807,
+      "step": 122500
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 4.4667695962671876e-05,
+      "loss": 0.7651,
+      "step": 123000
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 4.464395758616793e-05,
+      "loss": 0.7811,
+      "step": 123500
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 4.4620266686417e-05,
+      "loss": 0.7794,
+      "step": 124000
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 4.4596528309913054e-05,
+      "loss": 0.7763,
+      "step": 124500
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 4.457278993340911e-05,
+      "loss": 0.7757,
+      "step": 125000
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 4.4549051556905165e-05,
+      "loss": 0.7683,
+      "step": 125500
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 4.4525360657154225e-05,
+      "loss": 0.7764,
+      "step": 126000
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 4.450162228065028e-05,
+      "loss": 0.7661,
+      "step": 126500
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 4.4477883904146336e-05,
+      "loss": 0.7645,
+      "step": 127000
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 4.445414552764239e-05,
+      "loss": 0.7703,
+      "step": 127500
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 4.443040715113845e-05,
+      "loss": 0.7791,
+      "step": 128000
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 4.44066687746345e-05,
+      "loss": 0.7657,
+      "step": 128500
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 4.438293039813056e-05,
+      "loss": 0.7651,
+      "step": 129000
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 4.435919202162661e-05,
+      "loss": 0.7803,
+      "step": 129500
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 4.433545364512267e-05,
+      "loss": 0.7504,
+      "step": 130000
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 4.4311762745371735e-05,
+      "loss": 0.7785,
+      "step": 130500
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 4.4288024368867784e-05,
+      "loss": 0.7592,
+      "step": 131000
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 4.426428599236384e-05,
+      "loss": 0.7714,
+      "step": 131500
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 4.4240547615859895e-05,
+      "loss": 0.7652,
+      "step": 132000
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 4.421680923935595e-05,
+      "loss": 0.7608,
+      "step": 132500
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.4193070862852006e-05,
+      "loss": 0.7688,
+      "step": 133000
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.416933248634806e-05,
+      "loss": 0.79,
+      "step": 133500
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 4.414559410984412e-05,
+      "loss": 0.7526,
+      "step": 134000
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 4.412190321009318e-05,
+      "loss": 0.765,
+      "step": 134500
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 4.409816483358923e-05,
+      "loss": 0.7554,
+      "step": 135000
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 4.407442645708529e-05,
+      "loss": 0.7636,
+      "step": 135500
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 4.4050735557334354e-05,
+      "loss": 0.7582,
+      "step": 136000
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 4.402699718083041e-05,
+      "loss": 0.7504,
+      "step": 136500
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 4.400330628107947e-05,
+      "loss": 0.7591,
+      "step": 137000
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 4.3979567904575525e-05,
+      "loss": 0.7629,
+      "step": 137500
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 4.395582952807158e-05,
+      "loss": 0.7543,
+      "step": 138000
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 4.3932091151567636e-05,
+      "loss": 0.7634,
+      "step": 138500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 4.390835277506369e-05,
+      "loss": 0.7553,
+      "step": 139000
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 4.388461439855975e-05,
+      "loss": 0.7548,
+      "step": 139500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 4.38608760220558e-05,
+      "loss": 0.7554,
+      "step": 140000
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 4.383713764555186e-05,
+      "loss": 0.7539,
+      "step": 140500
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 4.3813399269047914e-05,
+      "loss": 0.7562,
+      "step": 141000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.378966089254396e-05,
+      "loss": 0.7537,
+      "step": 141500
+    },
+    {
+      "epoch": 2.0,
+      "eval_bleu": 42.7907,
+      "eval_gen_len": 13.7941,
+      "eval_loss": 0.8987888693809509,
+      "eval_runtime": 10044.1408,
+      "eval_samples_per_second": 14.121,
+      "eval_steps_per_second": 1.765,
+      "step": 141839
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 1063785,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 4.918136306736824e+18,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-141839/training_args.bin b/checkpoint-141839/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..dbbe22d1f032c0b0fd9eedfe9ae519ce9ccd36a7
--- /dev/null
+++ b/checkpoint-141839/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0adf1a980c6128833811b7e6eb546e117ffd3efb8c21dc7de95b5e76a5b21b8d
+size 4728
diff --git a/checkpoint-283678/config.json b/checkpoint-283678/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..c7d123d026fa1d1679aeb1220d9ae316d7c13a0b
--- /dev/null
+++ b/checkpoint-283678/config.json
@@ -0,0 +1,59 @@
+{
+  "_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "early_stopping": true,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 250014,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_length": 200,
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_beams": 5,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tokenizer_class": "MBart50Tokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.35.2",
+  "use_cache": true,
+  "vocab_size": 250054
+}
diff --git a/checkpoint-283678/generation_config.json b/checkpoint-283678/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..7e47c12e3900189593d4b56d0d776b58a7a55627
--- /dev/null
+++ b/checkpoint-283678/generation_config.json
@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "early_stopping": true,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 250014,
+  "forced_eos_token_id": 2,
+  "max_length": 200,
+  "num_beams": 5,
+  "pad_token_id": 1,
+  "transformers_version": "4.35.2"
+}
diff --git a/checkpoint-283678/model.safetensors b/checkpoint-283678/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..1c26786b712dd167cf27d25dbfd4ceab17f00742
--- /dev/null
+++ b/checkpoint-283678/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7d177f1e7493b532ebc4a1ffb3a41e55b3ecb907b70533411897f449b65ec4d
+size 2444578688
diff --git a/checkpoint-283678/optimizer.pt b/checkpoint-283678/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..a79d6a8be4b4435edfbbc46787be647ae3d21ad3
--- /dev/null
+++ b/checkpoint-283678/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8a229adfc30357fe66f4551a399a4369359fecbcb30285cc052cec1c69d75aa
+size 4887473903
diff --git a/checkpoint-283678/rng_state_0.pth b/checkpoint-283678/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..90ea071587d87dd218744ea5114c564b3fd8b0ca
--- /dev/null
+++ b/checkpoint-283678/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7aab6c9cf817dabab5a817aad8e8c0c027ecc31f80ee324829a1c30ad12c664
+size 15024
diff --git a/checkpoint-283678/rng_state_1.pth b/checkpoint-283678/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..9f8c0e3f5a27d8c02de745ab5c9b6d6ffbda8786
--- /dev/null
+++ b/checkpoint-283678/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7147cb5a020ed017a1409d253ffc4e524b7076e89b7ce45e07bc491c9282478b
+size 15024
diff --git a/checkpoint-283678/rng_state_2.pth b/checkpoint-283678/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d73eb3d68960dc5c755fda24dac4237caa30202b
--- /dev/null
+++ b/checkpoint-283678/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33ba14e4ce153d100896ad193ee1b3c6d4025f82859f4b9d697280b6cbbf489e
+size 15024
diff --git a/checkpoint-283678/rng_state_3.pth b/checkpoint-283678/rng_state_3.pth
new file mode 100644
index 0000000000000000000000000000000000000000..d848ebade4dc7a9452ab2fcf1ff8c735357f3158
--- /dev/null
+++ b/checkpoint-283678/rng_state_3.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68390f8e534ad10bc016408642d9d8145af981899f6818233529c940c786c6a1
+size 15024
diff --git a/checkpoint-283678/scheduler.pt b/checkpoint-283678/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..573f9c827ecad1ea6099d341e1084bb4646a2d37
--- /dev/null
+++ b/checkpoint-283678/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bed1f09e79aec328683ecbaa3c7f00d213503ab961ca4b0555a743ab9a30fe03
+size 1064
diff --git a/checkpoint-283678/sentencepiece.bpe.model b/checkpoint-283678/sentencepiece.bpe.model
new file mode 100644
index 0000000000000000000000000000000000000000..7a3f40a75f870bc1f21700cd414dc2acc431583c
--- /dev/null
+++ b/checkpoint-283678/sentencepiece.bpe.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051
diff --git a/checkpoint-283678/special_tokens_map.json b/checkpoint-283678/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92619141640d5fcbb4429807de2248352b0dca79
--- /dev/null
+++ b/checkpoint-283678/special_tokens_map.json
@@ -0,0 +1,69 @@
+{
+  "additional_special_tokens": [
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}
diff --git a/checkpoint-283678/tokenizer.json b/checkpoint-283678/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..d7c3ce71bb70639c3fb46702de9c8356f8e2f956
--- /dev/null
+++ b/checkpoint-283678/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d91c41f87c8dbce15b820b41232d0dcd26ba285c22362400d3dd771a711417d
+size 17110107
diff --git a/checkpoint-283678/tokenizer_config.json b/checkpoint-283678/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..733cf8031772d40c50da15c3fe56fe63f05c2a13
--- /dev/null
+++ b/checkpoint-283678/tokenizer_config.json
@@ -0,0 +1,528 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "ar_AR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250002": {
+      "content": "cs_CZ",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250003": {
+      "content": "de_DE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250004": {
+      "content": "en_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250005": {
+      "content": "es_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250006": {
+      "content": "et_EE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250007": {
+      "content": "fi_FI",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250008": {
+      "content": "fr_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250009": {
+      "content": "gu_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250010": {
+      "content": "hi_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250011": {
+      "content": "it_IT",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250012": {
+      "content": "ja_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250013": {
+      "content": "kk_KZ",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250014": {
+      "content": "ko_KR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250015": {
+      "content": "lt_LT",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250016": {
+      "content": "lv_LV",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250017": {
+      "content": "my_MM",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250018": {
+      "content": "ne_NP",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250019": {
+      "content": "nl_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250020": {
+      "content": "ro_RO",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250021": {
+      "content": "ru_RU",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250022": {
+      "content": "si_LK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250023": {
+      "content": "tr_TR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250024": {
+      "content": "vi_VN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250025": {
+      "content": "zh_CN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250026": {
+      "content": "af_ZA",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250027": {
+      "content": "az_AZ",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250028": {
+      "content": "bn_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250029": {
+      "content": "fa_IR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250030": {
+      "content": "he_IL",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250031": {
+      "content": "hr_HR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250032": {
+      "content": "id_ID",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250033": {
+      "content": "ka_GE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250034": {
+      "content": "km_KH",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250035": {
+      "content": "mk_MK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250036": {
+      "content": "ml_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250037": {
+      "content": "mn_MN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250038": {
+      "content": "mr_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250039": {
+      "content": "pl_PL",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250040": {
+      "content": "ps_AF",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250041": {
+      "content": "pt_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250042": {
+      "content": "sv_SE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250043": {
+      "content": "sw_KE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250044": {
+      "content": "ta_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250045": {
+      "content": "te_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250046": {
+      "content": "th_TH",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250047": {
+      "content": "tl_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250048": {
+      "content": "uk_UA",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250049": {
+      "content": "ur_PK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250050": {
+      "content": "xh_ZA",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250051": {
+      "content": "gl_ES",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250052": {
+      "content": "sl_SI",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250053": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "sp_model_kwargs": {},
+  "src_lang": "zh_CN",
+  "tgt_lang": "ko_KR",
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_token": "<unk>"
+}
diff --git a/checkpoint-283678/trainer_state.json b/checkpoint-283678/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..7446a9a953c737f5670df42df0db90c92445b868
--- /dev/null
+++ b/checkpoint-283678/trainer_state.json
@@ -0,0 +1,3461 @@
+{
+  "best_metric": 0.8987888693809509,
+  "best_model_checkpoint": "./zhko_mbartLarge_100p_run1/checkpoint-141839",
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 283678,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.335965407031397e-06,
+      "loss": 2.879,
+      "step": 500
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6860312088738486e-06,
+      "loss": 2.0241,
+      "step": 1000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.0360970107162996e-06,
+      "loss": 1.8819,
+      "step": 1500
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.386162812558751e-06,
+      "loss": 1.8009,
+      "step": 2000
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.1736228614401204e-05,
+      "loss": 1.7072,
+      "step": 2500
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.4086294416243657e-05,
+      "loss": 1.682,
+      "step": 3000
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.643636021808611e-05,
+      "loss": 1.6188,
+      "step": 3500
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.878642601992856e-05,
+      "loss": 1.5802,
+      "step": 4000
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 2.1136491821771008e-05,
+      "loss": 1.5547,
+      "step": 4500
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 2.348655762361346e-05,
+      "loss": 1.5422,
+      "step": 5000
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2.5831923293852228e-05,
+      "loss": 1.5194,
+      "step": 5500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2.8181989095694684e-05,
+      "loss": 1.4815,
+      "step": 6000
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 3.053205489753713e-05,
+      "loss": 1.4509,
+      "step": 6500
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 3.288212069937958e-05,
+      "loss": 1.458,
+      "step": 7000
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 3.5227486369618354e-05,
+      "loss": 1.4415,
+      "step": 7500
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 3.757285203985712e-05,
+      "loss": 1.4284,
+      "step": 8000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 3.991821771009588e-05,
+      "loss": 1.4297,
+      "step": 8500
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.2268283511938334e-05,
+      "loss": 1.4307,
+      "step": 9000
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.461834931378079e-05,
+      "loss": 1.3971,
+      "step": 9500
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.696371498401955e-05,
+      "loss": 1.4106,
+      "step": 10000
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.9313780785862004e-05,
+      "loss": 1.3999,
+      "step": 10500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.998319322943521e-05,
+      "loss": 1.3937,
+      "step": 11000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.995945485293127e-05,
+      "loss": 1.3773,
+      "step": 11500
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.9935716476427316e-05,
+      "loss": 1.3621,
+      "step": 12000
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.991197809992337e-05,
+      "loss": 1.3552,
+      "step": 12500
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.988823972341943e-05,
+      "loss": 1.3312,
+      "step": 13000
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.986450134691549e-05,
+      "loss": 1.3331,
+      "step": 13500
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.984081044716455e-05,
+      "loss": 1.3259,
+      "step": 14000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.9817072070660604e-05,
+      "loss": 1.2983,
+      "step": 14500
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.979333369415666e-05,
+      "loss": 1.2828,
+      "step": 15000
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.976959531765271e-05,
+      "loss": 1.2793,
+      "step": 15500
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.9745856941148764e-05,
+      "loss": 1.2968,
+      "step": 16000
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.9722118564644826e-05,
+      "loss": 1.2774,
+      "step": 16500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.969838018814088e-05,
+      "loss": 1.2507,
+      "step": 17000
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.967464181163694e-05,
+      "loss": 1.2521,
+      "step": 17500
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.965090343513299e-05,
+      "loss": 1.2337,
+      "step": 18000
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.962716505862904e-05,
+      "loss": 1.2238,
+      "step": 18500
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.96034266821251e-05,
+      "loss": 1.2111,
+      "step": 19000
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.957968830562115e-05,
+      "loss": 1.2336,
+      "step": 19500
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.955599740587022e-05,
+      "loss": 1.1975,
+      "step": 20000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.9532259029366274e-05,
+      "loss": 1.2092,
+      "step": 20500
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.950852065286233e-05,
+      "loss": 1.2079,
+      "step": 21000
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.948482975311139e-05,
+      "loss": 1.1831,
+      "step": 21500
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.946109137660745e-05,
+      "loss": 1.1965,
+      "step": 22000
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.94373530001035e-05,
+      "loss": 1.1928,
+      "step": 22500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.9413614623599556e-05,
+      "loss": 1.1779,
+      "step": 23000
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.938987624709561e-05,
+      "loss": 1.1759,
+      "step": 23500
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.936613787059167e-05,
+      "loss": 1.1871,
+      "step": 24000
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.934239949408772e-05,
+      "loss": 1.1782,
+      "step": 24500
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.931866111758378e-05,
+      "loss": 1.1619,
+      "step": 25000
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.9294922741079834e-05,
+      "loss": 1.1615,
+      "step": 25500
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.927118436457589e-05,
+      "loss": 1.1476,
+      "step": 26000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.924749346482495e-05,
+      "loss": 1.1507,
+      "step": 26500
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.922380256507401e-05,
+      "loss": 1.1489,
+      "step": 27000
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.9200064188570064e-05,
+      "loss": 1.144,
+      "step": 27500
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.917632581206613e-05,
+      "loss": 1.1223,
+      "step": 28000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.915258743556218e-05,
+      "loss": 1.1319,
+      "step": 28500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.912884905905824e-05,
+      "loss": 1.1235,
+      "step": 29000
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.9105110682554286e-05,
+      "loss": 1.1408,
+      "step": 29500
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.908141978280335e-05,
+      "loss": 1.1299,
+      "step": 30000
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.905772888305242e-05,
+      "loss": 1.1172,
+      "step": 30500
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.903399050654847e-05,
+      "loss": 1.1316,
+      "step": 31000
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.9010299606797535e-05,
+      "loss": 1.1154,
+      "step": 31500
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.8986561230293584e-05,
+      "loss": 1.1162,
+      "step": 32000
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.896282285378964e-05,
+      "loss": 1.101,
+      "step": 32500
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.89390844772857e-05,
+      "loss": 1.1153,
+      "step": 33000
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.891534610078176e-05,
+      "loss": 1.1023,
+      "step": 33500
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.889160772427781e-05,
+      "loss": 1.0935,
+      "step": 34000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.886786934777387e-05,
+      "loss": 1.0898,
+      "step": 34500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.884413097126992e-05,
+      "loss": 1.0846,
+      "step": 35000
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.882039259476597e-05,
+      "loss": 1.0963,
+      "step": 35500
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.879665421826203e-05,
+      "loss": 1.0764,
+      "step": 36000
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.877291584175809e-05,
+      "loss": 1.0649,
+      "step": 36500
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.8749177465254145e-05,
+      "loss": 1.0935,
+      "step": 37000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.8725439088750194e-05,
+      "loss": 1.0739,
+      "step": 37500
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.870170071224625e-05,
+      "loss": 1.0779,
+      "step": 38000
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.8677962335742305e-05,
+      "loss": 1.083,
+      "step": 38500
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.865422395923836e-05,
+      "loss": 1.0729,
+      "step": 39000
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.8630485582734416e-05,
+      "loss": 1.0587,
+      "step": 39500
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.860679468298348e-05,
+      "loss": 1.078,
+      "step": 40000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.858305630647954e-05,
+      "loss": 1.0802,
+      "step": 40500
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.855931792997559e-05,
+      "loss": 1.0538,
+      "step": 41000
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.853557955347164e-05,
+      "loss": 1.0562,
+      "step": 41500
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.85118411769677e-05,
+      "loss": 1.0516,
+      "step": 42000
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.848810280046375e-05,
+      "loss": 1.0624,
+      "step": 42500
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.8464364423959815e-05,
+      "loss": 1.0587,
+      "step": 43000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.8440626047455864e-05,
+      "loss": 1.0544,
+      "step": 43500
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.841688767095192e-05,
+      "loss": 1.0467,
+      "step": 44000
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.8393291724707e-05,
+      "loss": 1.0613,
+      "step": 44500
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.836955334820306e-05,
+      "loss": 1.0365,
+      "step": 45000
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.834581497169911e-05,
+      "loss": 1.0424,
+      "step": 45500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.832207659519516e-05,
+      "loss": 1.0442,
+      "step": 46000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.829833821869122e-05,
+      "loss": 1.0354,
+      "step": 46500
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.8274789749199304e-05,
+      "loss": 1.0543,
+      "step": 47000
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.825105137269536e-05,
+      "loss": 1.0476,
+      "step": 47500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.8227312996191415e-05,
+      "loss": 1.0427,
+      "step": 48000
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.820357461968748e-05,
+      "loss": 1.0174,
+      "step": 48500
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.8179836243183526e-05,
+      "loss": 1.0325,
+      "step": 49000
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.815609786667958e-05,
+      "loss": 1.0126,
+      "step": 49500
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.813240696692865e-05,
+      "loss": 1.0335,
+      "step": 50000
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.81086685904247e-05,
+      "loss": 1.0219,
+      "step": 50500
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.808493021392075e-05,
+      "loss": 1.0291,
+      "step": 51000
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.8061191837416815e-05,
+      "loss": 1.0251,
+      "step": 51500
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.803745346091287e-05,
+      "loss": 1.01,
+      "step": 52000
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.8013715084408925e-05,
+      "loss": 1.0224,
+      "step": 52500
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.7989976707904974e-05,
+      "loss": 1.0163,
+      "step": 53000
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.796623833140103e-05,
+      "loss": 1.0214,
+      "step": 53500
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.7942499954897085e-05,
+      "loss": 1.0187,
+      "step": 54000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.791876157839314e-05,
+      "loss": 1.0091,
+      "step": 54500
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.78950232018892e-05,
+      "loss": 1.0256,
+      "step": 55000
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.787128482538525e-05,
+      "loss": 1.0025,
+      "step": 55500
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.784754644888131e-05,
+      "loss": 1.0013,
+      "step": 56000
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.782380807237736e-05,
+      "loss": 1.0098,
+      "step": 56500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.780006969587342e-05,
+      "loss": 1.017,
+      "step": 57000
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.777637879612248e-05,
+      "loss": 1.0114,
+      "step": 57500
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.775264041961854e-05,
+      "loss": 0.9977,
+      "step": 58000
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.7728902043114596e-05,
+      "loss": 1.0107,
+      "step": 58500
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.7705163666610644e-05,
+      "loss": 0.9869,
+      "step": 59000
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.76814252901067e-05,
+      "loss": 1.008,
+      "step": 59500
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.7657686913602755e-05,
+      "loss": 1.0025,
+      "step": 60000
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.763394853709881e-05,
+      "loss": 0.9964,
+      "step": 60500
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.7610210160594866e-05,
+      "loss": 0.9921,
+      "step": 61000
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.758647178409092e-05,
+      "loss": 0.9938,
+      "step": 61500
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.756273340758698e-05,
+      "loss": 0.9934,
+      "step": 62000
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.753899503108303e-05,
+      "loss": 1.0249,
+      "step": 62500
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.751530413133209e-05,
+      "loss": 0.9832,
+      "step": 63000
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.749156575482815e-05,
+      "loss": 1.001,
+      "step": 63500
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.74678273783242e-05,
+      "loss": 0.9755,
+      "step": 64000
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4.7444089001820266e-05,
+      "loss": 0.995,
+      "step": 64500
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.742035062531632e-05,
+      "loss": 0.9824,
+      "step": 65000
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.739661224881237e-05,
+      "loss": 0.9759,
+      "step": 65500
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.7372873872308425e-05,
+      "loss": 0.98,
+      "step": 66000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.734913549580448e-05,
+      "loss": 0.9905,
+      "step": 66500
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.7325397119300536e-05,
+      "loss": 0.9811,
+      "step": 67000
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.730165874279659e-05,
+      "loss": 0.9873,
+      "step": 67500
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.727792036629265e-05,
+      "loss": 0.9804,
+      "step": 68000
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.7254229466541714e-05,
+      "loss": 0.9695,
+      "step": 68500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.723049109003776e-05,
+      "loss": 0.9751,
+      "step": 69000
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.720675271353382e-05,
+      "loss": 0.9735,
+      "step": 69500
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.7183061813782885e-05,
+      "loss": 0.9832,
+      "step": 70000
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.7159370914031945e-05,
+      "loss": 0.9719,
+      "step": 70500
+    },
+    {
+      "epoch": 1.0,
+      "eval_bleu": 40.8492,
+      "eval_gen_len": 13.8028,
+      "eval_loss": 0.9435123801231384,
+      "eval_runtime": 10127.6159,
+      "eval_samples_per_second": 14.005,
+      "eval_steps_per_second": 1.751,
+      "step": 70919
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.7135632537528e-05,
+      "loss": 0.9687,
+      "step": 71000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.7111894161024056e-05,
+      "loss": 0.9639,
+      "step": 71500
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.708815578452011e-05,
+      "loss": 0.9669,
+      "step": 72000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7064417408016166e-05,
+      "loss": 0.956,
+      "step": 72500
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 4.704067903151222e-05,
+      "loss": 0.9563,
+      "step": 73000
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 4.701694065500828e-05,
+      "loss": 0.9564,
+      "step": 73500
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 4.699320227850433e-05,
+      "loss": 0.9555,
+      "step": 74000
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 4.696946390200039e-05,
+      "loss": 0.9408,
+      "step": 74500
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.6945725525496444e-05,
+      "loss": 0.9347,
+      "step": 75000
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.69219871489925e-05,
+      "loss": 0.9368,
+      "step": 75500
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 4.6898296249241566e-05,
+      "loss": 0.9401,
+      "step": 76000
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 4.6874557872737615e-05,
+      "loss": 0.9383,
+      "step": 76500
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.685081949623367e-05,
+      "loss": 0.9232,
+      "step": 77000
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.6827081119729726e-05,
+      "loss": 0.9046,
+      "step": 77500
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 4.680334274322578e-05,
+      "loss": 0.9236,
+      "step": 78000
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.6779604366721837e-05,
+      "loss": 0.9091,
+      "step": 78500
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.675586599021789e-05,
+      "loss": 0.9036,
+      "step": 79000
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 4.673212761371395e-05,
+      "loss": 0.909,
+      "step": 79500
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 4.670838923721e-05,
+      "loss": 0.9003,
+      "step": 80000
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.668469833745906e-05,
+      "loss": 0.8957,
+      "step": 80500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.666095996095512e-05,
+      "loss": 0.8928,
+      "step": 81000
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 4.66373640147102e-05,
+      "loss": 0.8842,
+      "step": 81500
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.6613625638206256e-05,
+      "loss": 0.885,
+      "step": 82000
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.658988726170231e-05,
+      "loss": 0.8821,
+      "step": 82500
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 4.656614888519836e-05,
+      "loss": 0.8769,
+      "step": 83000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 4.6542410508694416e-05,
+      "loss": 0.8681,
+      "step": 83500
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 4.651867213219048e-05,
+      "loss": 0.8767,
+      "step": 84000
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 4.6494933755686534e-05,
+      "loss": 0.8702,
+      "step": 84500
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 4.647119537918259e-05,
+      "loss": 0.8753,
+      "step": 85000
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 4.644745700267864e-05,
+      "loss": 0.8587,
+      "step": 85500
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 4.642371862617469e-05,
+      "loss": 0.8459,
+      "step": 86000
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 4.639998024967075e-05,
+      "loss": 0.853,
+      "step": 86500
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 4.6376241873166804e-05,
+      "loss": 0.8811,
+      "step": 87000
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 4.635255097341587e-05,
+      "loss": 0.8618,
+      "step": 87500
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 4.6328812596911926e-05,
+      "loss": 0.8493,
+      "step": 88000
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 4.630507422040798e-05,
+      "loss": 0.8603,
+      "step": 88500
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 4.628133584390403e-05,
+      "loss": 0.8333,
+      "step": 89000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 4.6257597467400086e-05,
+      "loss": 0.8386,
+      "step": 89500
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 4.623385909089614e-05,
+      "loss": 0.8424,
+      "step": 90000
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 4.6210120714392204e-05,
+      "loss": 0.8519,
+      "step": 90500
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 4.618638233788826e-05,
+      "loss": 0.8297,
+      "step": 91000
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 4.616264396138431e-05,
+      "loss": 0.8406,
+      "step": 91500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 4.613890558488036e-05,
+      "loss": 0.8392,
+      "step": 92000
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 4.611516720837642e-05,
+      "loss": 0.8269,
+      "step": 92500
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 4.6091476308625486e-05,
+      "loss": 0.8392,
+      "step": 93000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 4.6067785408874545e-05,
+      "loss": 0.8474,
+      "step": 93500
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 4.60440470323706e-05,
+      "loss": 0.8329,
+      "step": 94000
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 4.6020308655866656e-05,
+      "loss": 0.8327,
+      "step": 94500
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 4.599657027936271e-05,
+      "loss": 0.8435,
+      "step": 95000
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 4.597283190285877e-05,
+      "loss": 0.8343,
+      "step": 95500
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 4.5949141003107834e-05,
+      "loss": 0.8316,
+      "step": 96000
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 4.592540262660389e-05,
+      "loss": 0.8339,
+      "step": 96500
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 4.590166425009994e-05,
+      "loss": 0.8199,
+      "step": 97000
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 4.5877925873595994e-05,
+      "loss": 0.8162,
+      "step": 97500
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 4.585418749709205e-05,
+      "loss": 0.8239,
+      "step": 98000
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 4.5830496597341116e-05,
+      "loss": 0.8197,
+      "step": 98500
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 4.5806805697590176e-05,
+      "loss": 0.8164,
+      "step": 99000
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 4.578306732108623e-05,
+      "loss": 0.818,
+      "step": 99500
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 4.575932894458229e-05,
+      "loss": 0.8189,
+      "step": 100000
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 4.573559056807834e-05,
+      "loss": 0.8303,
+      "step": 100500
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 4.57118521915744e-05,
+      "loss": 0.8174,
+      "step": 101000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 4.568811381507045e-05,
+      "loss": 0.814,
+      "step": 101500
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 4.566437543856651e-05,
+      "loss": 0.8226,
+      "step": 102000
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 4.5640637062062564e-05,
+      "loss": 0.818,
+      "step": 102500
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 4.561689868555862e-05,
+      "loss": 0.8054,
+      "step": 103000
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 4.5593160309054675e-05,
+      "loss": 0.8191,
+      "step": 103500
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 4.556946940930374e-05,
+      "loss": 0.8134,
+      "step": 104000
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 4.554573103279979e-05,
+      "loss": 0.8077,
+      "step": 104500
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 4.5521992656295846e-05,
+      "loss": 0.8081,
+      "step": 105000
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 4.54982542797919e-05,
+      "loss": 0.7985,
+      "step": 105500
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 4.547451590328796e-05,
+      "loss": 0.8064,
+      "step": 106000
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 4.545077752678401e-05,
+      "loss": 0.8008,
+      "step": 106500
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 4.542703915028007e-05,
+      "loss": 0.7934,
+      "step": 107000
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 4.540330077377612e-05,
+      "loss": 0.7915,
+      "step": 107500
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 4.537956239727218e-05,
+      "loss": 0.8087,
+      "step": 108000
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 4.5355824020768234e-05,
+      "loss": 0.7941,
+      "step": 108500
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 4.533208564426429e-05,
+      "loss": 0.8032,
+      "step": 109000
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 4.5308347267760345e-05,
+      "loss": 0.8061,
+      "step": 109500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 4.5284608891256394e-05,
+      "loss": 0.7974,
+      "step": 110000
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 4.526091799150546e-05,
+      "loss": 0.7915,
+      "step": 110500
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 4.523722709175453e-05,
+      "loss": 0.7998,
+      "step": 111000
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 4.5213488715250576e-05,
+      "loss": 0.8014,
+      "step": 111500
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 4.518975033874663e-05,
+      "loss": 0.7848,
+      "step": 112000
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.516601196224269e-05,
+      "loss": 0.7914,
+      "step": 112500
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.5142321062491754e-05,
+      "loss": 0.795,
+      "step": 113000
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.5118630162740814e-05,
+      "loss": 0.7913,
+      "step": 113500
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.509489178623687e-05,
+      "loss": 0.7936,
+      "step": 114000
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.5071153409732924e-05,
+      "loss": 0.7932,
+      "step": 114500
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 4.504741503322898e-05,
+      "loss": 0.7835,
+      "step": 115000
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 4.502367665672504e-05,
+      "loss": 0.7942,
+      "step": 115500
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 4.499993828022109e-05,
+      "loss": 0.7771,
+      "step": 116000
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 4.4976199903717146e-05,
+      "loss": 0.7805,
+      "step": 116500
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 4.4952509003966206e-05,
+      "loss": 0.7952,
+      "step": 117000
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 4.492877062746226e-05,
+      "loss": 0.7799,
+      "step": 117500
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 4.490503225095832e-05,
+      "loss": 0.7932,
+      "step": 118000
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 4.488129387445438e-05,
+      "loss": 0.7919,
+      "step": 118500
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 4.4857555497950435e-05,
+      "loss": 0.7776,
+      "step": 119000
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 4.4833864598199495e-05,
+      "loss": 0.7731,
+      "step": 119500
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 4.481012622169555e-05,
+      "loss": 0.7778,
+      "step": 120000
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 4.47863878451916e-05,
+      "loss": 0.7638,
+      "step": 120500
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 4.476264946868766e-05,
+      "loss": 0.7819,
+      "step": 121000
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 4.473891109218372e-05,
+      "loss": 0.7717,
+      "step": 121500
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 4.471517271567977e-05,
+      "loss": 0.7803,
+      "step": 122000
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 4.469143433917583e-05,
+      "loss": 0.7807,
+      "step": 122500
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 4.4667695962671876e-05,
+      "loss": 0.7651,
+      "step": 123000
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 4.464395758616793e-05,
+      "loss": 0.7811,
+      "step": 123500
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 4.4620266686417e-05,
+      "loss": 0.7794,
+      "step": 124000
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 4.4596528309913054e-05,
+      "loss": 0.7763,
+      "step": 124500
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 4.457278993340911e-05,
+      "loss": 0.7757,
+      "step": 125000
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 4.4549051556905165e-05,
+      "loss": 0.7683,
+      "step": 125500
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 4.4525360657154225e-05,
+      "loss": 0.7764,
+      "step": 126000
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 4.450162228065028e-05,
+      "loss": 0.7661,
+      "step": 126500
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 4.4477883904146336e-05,
+      "loss": 0.7645,
+      "step": 127000
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 4.445414552764239e-05,
+      "loss": 0.7703,
+      "step": 127500
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 4.443040715113845e-05,
+      "loss": 0.7791,
+      "step": 128000
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 4.44066687746345e-05,
+      "loss": 0.7657,
+      "step": 128500
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 4.438293039813056e-05,
+      "loss": 0.7651,
+      "step": 129000
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 4.435919202162661e-05,
+      "loss": 0.7803,
+      "step": 129500
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 4.433545364512267e-05,
+      "loss": 0.7504,
+      "step": 130000
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 4.4311762745371735e-05,
+      "loss": 0.7785,
+      "step": 130500
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 4.4288024368867784e-05,
+      "loss": 0.7592,
+      "step": 131000
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 4.426428599236384e-05,
+      "loss": 0.7714,
+      "step": 131500
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 4.4240547615859895e-05,
+      "loss": 0.7652,
+      "step": 132000
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 4.421680923935595e-05,
+      "loss": 0.7608,
+      "step": 132500
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.4193070862852006e-05,
+      "loss": 0.7688,
+      "step": 133000
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.416933248634806e-05,
+      "loss": 0.79,
+      "step": 133500
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 4.414559410984412e-05,
+      "loss": 0.7526,
+      "step": 134000
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 4.412190321009318e-05,
+      "loss": 0.765,
+      "step": 134500
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 4.409816483358923e-05,
+      "loss": 0.7554,
+      "step": 135000
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 4.407442645708529e-05,
+      "loss": 0.7636,
+      "step": 135500
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 4.4050735557334354e-05,
+      "loss": 0.7582,
+      "step": 136000
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 4.402699718083041e-05,
+      "loss": 0.7504,
+      "step": 136500
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 4.400330628107947e-05,
+      "loss": 0.7591,
+      "step": 137000
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 4.3979567904575525e-05,
+      "loss": 0.7629,
+      "step": 137500
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 4.395582952807158e-05,
+      "loss": 0.7543,
+      "step": 138000
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 4.3932091151567636e-05,
+      "loss": 0.7634,
+      "step": 138500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 4.390835277506369e-05,
+      "loss": 0.7553,
+      "step": 139000
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 4.388461439855975e-05,
+      "loss": 0.7548,
+      "step": 139500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 4.38608760220558e-05,
+      "loss": 0.7554,
+      "step": 140000
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 4.383713764555186e-05,
+      "loss": 0.7539,
+      "step": 140500
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 4.3813399269047914e-05,
+      "loss": 0.7562,
+      "step": 141000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.378966089254396e-05,
+      "loss": 0.7537,
+      "step": 141500
+    },
+    {
+      "epoch": 2.0,
+      "eval_bleu": 42.7907,
+      "eval_gen_len": 13.7941,
+      "eval_loss": 0.8987888693809509,
+      "eval_runtime": 10044.1408,
+      "eval_samples_per_second": 14.121,
+      "eval_steps_per_second": 1.765,
+      "step": 141839
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.376596999279303e-05,
+      "loss": 0.7511,
+      "step": 142000
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.3742231616289084e-05,
+      "loss": 0.75,
+      "step": 142500
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.371849323978514e-05,
+      "loss": 0.7489,
+      "step": 143000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3694754863281195e-05,
+      "loss": 0.7359,
+      "step": 143500
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 4.367101648677725e-05,
+      "loss": 0.7416,
+      "step": 144000
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.364732558702632e-05,
+      "loss": 0.7421,
+      "step": 144500
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.362358721052237e-05,
+      "loss": 0.7448,
+      "step": 145000
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 4.359984883401843e-05,
+      "loss": 0.7314,
+      "step": 145500
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 4.357611045751448e-05,
+      "loss": 0.731,
+      "step": 146000
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 4.3552419557763544e-05,
+      "loss": 0.7307,
+      "step": 146500
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 4.35286811812596e-05,
+      "loss": 0.7411,
+      "step": 147000
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.3504942804755655e-05,
+      "loss": 0.7298,
+      "step": 147500
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 4.348120442825171e-05,
+      "loss": 0.7144,
+      "step": 148000
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 4.3457466051747766e-05,
+      "loss": 0.7146,
+      "step": 148500
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 4.343372767524382e-05,
+      "loss": 0.7157,
+      "step": 149000
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 4.341003677549288e-05,
+      "loss": 0.7116,
+      "step": 149500
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.338629839898894e-05,
+      "loss": 0.7105,
+      "step": 150000
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.336256002248499e-05,
+      "loss": 0.7096,
+      "step": 150500
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 4.3338916599487063e-05,
+      "loss": 0.7026,
+      "step": 151000
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 4.331517822298312e-05,
+      "loss": 0.706,
+      "step": 151500
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 4.3291439846479174e-05,
+      "loss": 0.6996,
+      "step": 152000
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 4.326770146997523e-05,
+      "loss": 0.6965,
+      "step": 152500
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 4.3243963093471285e-05,
+      "loss": 0.699,
+      "step": 153000
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 4.322022471696734e-05,
+      "loss": 0.6926,
+      "step": 153500
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 4.3196486340463396e-05,
+      "loss": 0.6925,
+      "step": 154000
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 4.3172747963959445e-05,
+      "loss": 0.6825,
+      "step": 154500
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 4.31490095874555e-05,
+      "loss": 0.6869,
+      "step": 155000
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 4.312527121095156e-05,
+      "loss": 0.691,
+      "step": 155500
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 4.310153283444762e-05,
+      "loss": 0.682,
+      "step": 156000
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 4.3077794457943673e-05,
+      "loss": 0.6822,
+      "step": 156500
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 4.305405608143972e-05,
+      "loss": 0.6666,
+      "step": 157000
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 4.303031770493578e-05,
+      "loss": 0.6749,
+      "step": 157500
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 4.300657932843183e-05,
+      "loss": 0.6903,
+      "step": 158000
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 4.29828884286809e-05,
+      "loss": 0.6816,
+      "step": 158500
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 4.2959150052176955e-05,
+      "loss": 0.6776,
+      "step": 159000
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 4.2935459152426015e-05,
+      "loss": 0.6787,
+      "step": 159500
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 4.2911768252675075e-05,
+      "loss": 0.6559,
+      "step": 160000
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 4.288807735292414e-05,
+      "loss": 0.668,
+      "step": 160500
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 4.28643389764202e-05,
+      "loss": 0.6755,
+      "step": 161000
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 4.284060059991625e-05,
+      "loss": 0.6751,
+      "step": 161500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 4.281686222341231e-05,
+      "loss": 0.6608,
+      "step": 162000
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 4.2793123846908364e-05,
+      "loss": 0.6562,
+      "step": 162500
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 4.276938547040442e-05,
+      "loss": 0.6672,
+      "step": 163000
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 4.2745647093900475e-05,
+      "loss": 0.6577,
+      "step": 163500
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 4.272190871739653e-05,
+      "loss": 0.6714,
+      "step": 164000
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 4.2698170340892586e-05,
+      "loss": 0.6706,
+      "step": 164500
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 4.2674479441141646e-05,
+      "loss": 0.6593,
+      "step": 165000
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 4.26507410646377e-05,
+      "loss": 0.6633,
+      "step": 165500
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 4.262705016488677e-05,
+      "loss": 0.6672,
+      "step": 166000
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 4.260331178838282e-05,
+      "loss": 0.6584,
+      "step": 166500
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 4.257957341187888e-05,
+      "loss": 0.6593,
+      "step": 167000
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 4.255583503537493e-05,
+      "loss": 0.6651,
+      "step": 167500
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 4.253209665887098e-05,
+      "loss": 0.6519,
+      "step": 168000
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 4.250835828236704e-05,
+      "loss": 0.6518,
+      "step": 168500
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 4.24846199058631e-05,
+      "loss": 0.6598,
+      "step": 169000
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 4.2460881529359156e-05,
+      "loss": 0.6445,
+      "step": 169500
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 4.2437143152855205e-05,
+      "loss": 0.6499,
+      "step": 170000
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 4.241340477635126e-05,
+      "loss": 0.6516,
+      "step": 170500
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 4.2389666399847316e-05,
+      "loss": 0.6553,
+      "step": 171000
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 4.236592802334337e-05,
+      "loss": 0.668,
+      "step": 171500
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 4.2342189646839427e-05,
+      "loss": 0.6447,
+      "step": 172000
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 4.231845127033548e-05,
+      "loss": 0.6474,
+      "step": 172500
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 4.229471289383154e-05,
+      "loss": 0.6594,
+      "step": 173000
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 4.227097451732759e-05,
+      "loss": 0.6516,
+      "step": 173500
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 4.224723614082365e-05,
+      "loss": 0.6404,
+      "step": 174000
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 4.222354524107271e-05,
+      "loss": 0.6526,
+      "step": 174500
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 4.2199806864568764e-05,
+      "loss": 0.6459,
+      "step": 175000
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 4.2176068488064826e-05,
+      "loss": 0.6519,
+      "step": 175500
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 4.2152377588313886e-05,
+      "loss": 0.6391,
+      "step": 176000
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 4.212863921180994e-05,
+      "loss": 0.6346,
+      "step": 176500
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 4.2104900835306e-05,
+      "loss": 0.6395,
+      "step": 177000
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 4.2081162458802046e-05,
+      "loss": 0.6409,
+      "step": 177500
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 4.20574240822981e-05,
+      "loss": 0.6285,
+      "step": 178000
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 4.2033685705794163e-05,
+      "loss": 0.6307,
+      "step": 178500
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 4.200999480604322e-05,
+      "loss": 0.6506,
+      "step": 179000
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 4.198625642953928e-05,
+      "loss": 0.6307,
+      "step": 179500
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 4.1962518053035334e-05,
+      "loss": 0.6446,
+      "step": 180000
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 4.193877967653139e-05,
+      "loss": 0.6422,
+      "step": 180500
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 4.1915088776780457e-05,
+      "loss": 0.6398,
+      "step": 181000
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 4.1891397877029516e-05,
+      "loss": 0.6333,
+      "step": 181500
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 4.186765950052557e-05,
+      "loss": 0.6363,
+      "step": 182000
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 4.184392112402162e-05,
+      "loss": 0.6319,
+      "step": 182500
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 4.1820182747517676e-05,
+      "loss": 0.6294,
+      "step": 183000
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 4.179649184776674e-05,
+      "loss": 0.6366,
+      "step": 183500
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 4.17727534712628e-05,
+      "loss": 0.6384,
+      "step": 184000
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 4.1749015094758854e-05,
+      "loss": 0.6365,
+      "step": 184500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 4.172527671825491e-05,
+      "loss": 0.6331,
+      "step": 185000
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 4.1701538341750965e-05,
+      "loss": 0.6327,
+      "step": 185500
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 4.167779996524701e-05,
+      "loss": 0.628,
+      "step": 186000
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 4.1654061588743076e-05,
+      "loss": 0.6309,
+      "step": 186500
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 4.163032321223913e-05,
+      "loss": 0.6159,
+      "step": 187000
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 4.1606584835735187e-05,
+      "loss": 0.6307,
+      "step": 187500
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 4.158284645923124e-05,
+      "loss": 0.6356,
+      "step": 188000
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 4.155910808272729e-05,
+      "loss": 0.6269,
+      "step": 188500
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 4.1535369706223346e-05,
+      "loss": 0.6368,
+      "step": 189000
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 4.15116313297194e-05,
+      "loss": 0.6282,
+      "step": 189500
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 4.1487892953215464e-05,
+      "loss": 0.6226,
+      "step": 190000
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 4.146415457671152e-05,
+      "loss": 0.6181,
+      "step": 190500
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 4.144041620020757e-05,
+      "loss": 0.6138,
+      "step": 191000
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 4.1416677823703623e-05,
+      "loss": 0.6184,
+      "step": 191500
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 4.139298692395269e-05,
+      "loss": 0.624,
+      "step": 192000
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 4.136924854744874e-05,
+      "loss": 0.6125,
+      "step": 192500
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 4.13455101709448e-05,
+      "loss": 0.6261,
+      "step": 193000
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 4.1321771794440857e-05,
+      "loss": 0.6189,
+      "step": 193500
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 4.129803341793691e-05,
+      "loss": 0.6149,
+      "step": 194000
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 4.127429504143297e-05,
+      "loss": 0.6296,
+      "step": 194500
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 4.1250556664929016e-05,
+      "loss": 0.6263,
+      "step": 195000
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 4.122681828842507e-05,
+      "loss": 0.6298,
+      "step": 195500
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 4.120312738867414e-05,
+      "loss": 0.6102,
+      "step": 196000
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 4.1179389012170194e-05,
+      "loss": 0.6248,
+      "step": 196500
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 4.115565063566625e-05,
+      "loss": 0.6121,
+      "step": 197000
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 4.1131912259162305e-05,
+      "loss": 0.6156,
+      "step": 197500
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 4.1108221359411365e-05,
+      "loss": 0.6076,
+      "step": 198000
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 4.1084577936413436e-05,
+      "loss": 0.6231,
+      "step": 198500
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 4.106083955990949e-05,
+      "loss": 0.6281,
+      "step": 199000
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 4.103710118340555e-05,
+      "loss": 0.6157,
+      "step": 199500
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 4.10133628069016e-05,
+      "loss": 0.6086,
+      "step": 200000
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 4.098962443039766e-05,
+      "loss": 0.6225,
+      "step": 200500
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 4.096588605389371e-05,
+      "loss": 0.6078,
+      "step": 201000
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 4.094214767738977e-05,
+      "loss": 0.6217,
+      "step": 201500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 4.0918409300885824e-05,
+      "loss": 0.6096,
+      "step": 202000
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 4.089467092438188e-05,
+      "loss": 0.6253,
+      "step": 202500
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 4.087102750138395e-05,
+      "loss": 0.6068,
+      "step": 203000
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 4.0847289124880006e-05,
+      "loss": 0.613,
+      "step": 203500
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 4.082355074837606e-05,
+      "loss": 0.6179,
+      "step": 204000
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 4.079981237187212e-05,
+      "loss": 0.6312,
+      "step": 204500
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 4.077612147212118e-05,
+      "loss": 0.601,
+      "step": 205000
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 4.075238309561723e-05,
+      "loss": 0.615,
+      "step": 205500
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 4.072864471911329e-05,
+      "loss": 0.6071,
+      "step": 206000
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 4.0704906342609344e-05,
+      "loss": 0.6135,
+      "step": 206500
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 4.06811679661054e-05,
+      "loss": 0.6073,
+      "step": 207000
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 4.065747706635446e-05,
+      "loss": 0.6073,
+      "step": 207500
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 4.0633738689850515e-05,
+      "loss": 0.6053,
+      "step": 208000
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 4.061000031334658e-05,
+      "loss": 0.6116,
+      "step": 208500
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 4.058626193684263e-05,
+      "loss": 0.606,
+      "step": 209000
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 4.056252356033868e-05,
+      "loss": 0.6183,
+      "step": 209500
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 4.0538785183834736e-05,
+      "loss": 0.601,
+      "step": 210000
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 4.051504680733079e-05,
+      "loss": 0.6041,
+      "step": 210500
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 4.049130843082685e-05,
+      "loss": 0.6073,
+      "step": 211000
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 4.04675700543229e-05,
+      "loss": 0.6106,
+      "step": 211500
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 4.044383167781896e-05,
+      "loss": 0.6105,
+      "step": 212000
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.0420093301315014e-05,
+      "loss": 0.5973,
+      "step": 212500
+    },
+    {
+      "epoch": 3.0,
+      "eval_bleu": 43.4697,
+      "eval_gen_len": 13.6556,
+      "eval_loss": 0.9142627120018005,
+      "eval_runtime": 9819.0968,
+      "eval_samples_per_second": 14.445,
+      "eval_steps_per_second": 1.806,
+      "step": 212758
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.039635492481107e-05,
+      "loss": 0.6014,
+      "step": 213000
+    },
+    {
+      "epoch": 3.01,
+      "learning_rate": 4.037266402506013e-05,
+      "loss": 0.6099,
+      "step": 213500
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 4.0348925648556185e-05,
+      "loss": 0.5951,
+      "step": 214000
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 4.032518727205224e-05,
+      "loss": 0.597,
+      "step": 214500
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 4.03014488955483e-05,
+      "loss": 0.596,
+      "step": 215000
+    },
+    {
+      "epoch": 3.04,
+      "learning_rate": 4.027771051904435e-05,
+      "loss": 0.5914,
+      "step": 215500
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 4.0253972142540406e-05,
+      "loss": 0.5979,
+      "step": 216000
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 4.023023376603646e-05,
+      "loss": 0.5855,
+      "step": 216500
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 4.020649538953252e-05,
+      "loss": 0.5887,
+      "step": 217000
+    },
+    {
+      "epoch": 3.07,
+      "learning_rate": 4.018275701302857e-05,
+      "loss": 0.5853,
+      "step": 217500
+    },
+    {
+      "epoch": 3.07,
+      "learning_rate": 4.015901863652463e-05,
+      "loss": 0.5921,
+      "step": 218000
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 4.013528026002068e-05,
+      "loss": 0.5873,
+      "step": 218500
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 4.0111589360269744e-05,
+      "loss": 0.5708,
+      "step": 219000
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 4.00878509837658e-05,
+      "loss": 0.5777,
+      "step": 219500
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 4.0064112607261855e-05,
+      "loss": 0.5772,
+      "step": 220000
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 4.004037423075791e-05,
+      "loss": 0.573,
+      "step": 220500
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 4.0016635854253966e-05,
+      "loss": 0.5734,
+      "step": 221000
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 3.999289747775002e-05,
+      "loss": 0.5682,
+      "step": 221500
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 3.996920657799909e-05,
+      "loss": 0.5628,
+      "step": 222000
+    },
+    {
+      "epoch": 3.14,
+      "learning_rate": 3.9945468201495136e-05,
+      "loss": 0.5687,
+      "step": 222500
+    },
+    {
+      "epoch": 3.14,
+      "learning_rate": 3.992172982499119e-05,
+      "loss": 0.5569,
+      "step": 223000
+    },
+    {
+      "epoch": 3.15,
+      "learning_rate": 3.989799144848725e-05,
+      "loss": 0.5571,
+      "step": 223500
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 3.98742530719833e-05,
+      "loss": 0.5619,
+      "step": 224000
+    },
+    {
+      "epoch": 3.17,
+      "learning_rate": 3.9850514695479365e-05,
+      "loss": 0.5635,
+      "step": 224500
+    },
+    {
+      "epoch": 3.17,
+      "learning_rate": 3.982677631897542e-05,
+      "loss": 0.5538,
+      "step": 225000
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 3.980303794247147e-05,
+      "loss": 0.5494,
+      "step": 225500
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 3.9779299565967525e-05,
+      "loss": 0.5532,
+      "step": 226000
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 3.975556118946358e-05,
+      "loss": 0.5576,
+      "step": 226500
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 3.973187028971264e-05,
+      "loss": 0.5499,
+      "step": 227000
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 3.970822686671472e-05,
+      "loss": 0.5434,
+      "step": 227500
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 3.968448849021077e-05,
+      "loss": 0.5421,
+      "step": 228000
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 3.966075011370682e-05,
+      "loss": 0.5377,
+      "step": 228500
+    },
+    {
+      "epoch": 3.23,
+      "learning_rate": 3.963701173720288e-05,
+      "loss": 0.5614,
+      "step": 229000
+    },
+    {
+      "epoch": 3.24,
+      "learning_rate": 3.961327336069894e-05,
+      "loss": 0.5458,
+      "step": 229500
+    },
+    {
+      "epoch": 3.24,
+      "learning_rate": 3.9589534984194995e-05,
+      "loss": 0.5483,
+      "step": 230000
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 3.9565796607691044e-05,
+      "loss": 0.5423,
+      "step": 230500
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 3.95420582311871e-05,
+      "loss": 0.5346,
+      "step": 231000
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 3.9518319854683155e-05,
+      "loss": 0.5334,
+      "step": 231500
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 3.9494628954932215e-05,
+      "loss": 0.5451,
+      "step": 232000
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 3.947089057842828e-05,
+      "loss": 0.5378,
+      "step": 232500
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 3.944715220192433e-05,
+      "loss": 0.5323,
+      "step": 233000
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 3.942346130217339e-05,
+      "loss": 0.5346,
+      "step": 233500
+    },
+    {
+      "epoch": 3.3,
+      "learning_rate": 3.939972292566945e-05,
+      "loss": 0.5334,
+      "step": 234000
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 3.9375984549165504e-05,
+      "loss": 0.5292,
+      "step": 234500
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 3.935224617266155e-05,
+      "loss": 0.5442,
+      "step": 235000
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 3.9328507796157615e-05,
+      "loss": 0.5423,
+      "step": 235500
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 3.930476941965367e-05,
+      "loss": 0.53,
+      "step": 236000
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 3.9281031043149725e-05,
+      "loss": 0.5359,
+      "step": 236500
+    },
+    {
+      "epoch": 3.34,
+      "learning_rate": 3.9257340143398785e-05,
+      "loss": 0.5432,
+      "step": 237000
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 3.923360176689484e-05,
+      "loss": 0.5325,
+      "step": 237500
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 3.9209863390390896e-05,
+      "loss": 0.53,
+      "step": 238000
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 3.918612501388695e-05,
+      "loss": 0.5347,
+      "step": 238500
+    },
+    {
+      "epoch": 3.37,
+      "learning_rate": 3.916238663738301e-05,
+      "loss": 0.5268,
+      "step": 239000
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 3.913864826087906e-05,
+      "loss": 0.5233,
+      "step": 239500
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 3.911490988437512e-05,
+      "loss": 0.5281,
+      "step": 240000
+    },
+    {
+      "epoch": 3.39,
+      "learning_rate": 3.9091171507871174e-05,
+      "loss": 0.5191,
+      "step": 240500
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 3.906743313136723e-05,
+      "loss": 0.5226,
+      "step": 241000
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 3.904369475486328e-05,
+      "loss": 0.525,
+      "step": 241500
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 3.901995637835934e-05,
+      "loss": 0.5252,
+      "step": 242000
+    },
+    {
+      "epoch": 3.42,
+      "learning_rate": 3.8996218001855396e-05,
+      "loss": 0.5406,
+      "step": 242500
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 3.897257457885746e-05,
+      "loss": 0.5168,
+      "step": 243000
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 3.8948836202353515e-05,
+      "loss": 0.5218,
+      "step": 243500
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 3.892509782584958e-05,
+      "loss": 0.5304,
+      "step": 244000
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 3.890135944934563e-05,
+      "loss": 0.5217,
+      "step": 244500
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 3.887762107284169e-05,
+      "loss": 0.5143,
+      "step": 245000
+    },
+    {
+      "epoch": 3.46,
+      "learning_rate": 3.885388269633774e-05,
+      "loss": 0.5326,
+      "step": 245500
+    },
+    {
+      "epoch": 3.47,
+      "learning_rate": 3.883014431983379e-05,
+      "loss": 0.5152,
+      "step": 246000
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 3.880640594332985e-05,
+      "loss": 0.5288,
+      "step": 246500
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 3.8782667566825904e-05,
+      "loss": 0.517,
+      "step": 247000
+    },
+    {
+      "epoch": 3.49,
+      "learning_rate": 3.8758929190321966e-05,
+      "loss": 0.5092,
+      "step": 247500
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 3.8735238290571026e-05,
+      "loss": 0.5147,
+      "step": 248000
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 3.871149991406708e-05,
+      "loss": 0.5175,
+      "step": 248500
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 3.868780901431614e-05,
+      "loss": 0.4983,
+      "step": 249000
+    },
+    {
+      "epoch": 3.52,
+      "learning_rate": 3.866411811456521e-05,
+      "loss": 0.5145,
+      "step": 249500
+    },
+    {
+      "epoch": 3.53,
+      "learning_rate": 3.8640379738061264e-05,
+      "loss": 0.5226,
+      "step": 250000
+    },
+    {
+      "epoch": 3.53,
+      "learning_rate": 3.861664136155731e-05,
+      "loss": 0.5148,
+      "step": 250500
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 3.859290298505337e-05,
+      "loss": 0.5172,
+      "step": 251000
+    },
+    {
+      "epoch": 3.55,
+      "learning_rate": 3.856916460854942e-05,
+      "loss": 0.5245,
+      "step": 251500
+    },
+    {
+      "epoch": 3.55,
+      "learning_rate": 3.854547370879849e-05,
+      "loss": 0.5164,
+      "step": 252000
+    },
+    {
+      "epoch": 3.56,
+      "learning_rate": 3.8521735332294545e-05,
+      "loss": 0.5087,
+      "step": 252500
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 3.84979969557906e-05,
+      "loss": 0.517,
+      "step": 253000
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 3.8474258579286656e-05,
+      "loss": 0.5089,
+      "step": 253500
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 3.8450520202782705e-05,
+      "loss": 0.5045,
+      "step": 254000
+    },
+    {
+      "epoch": 3.59,
+      "learning_rate": 3.842682930303177e-05,
+      "loss": 0.5169,
+      "step": 254500
+    },
+    {
+      "epoch": 3.6,
+      "learning_rate": 3.840309092652783e-05,
+      "loss": 0.5204,
+      "step": 255000
+    },
+    {
+      "epoch": 3.6,
+      "learning_rate": 3.837935255002388e-05,
+      "loss": 0.5064,
+      "step": 255500
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 3.835561417351994e-05,
+      "loss": 0.5074,
+      "step": 256000
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 3.8331875797015994e-05,
+      "loss": 0.5147,
+      "step": 256500
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 3.830813742051205e-05,
+      "loss": 0.5031,
+      "step": 257000
+    },
+    {
+      "epoch": 3.63,
+      "learning_rate": 3.8284399044008104e-05,
+      "loss": 0.5046,
+      "step": 257500
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 3.826066066750415e-05,
+      "loss": 0.5023,
+      "step": 258000
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 3.8236922291000215e-05,
+      "loss": 0.5053,
+      "step": 258500
+    },
+    {
+      "epoch": 3.65,
+      "learning_rate": 3.821318391449627e-05,
+      "loss": 0.5145,
+      "step": 259000
+    },
+    {
+      "epoch": 3.66,
+      "learning_rate": 3.8189445537992326e-05,
+      "loss": 0.5037,
+      "step": 259500
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 3.816570716148838e-05,
+      "loss": 0.5164,
+      "step": 260000
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 3.814196878498443e-05,
+      "loss": 0.5089,
+      "step": 260500
+    },
+    {
+      "epoch": 3.68,
+      "learning_rate": 3.8118230408480486e-05,
+      "loss": 0.499,
+      "step": 261000
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 3.809453950872955e-05,
+      "loss": 0.5004,
+      "step": 261500
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 3.807084860897861e-05,
+      "loss": 0.4955,
+      "step": 262000
+    },
+    {
+      "epoch": 3.7,
+      "learning_rate": 3.804715770922768e-05,
+      "loss": 0.5021,
+      "step": 262500
+    },
+    {
+      "epoch": 3.71,
+      "learning_rate": 3.802341933272373e-05,
+      "loss": 0.5064,
+      "step": 263000
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 3.799968095621979e-05,
+      "loss": 0.4947,
+      "step": 263500
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 3.7975942579715846e-05,
+      "loss": 0.5033,
+      "step": 264000
+    },
+    {
+      "epoch": 3.73,
+      "learning_rate": 3.79522042032119e-05,
+      "loss": 0.5022,
+      "step": 264500
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 3.792846582670796e-05,
+      "loss": 0.4956,
+      "step": 265000
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 3.790477492695702e-05,
+      "loss": 0.5102,
+      "step": 265500
+    },
+    {
+      "epoch": 3.75,
+      "learning_rate": 3.788103655045307e-05,
+      "loss": 0.509,
+      "step": 266000
+    },
+    {
+      "epoch": 3.76,
+      "learning_rate": 3.785729817394913e-05,
+      "loss": 0.5048,
+      "step": 266500
+    },
+    {
+      "epoch": 3.76,
+      "learning_rate": 3.783355979744518e-05,
+      "loss": 0.4954,
+      "step": 267000
+    },
+    {
+      "epoch": 3.77,
+      "learning_rate": 3.780982142094124e-05,
+      "loss": 0.5009,
+      "step": 267500
+    },
+    {
+      "epoch": 3.78,
+      "learning_rate": 3.7786083044437294e-05,
+      "loss": 0.4909,
+      "step": 268000
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 3.776234466793335e-05,
+      "loss": 0.5018,
+      "step": 268500
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 3.77386062914294e-05,
+      "loss": 0.4903,
+      "step": 269000
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 3.7714867914925454e-05,
+      "loss": 0.504,
+      "step": 269500
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 3.7691129538421516e-05,
+      "loss": 0.5088,
+      "step": 270000
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 3.7667438638670576e-05,
+      "loss": 0.5009,
+      "step": 270500
+    },
+    {
+      "epoch": 3.82,
+      "learning_rate": 3.7643747738919636e-05,
+      "loss": 0.4924,
+      "step": 271000
+    },
+    {
+      "epoch": 3.83,
+      "learning_rate": 3.76200568391687e-05,
+      "loss": 0.4999,
+      "step": 271500
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 3.759631846266476e-05,
+      "loss": 0.4997,
+      "step": 272000
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 3.7572580086160813e-05,
+      "loss": 0.4968,
+      "step": 272500
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 3.754884170965687e-05,
+      "loss": 0.4905,
+      "step": 273000
+    },
+    {
+      "epoch": 3.86,
+      "learning_rate": 3.7525103333152924e-05,
+      "loss": 0.5055,
+      "step": 273500
+    },
+    {
+      "epoch": 3.86,
+      "learning_rate": 3.750136495664898e-05,
+      "loss": 0.4918,
+      "step": 274000
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 3.747762658014503e-05,
+      "loss": 0.4947,
+      "step": 274500
+    },
+    {
+      "epoch": 3.88,
+      "learning_rate": 3.745388820364109e-05,
+      "loss": 0.5016,
+      "step": 275000
+    },
+    {
+      "epoch": 3.88,
+      "learning_rate": 3.7430149827137146e-05,
+      "loss": 0.5084,
+      "step": 275500
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 3.74064114506332e-05,
+      "loss": 0.4882,
+      "step": 276000
+    },
+    {
+      "epoch": 3.9,
+      "learning_rate": 3.738272055088226e-05,
+      "loss": 0.4948,
+      "step": 276500
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 3.735898217437832e-05,
+      "loss": 0.4971,
+      "step": 277000
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 3.733524379787437e-05,
+      "loss": 0.4867,
+      "step": 277500
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 3.731150542137043e-05,
+      "loss": 0.4917,
+      "step": 278000
+    },
+    {
+      "epoch": 3.93,
+      "learning_rate": 3.7287767044866484e-05,
+      "loss": 0.4853,
+      "step": 278500
+    },
+    {
+      "epoch": 3.93,
+      "learning_rate": 3.726402866836254e-05,
+      "loss": 0.4948,
+      "step": 279000
+    },
+    {
+      "epoch": 3.94,
+      "learning_rate": 3.7240290291858594e-05,
+      "loss": 0.4935,
+      "step": 279500
+    },
+    {
+      "epoch": 3.95,
+      "learning_rate": 3.721655191535465e-05,
+      "loss": 0.4906,
+      "step": 280000
+    },
+    {
+      "epoch": 3.96,
+      "learning_rate": 3.71928135388507e-05,
+      "loss": 0.4992,
+      "step": 280500
+    },
+    {
+      "epoch": 3.96,
+      "learning_rate": 3.716907516234676e-05,
+      "loss": 0.4814,
+      "step": 281000
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 3.7145336785842816e-05,
+      "loss": 0.4903,
+      "step": 281500
+    },
+    {
+      "epoch": 3.98,
+      "learning_rate": 3.712159840933887e-05,
+      "loss": 0.4869,
+      "step": 282000
+    },
+    {
+      "epoch": 3.98,
+      "learning_rate": 3.709786003283493e-05,
+      "loss": 0.493,
+      "step": 282500
+    },
+    {
+      "epoch": 3.99,
+      "learning_rate": 3.7074121656330976e-05,
+      "loss": 0.4909,
+      "step": 283000
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.705038327982703e-05,
+      "loss": 0.4873,
+      "step": 283500
+    },
+    {
+      "epoch": 4.0,
+      "eval_bleu": 43.6153,
+      "eval_gen_len": 13.6544,
+      "eval_loss": 0.9757916927337646,
+      "eval_runtime": 9810.7798,
+      "eval_samples_per_second": 14.457,
+      "eval_steps_per_second": 1.807,
+      "step": 283678
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 1063785,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 9.836272613473649e+18,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-283678/training_args.bin b/checkpoint-283678/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..dbbe22d1f032c0b0fd9eedfe9ae519ce9ccd36a7
--- /dev/null
+++ b/checkpoint-283678/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0adf1a980c6128833811b7e6eb546e117ffd3efb8c21dc7de95b5e76a5b21b8d
+size 4728
diff --git a/checkpoint-354597/config.json b/checkpoint-354597/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..c7d123d026fa1d1679aeb1220d9ae316d7c13a0b
--- /dev/null
+++ b/checkpoint-354597/config.json
@@ -0,0 +1,59 @@
+{
+  "_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "early_stopping": true,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 250014,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_length": 200,
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_beams": 5,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tokenizer_class": "MBart50Tokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.35.2",
+  "use_cache": true,
+  "vocab_size": 250054
+}
diff --git a/checkpoint-354597/generation_config.json b/checkpoint-354597/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..7e47c12e3900189593d4b56d0d776b58a7a55627
--- /dev/null
+++ b/checkpoint-354597/generation_config.json
@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "early_stopping": true,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 250014,
+  "forced_eos_token_id": 2,
+  "max_length": 200,
+  "num_beams": 5,
+  "pad_token_id": 1,
+  "transformers_version": "4.35.2"
+}
diff --git a/checkpoint-354597/model.safetensors b/checkpoint-354597/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..35347377c4cbab4100bb9408dce785cb4961ce19
--- /dev/null
+++ b/checkpoint-354597/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ce9e9eb263fd3cc9b36e2fdc0918ce420101f814a240d91ff77a49f82e41db0
+size 2444578688
diff --git a/checkpoint-354597/optimizer.pt b/checkpoint-354597/optimizer.pt
new file mode 100644
index 0000000000000000000000000000000000000000..4d9f0882c3be9653f2ef5162c8f3f26ac452d6fe
--- /dev/null
+++ b/checkpoint-354597/optimizer.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25bb0d2ae19fe22e534b6fa5026eaf9d323d2eca340fcc7e01dbf5601c39754e
+size 4887473903
diff --git a/checkpoint-354597/rng_state_0.pth b/checkpoint-354597/rng_state_0.pth
new file mode 100644
index 0000000000000000000000000000000000000000..365333630d552b8fa01b89d08037ba79e215ad69
--- /dev/null
+++ b/checkpoint-354597/rng_state_0.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd09c3aa3224ca285750e8bdd73139b56096392004cf3cd2dcac1a92be8986bb
+size 15024
diff --git a/checkpoint-354597/rng_state_1.pth b/checkpoint-354597/rng_state_1.pth
new file mode 100644
index 0000000000000000000000000000000000000000..1e3e94ab14b535ba32c8913d2a953a73487c6618
--- /dev/null
+++ b/checkpoint-354597/rng_state_1.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:476f94429eeae0e99cf9e686bcf4841b726c55b9a4f5b4a19553928f1f8fa1da
+size 15024
diff --git a/checkpoint-354597/rng_state_2.pth b/checkpoint-354597/rng_state_2.pth
new file mode 100644
index 0000000000000000000000000000000000000000..7ab3422246eaf3a1fdf1db3166abc1b060e12352
--- /dev/null
+++ b/checkpoint-354597/rng_state_2.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95d07add4fc4c4215c42f55301a6b8884a9fab1e2fcf651ecb89d629cecfc0c2
+size 15024
diff --git a/checkpoint-354597/rng_state_3.pth b/checkpoint-354597/rng_state_3.pth
new file mode 100644
index 0000000000000000000000000000000000000000..916b2da7489871e476da3117e6c2f0c77da5a5cc
--- /dev/null
+++ b/checkpoint-354597/rng_state_3.pth
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc76d4ca55054826774d54f7bfa258338bf25a99ce8d5b5dd025a77f653f14ae
+size 15024
diff --git a/checkpoint-354597/scheduler.pt b/checkpoint-354597/scheduler.pt
new file mode 100644
index 0000000000000000000000000000000000000000..4af85819449fb72aa0eae638d3828cfef3473b5d
--- /dev/null
+++ b/checkpoint-354597/scheduler.pt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bed9e4256d2ef527f867c56827da062cd2b96d2d096480d53aa75d2f1d2e19e
+size 1064
diff --git a/checkpoint-354597/sentencepiece.bpe.model b/checkpoint-354597/sentencepiece.bpe.model
new file mode 100644
index 0000000000000000000000000000000000000000..7a3f40a75f870bc1f21700cd414dc2acc431583c
--- /dev/null
+++ b/checkpoint-354597/sentencepiece.bpe.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051
diff --git a/checkpoint-354597/special_tokens_map.json b/checkpoint-354597/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92619141640d5fcbb4429807de2248352b0dca79
--- /dev/null
+++ b/checkpoint-354597/special_tokens_map.json
@@ -0,0 +1,69 @@
+{
+  "additional_special_tokens": [
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}
diff --git a/checkpoint-354597/tokenizer.json b/checkpoint-354597/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..d7c3ce71bb70639c3fb46702de9c8356f8e2f956
--- /dev/null
+++ b/checkpoint-354597/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d91c41f87c8dbce15b820b41232d0dcd26ba285c22362400d3dd771a711417d
+size 17110107
diff --git a/checkpoint-354597/tokenizer_config.json b/checkpoint-354597/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..733cf8031772d40c50da15c3fe56fe63f05c2a13
--- /dev/null
+++ b/checkpoint-354597/tokenizer_config.json
@@ -0,0 +1,528 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "ar_AR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250002": {
+      "content": "cs_CZ",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250003": {
+      "content": "de_DE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250004": {
+      "content": "en_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250005": {
+      "content": "es_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250006": {
+      "content": "et_EE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250007": {
+      "content": "fi_FI",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250008": {
+      "content": "fr_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250009": {
+      "content": "gu_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250010": {
+      "content": "hi_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250011": {
+      "content": "it_IT",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250012": {
+      "content": "ja_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250013": {
+      "content": "kk_KZ",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250014": {
+      "content": "ko_KR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250015": {
+      "content": "lt_LT",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250016": {
+      "content": "lv_LV",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250017": {
+      "content": "my_MM",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250018": {
+      "content": "ne_NP",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250019": {
+      "content": "nl_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250020": {
+      "content": "ro_RO",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250021": {
+      "content": "ru_RU",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250022": {
+      "content": "si_LK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250023": {
+      "content": "tr_TR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250024": {
+      "content": "vi_VN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250025": {
+      "content": "zh_CN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250026": {
+      "content": "af_ZA",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250027": {
+      "content": "az_AZ",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250028": {
+      "content": "bn_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250029": {
+      "content": "fa_IR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250030": {
+      "content": "he_IL",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250031": {
+      "content": "hr_HR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250032": {
+      "content": "id_ID",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250033": {
+      "content": "ka_GE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250034": {
+      "content": "km_KH",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250035": {
+      "content": "mk_MK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250036": {
+      "content": "ml_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250037": {
+      "content": "mn_MN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250038": {
+      "content": "mr_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250039": {
+      "content": "pl_PL",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250040": {
+      "content": "ps_AF",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250041": {
+      "content": "pt_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250042": {
+      "content": "sv_SE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250043": {
+      "content": "sw_KE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250044": {
+      "content": "ta_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250045": {
+      "content": "te_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250046": {
+      "content": "th_TH",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250047": {
+      "content": "tl_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250048": {
+      "content": "uk_UA",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250049": {
+      "content": "ur_PK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250050": {
+      "content": "xh_ZA",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250051": {
+      "content": "gl_ES",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250052": {
+      "content": "sl_SI",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250053": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "sp_model_kwargs": {},
+  "src_lang": "zh_CN",
+  "tgt_lang": "ko_KR",
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_token": "<unk>"
+}
diff --git a/checkpoint-354597/trainer_state.json b/checkpoint-354597/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..1c81628f567b95abc746cb473a57edd47d4491bb
--- /dev/null
+++ b/checkpoint-354597/trainer_state.json
@@ -0,0 +1,4323 @@
+{
+  "best_metric": 0.8987888693809509,
+  "best_model_checkpoint": "./zhko_mbartLarge_100p_run1/checkpoint-141839",
+  "epoch": 4.9999929497528885,
+  "eval_steps": 500,
+  "global_step": 354597,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.335965407031397e-06,
+      "loss": 2.879,
+      "step": 500
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6860312088738486e-06,
+      "loss": 2.0241,
+      "step": 1000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.0360970107162996e-06,
+      "loss": 1.8819,
+      "step": 1500
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.386162812558751e-06,
+      "loss": 1.8009,
+      "step": 2000
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.1736228614401204e-05,
+      "loss": 1.7072,
+      "step": 2500
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.4086294416243657e-05,
+      "loss": 1.682,
+      "step": 3000
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.643636021808611e-05,
+      "loss": 1.6188,
+      "step": 3500
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.878642601992856e-05,
+      "loss": 1.5802,
+      "step": 4000
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 2.1136491821771008e-05,
+      "loss": 1.5547,
+      "step": 4500
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 2.348655762361346e-05,
+      "loss": 1.5422,
+      "step": 5000
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2.5831923293852228e-05,
+      "loss": 1.5194,
+      "step": 5500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2.8181989095694684e-05,
+      "loss": 1.4815,
+      "step": 6000
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 3.053205489753713e-05,
+      "loss": 1.4509,
+      "step": 6500
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 3.288212069937958e-05,
+      "loss": 1.458,
+      "step": 7000
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 3.5227486369618354e-05,
+      "loss": 1.4415,
+      "step": 7500
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 3.757285203985712e-05,
+      "loss": 1.4284,
+      "step": 8000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 3.991821771009588e-05,
+      "loss": 1.4297,
+      "step": 8500
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.2268283511938334e-05,
+      "loss": 1.4307,
+      "step": 9000
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.461834931378079e-05,
+      "loss": 1.3971,
+      "step": 9500
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.696371498401955e-05,
+      "loss": 1.4106,
+      "step": 10000
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.9313780785862004e-05,
+      "loss": 1.3999,
+      "step": 10500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.998319322943521e-05,
+      "loss": 1.3937,
+      "step": 11000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.995945485293127e-05,
+      "loss": 1.3773,
+      "step": 11500
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.9935716476427316e-05,
+      "loss": 1.3621,
+      "step": 12000
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.991197809992337e-05,
+      "loss": 1.3552,
+      "step": 12500
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.988823972341943e-05,
+      "loss": 1.3312,
+      "step": 13000
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.986450134691549e-05,
+      "loss": 1.3331,
+      "step": 13500
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.984081044716455e-05,
+      "loss": 1.3259,
+      "step": 14000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.9817072070660604e-05,
+      "loss": 1.2983,
+      "step": 14500
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.979333369415666e-05,
+      "loss": 1.2828,
+      "step": 15000
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.976959531765271e-05,
+      "loss": 1.2793,
+      "step": 15500
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.9745856941148764e-05,
+      "loss": 1.2968,
+      "step": 16000
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.9722118564644826e-05,
+      "loss": 1.2774,
+      "step": 16500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.969838018814088e-05,
+      "loss": 1.2507,
+      "step": 17000
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.967464181163694e-05,
+      "loss": 1.2521,
+      "step": 17500
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.965090343513299e-05,
+      "loss": 1.2337,
+      "step": 18000
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.962716505862904e-05,
+      "loss": 1.2238,
+      "step": 18500
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.96034266821251e-05,
+      "loss": 1.2111,
+      "step": 19000
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.957968830562115e-05,
+      "loss": 1.2336,
+      "step": 19500
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.955599740587022e-05,
+      "loss": 1.1975,
+      "step": 20000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.9532259029366274e-05,
+      "loss": 1.2092,
+      "step": 20500
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.950852065286233e-05,
+      "loss": 1.2079,
+      "step": 21000
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.948482975311139e-05,
+      "loss": 1.1831,
+      "step": 21500
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.946109137660745e-05,
+      "loss": 1.1965,
+      "step": 22000
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.94373530001035e-05,
+      "loss": 1.1928,
+      "step": 22500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.9413614623599556e-05,
+      "loss": 1.1779,
+      "step": 23000
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.938987624709561e-05,
+      "loss": 1.1759,
+      "step": 23500
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.936613787059167e-05,
+      "loss": 1.1871,
+      "step": 24000
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.934239949408772e-05,
+      "loss": 1.1782,
+      "step": 24500
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.931866111758378e-05,
+      "loss": 1.1619,
+      "step": 25000
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.9294922741079834e-05,
+      "loss": 1.1615,
+      "step": 25500
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.927118436457589e-05,
+      "loss": 1.1476,
+      "step": 26000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.924749346482495e-05,
+      "loss": 1.1507,
+      "step": 26500
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.922380256507401e-05,
+      "loss": 1.1489,
+      "step": 27000
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.9200064188570064e-05,
+      "loss": 1.144,
+      "step": 27500
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.917632581206613e-05,
+      "loss": 1.1223,
+      "step": 28000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.915258743556218e-05,
+      "loss": 1.1319,
+      "step": 28500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.912884905905824e-05,
+      "loss": 1.1235,
+      "step": 29000
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.9105110682554286e-05,
+      "loss": 1.1408,
+      "step": 29500
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.908141978280335e-05,
+      "loss": 1.1299,
+      "step": 30000
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.905772888305242e-05,
+      "loss": 1.1172,
+      "step": 30500
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.903399050654847e-05,
+      "loss": 1.1316,
+      "step": 31000
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.9010299606797535e-05,
+      "loss": 1.1154,
+      "step": 31500
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.8986561230293584e-05,
+      "loss": 1.1162,
+      "step": 32000
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.896282285378964e-05,
+      "loss": 1.101,
+      "step": 32500
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.89390844772857e-05,
+      "loss": 1.1153,
+      "step": 33000
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.891534610078176e-05,
+      "loss": 1.1023,
+      "step": 33500
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.889160772427781e-05,
+      "loss": 1.0935,
+      "step": 34000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.886786934777387e-05,
+      "loss": 1.0898,
+      "step": 34500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.884413097126992e-05,
+      "loss": 1.0846,
+      "step": 35000
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.882039259476597e-05,
+      "loss": 1.0963,
+      "step": 35500
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.879665421826203e-05,
+      "loss": 1.0764,
+      "step": 36000
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.877291584175809e-05,
+      "loss": 1.0649,
+      "step": 36500
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.8749177465254145e-05,
+      "loss": 1.0935,
+      "step": 37000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.8725439088750194e-05,
+      "loss": 1.0739,
+      "step": 37500
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.870170071224625e-05,
+      "loss": 1.0779,
+      "step": 38000
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.8677962335742305e-05,
+      "loss": 1.083,
+      "step": 38500
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.865422395923836e-05,
+      "loss": 1.0729,
+      "step": 39000
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.8630485582734416e-05,
+      "loss": 1.0587,
+      "step": 39500
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.860679468298348e-05,
+      "loss": 1.078,
+      "step": 40000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.858305630647954e-05,
+      "loss": 1.0802,
+      "step": 40500
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.855931792997559e-05,
+      "loss": 1.0538,
+      "step": 41000
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.853557955347164e-05,
+      "loss": 1.0562,
+      "step": 41500
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.85118411769677e-05,
+      "loss": 1.0516,
+      "step": 42000
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.848810280046375e-05,
+      "loss": 1.0624,
+      "step": 42500
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.8464364423959815e-05,
+      "loss": 1.0587,
+      "step": 43000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.8440626047455864e-05,
+      "loss": 1.0544,
+      "step": 43500
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.841688767095192e-05,
+      "loss": 1.0467,
+      "step": 44000
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.8393291724707e-05,
+      "loss": 1.0613,
+      "step": 44500
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.836955334820306e-05,
+      "loss": 1.0365,
+      "step": 45000
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.834581497169911e-05,
+      "loss": 1.0424,
+      "step": 45500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.832207659519516e-05,
+      "loss": 1.0442,
+      "step": 46000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.829833821869122e-05,
+      "loss": 1.0354,
+      "step": 46500
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.8274789749199304e-05,
+      "loss": 1.0543,
+      "step": 47000
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.825105137269536e-05,
+      "loss": 1.0476,
+      "step": 47500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.8227312996191415e-05,
+      "loss": 1.0427,
+      "step": 48000
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.820357461968748e-05,
+      "loss": 1.0174,
+      "step": 48500
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.8179836243183526e-05,
+      "loss": 1.0325,
+      "step": 49000
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.815609786667958e-05,
+      "loss": 1.0126,
+      "step": 49500
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.813240696692865e-05,
+      "loss": 1.0335,
+      "step": 50000
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.81086685904247e-05,
+      "loss": 1.0219,
+      "step": 50500
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.808493021392075e-05,
+      "loss": 1.0291,
+      "step": 51000
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.8061191837416815e-05,
+      "loss": 1.0251,
+      "step": 51500
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.803745346091287e-05,
+      "loss": 1.01,
+      "step": 52000
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.8013715084408925e-05,
+      "loss": 1.0224,
+      "step": 52500
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.7989976707904974e-05,
+      "loss": 1.0163,
+      "step": 53000
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.796623833140103e-05,
+      "loss": 1.0214,
+      "step": 53500
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.7942499954897085e-05,
+      "loss": 1.0187,
+      "step": 54000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.791876157839314e-05,
+      "loss": 1.0091,
+      "step": 54500
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.78950232018892e-05,
+      "loss": 1.0256,
+      "step": 55000
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.787128482538525e-05,
+      "loss": 1.0025,
+      "step": 55500
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.784754644888131e-05,
+      "loss": 1.0013,
+      "step": 56000
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.782380807237736e-05,
+      "loss": 1.0098,
+      "step": 56500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.780006969587342e-05,
+      "loss": 1.017,
+      "step": 57000
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.777637879612248e-05,
+      "loss": 1.0114,
+      "step": 57500
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.775264041961854e-05,
+      "loss": 0.9977,
+      "step": 58000
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.7728902043114596e-05,
+      "loss": 1.0107,
+      "step": 58500
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.7705163666610644e-05,
+      "loss": 0.9869,
+      "step": 59000
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.76814252901067e-05,
+      "loss": 1.008,
+      "step": 59500
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.7657686913602755e-05,
+      "loss": 1.0025,
+      "step": 60000
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.763394853709881e-05,
+      "loss": 0.9964,
+      "step": 60500
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.7610210160594866e-05,
+      "loss": 0.9921,
+      "step": 61000
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.758647178409092e-05,
+      "loss": 0.9938,
+      "step": 61500
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.756273340758698e-05,
+      "loss": 0.9934,
+      "step": 62000
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.753899503108303e-05,
+      "loss": 1.0249,
+      "step": 62500
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.751530413133209e-05,
+      "loss": 0.9832,
+      "step": 63000
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.749156575482815e-05,
+      "loss": 1.001,
+      "step": 63500
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.74678273783242e-05,
+      "loss": 0.9755,
+      "step": 64000
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4.7444089001820266e-05,
+      "loss": 0.995,
+      "step": 64500
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.742035062531632e-05,
+      "loss": 0.9824,
+      "step": 65000
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.739661224881237e-05,
+      "loss": 0.9759,
+      "step": 65500
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.7372873872308425e-05,
+      "loss": 0.98,
+      "step": 66000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.734913549580448e-05,
+      "loss": 0.9905,
+      "step": 66500
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.7325397119300536e-05,
+      "loss": 0.9811,
+      "step": 67000
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.730165874279659e-05,
+      "loss": 0.9873,
+      "step": 67500
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.727792036629265e-05,
+      "loss": 0.9804,
+      "step": 68000
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.7254229466541714e-05,
+      "loss": 0.9695,
+      "step": 68500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.723049109003776e-05,
+      "loss": 0.9751,
+      "step": 69000
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.720675271353382e-05,
+      "loss": 0.9735,
+      "step": 69500
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.7183061813782885e-05,
+      "loss": 0.9832,
+      "step": 70000
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.7159370914031945e-05,
+      "loss": 0.9719,
+      "step": 70500
+    },
+    {
+      "epoch": 1.0,
+      "eval_bleu": 40.8492,
+      "eval_gen_len": 13.8028,
+      "eval_loss": 0.9435123801231384,
+      "eval_runtime": 10127.6159,
+      "eval_samples_per_second": 14.005,
+      "eval_steps_per_second": 1.751,
+      "step": 70919
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.7135632537528e-05,
+      "loss": 0.9687,
+      "step": 71000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.7111894161024056e-05,
+      "loss": 0.9639,
+      "step": 71500
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.708815578452011e-05,
+      "loss": 0.9669,
+      "step": 72000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7064417408016166e-05,
+      "loss": 0.956,
+      "step": 72500
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 4.704067903151222e-05,
+      "loss": 0.9563,
+      "step": 73000
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 4.701694065500828e-05,
+      "loss": 0.9564,
+      "step": 73500
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 4.699320227850433e-05,
+      "loss": 0.9555,
+      "step": 74000
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 4.696946390200039e-05,
+      "loss": 0.9408,
+      "step": 74500
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.6945725525496444e-05,
+      "loss": 0.9347,
+      "step": 75000
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.69219871489925e-05,
+      "loss": 0.9368,
+      "step": 75500
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 4.6898296249241566e-05,
+      "loss": 0.9401,
+      "step": 76000
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 4.6874557872737615e-05,
+      "loss": 0.9383,
+      "step": 76500
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.685081949623367e-05,
+      "loss": 0.9232,
+      "step": 77000
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.6827081119729726e-05,
+      "loss": 0.9046,
+      "step": 77500
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 4.680334274322578e-05,
+      "loss": 0.9236,
+      "step": 78000
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.6779604366721837e-05,
+      "loss": 0.9091,
+      "step": 78500
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.675586599021789e-05,
+      "loss": 0.9036,
+      "step": 79000
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 4.673212761371395e-05,
+      "loss": 0.909,
+      "step": 79500
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 4.670838923721e-05,
+      "loss": 0.9003,
+      "step": 80000
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.668469833745906e-05,
+      "loss": 0.8957,
+      "step": 80500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.666095996095512e-05,
+      "loss": 0.8928,
+      "step": 81000
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 4.66373640147102e-05,
+      "loss": 0.8842,
+      "step": 81500
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.6613625638206256e-05,
+      "loss": 0.885,
+      "step": 82000
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.658988726170231e-05,
+      "loss": 0.8821,
+      "step": 82500
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 4.656614888519836e-05,
+      "loss": 0.8769,
+      "step": 83000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 4.6542410508694416e-05,
+      "loss": 0.8681,
+      "step": 83500
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 4.651867213219048e-05,
+      "loss": 0.8767,
+      "step": 84000
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 4.6494933755686534e-05,
+      "loss": 0.8702,
+      "step": 84500
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 4.647119537918259e-05,
+      "loss": 0.8753,
+      "step": 85000
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 4.644745700267864e-05,
+      "loss": 0.8587,
+      "step": 85500
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 4.642371862617469e-05,
+      "loss": 0.8459,
+      "step": 86000
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 4.639998024967075e-05,
+      "loss": 0.853,
+      "step": 86500
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 4.6376241873166804e-05,
+      "loss": 0.8811,
+      "step": 87000
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 4.635255097341587e-05,
+      "loss": 0.8618,
+      "step": 87500
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 4.6328812596911926e-05,
+      "loss": 0.8493,
+      "step": 88000
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 4.630507422040798e-05,
+      "loss": 0.8603,
+      "step": 88500
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 4.628133584390403e-05,
+      "loss": 0.8333,
+      "step": 89000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 4.6257597467400086e-05,
+      "loss": 0.8386,
+      "step": 89500
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 4.623385909089614e-05,
+      "loss": 0.8424,
+      "step": 90000
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 4.6210120714392204e-05,
+      "loss": 0.8519,
+      "step": 90500
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 4.618638233788826e-05,
+      "loss": 0.8297,
+      "step": 91000
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 4.616264396138431e-05,
+      "loss": 0.8406,
+      "step": 91500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 4.613890558488036e-05,
+      "loss": 0.8392,
+      "step": 92000
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 4.611516720837642e-05,
+      "loss": 0.8269,
+      "step": 92500
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 4.6091476308625486e-05,
+      "loss": 0.8392,
+      "step": 93000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 4.6067785408874545e-05,
+      "loss": 0.8474,
+      "step": 93500
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 4.60440470323706e-05,
+      "loss": 0.8329,
+      "step": 94000
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 4.6020308655866656e-05,
+      "loss": 0.8327,
+      "step": 94500
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 4.599657027936271e-05,
+      "loss": 0.8435,
+      "step": 95000
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 4.597283190285877e-05,
+      "loss": 0.8343,
+      "step": 95500
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 4.5949141003107834e-05,
+      "loss": 0.8316,
+      "step": 96000
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 4.592540262660389e-05,
+      "loss": 0.8339,
+      "step": 96500
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 4.590166425009994e-05,
+      "loss": 0.8199,
+      "step": 97000
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 4.5877925873595994e-05,
+      "loss": 0.8162,
+      "step": 97500
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 4.585418749709205e-05,
+      "loss": 0.8239,
+      "step": 98000
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 4.5830496597341116e-05,
+      "loss": 0.8197,
+      "step": 98500
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 4.5806805697590176e-05,
+      "loss": 0.8164,
+      "step": 99000
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 4.578306732108623e-05,
+      "loss": 0.818,
+      "step": 99500
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 4.575932894458229e-05,
+      "loss": 0.8189,
+      "step": 100000
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 4.573559056807834e-05,
+      "loss": 0.8303,
+      "step": 100500
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 4.57118521915744e-05,
+      "loss": 0.8174,
+      "step": 101000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 4.568811381507045e-05,
+      "loss": 0.814,
+      "step": 101500
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 4.566437543856651e-05,
+      "loss": 0.8226,
+      "step": 102000
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 4.5640637062062564e-05,
+      "loss": 0.818,
+      "step": 102500
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 4.561689868555862e-05,
+      "loss": 0.8054,
+      "step": 103000
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 4.5593160309054675e-05,
+      "loss": 0.8191,
+      "step": 103500
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 4.556946940930374e-05,
+      "loss": 0.8134,
+      "step": 104000
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 4.554573103279979e-05,
+      "loss": 0.8077,
+      "step": 104500
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 4.5521992656295846e-05,
+      "loss": 0.8081,
+      "step": 105000
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 4.54982542797919e-05,
+      "loss": 0.7985,
+      "step": 105500
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 4.547451590328796e-05,
+      "loss": 0.8064,
+      "step": 106000
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 4.545077752678401e-05,
+      "loss": 0.8008,
+      "step": 106500
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 4.542703915028007e-05,
+      "loss": 0.7934,
+      "step": 107000
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 4.540330077377612e-05,
+      "loss": 0.7915,
+      "step": 107500
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 4.537956239727218e-05,
+      "loss": 0.8087,
+      "step": 108000
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 4.5355824020768234e-05,
+      "loss": 0.7941,
+      "step": 108500
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 4.533208564426429e-05,
+      "loss": 0.8032,
+      "step": 109000
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 4.5308347267760345e-05,
+      "loss": 0.8061,
+      "step": 109500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 4.5284608891256394e-05,
+      "loss": 0.7974,
+      "step": 110000
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 4.526091799150546e-05,
+      "loss": 0.7915,
+      "step": 110500
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 4.523722709175453e-05,
+      "loss": 0.7998,
+      "step": 111000
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 4.5213488715250576e-05,
+      "loss": 0.8014,
+      "step": 111500
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 4.518975033874663e-05,
+      "loss": 0.7848,
+      "step": 112000
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.516601196224269e-05,
+      "loss": 0.7914,
+      "step": 112500
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.5142321062491754e-05,
+      "loss": 0.795,
+      "step": 113000
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.5118630162740814e-05,
+      "loss": 0.7913,
+      "step": 113500
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.509489178623687e-05,
+      "loss": 0.7936,
+      "step": 114000
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.5071153409732924e-05,
+      "loss": 0.7932,
+      "step": 114500
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 4.504741503322898e-05,
+      "loss": 0.7835,
+      "step": 115000
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 4.502367665672504e-05,
+      "loss": 0.7942,
+      "step": 115500
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 4.499993828022109e-05,
+      "loss": 0.7771,
+      "step": 116000
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 4.4976199903717146e-05,
+      "loss": 0.7805,
+      "step": 116500
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 4.4952509003966206e-05,
+      "loss": 0.7952,
+      "step": 117000
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 4.492877062746226e-05,
+      "loss": 0.7799,
+      "step": 117500
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 4.490503225095832e-05,
+      "loss": 0.7932,
+      "step": 118000
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 4.488129387445438e-05,
+      "loss": 0.7919,
+      "step": 118500
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 4.4857555497950435e-05,
+      "loss": 0.7776,
+      "step": 119000
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 4.4833864598199495e-05,
+      "loss": 0.7731,
+      "step": 119500
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 4.481012622169555e-05,
+      "loss": 0.7778,
+      "step": 120000
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 4.47863878451916e-05,
+      "loss": 0.7638,
+      "step": 120500
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 4.476264946868766e-05,
+      "loss": 0.7819,
+      "step": 121000
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 4.473891109218372e-05,
+      "loss": 0.7717,
+      "step": 121500
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 4.471517271567977e-05,
+      "loss": 0.7803,
+      "step": 122000
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 4.469143433917583e-05,
+      "loss": 0.7807,
+      "step": 122500
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 4.4667695962671876e-05,
+      "loss": 0.7651,
+      "step": 123000
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 4.464395758616793e-05,
+      "loss": 0.7811,
+      "step": 123500
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 4.4620266686417e-05,
+      "loss": 0.7794,
+      "step": 124000
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 4.4596528309913054e-05,
+      "loss": 0.7763,
+      "step": 124500
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 4.457278993340911e-05,
+      "loss": 0.7757,
+      "step": 125000
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 4.4549051556905165e-05,
+      "loss": 0.7683,
+      "step": 125500
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 4.4525360657154225e-05,
+      "loss": 0.7764,
+      "step": 126000
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 4.450162228065028e-05,
+      "loss": 0.7661,
+      "step": 126500
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 4.4477883904146336e-05,
+      "loss": 0.7645,
+      "step": 127000
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 4.445414552764239e-05,
+      "loss": 0.7703,
+      "step": 127500
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 4.443040715113845e-05,
+      "loss": 0.7791,
+      "step": 128000
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 4.44066687746345e-05,
+      "loss": 0.7657,
+      "step": 128500
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 4.438293039813056e-05,
+      "loss": 0.7651,
+      "step": 129000
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 4.435919202162661e-05,
+      "loss": 0.7803,
+      "step": 129500
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 4.433545364512267e-05,
+      "loss": 0.7504,
+      "step": 130000
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 4.4311762745371735e-05,
+      "loss": 0.7785,
+      "step": 130500
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 4.4288024368867784e-05,
+      "loss": 0.7592,
+      "step": 131000
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 4.426428599236384e-05,
+      "loss": 0.7714,
+      "step": 131500
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 4.4240547615859895e-05,
+      "loss": 0.7652,
+      "step": 132000
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 4.421680923935595e-05,
+      "loss": 0.7608,
+      "step": 132500
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.4193070862852006e-05,
+      "loss": 0.7688,
+      "step": 133000
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.416933248634806e-05,
+      "loss": 0.79,
+      "step": 133500
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 4.414559410984412e-05,
+      "loss": 0.7526,
+      "step": 134000
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 4.412190321009318e-05,
+      "loss": 0.765,
+      "step": 134500
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 4.409816483358923e-05,
+      "loss": 0.7554,
+      "step": 135000
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 4.407442645708529e-05,
+      "loss": 0.7636,
+      "step": 135500
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 4.4050735557334354e-05,
+      "loss": 0.7582,
+      "step": 136000
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 4.402699718083041e-05,
+      "loss": 0.7504,
+      "step": 136500
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 4.400330628107947e-05,
+      "loss": 0.7591,
+      "step": 137000
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 4.3979567904575525e-05,
+      "loss": 0.7629,
+      "step": 137500
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 4.395582952807158e-05,
+      "loss": 0.7543,
+      "step": 138000
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 4.3932091151567636e-05,
+      "loss": 0.7634,
+      "step": 138500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 4.390835277506369e-05,
+      "loss": 0.7553,
+      "step": 139000
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 4.388461439855975e-05,
+      "loss": 0.7548,
+      "step": 139500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 4.38608760220558e-05,
+      "loss": 0.7554,
+      "step": 140000
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 4.383713764555186e-05,
+      "loss": 0.7539,
+      "step": 140500
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 4.3813399269047914e-05,
+      "loss": 0.7562,
+      "step": 141000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.378966089254396e-05,
+      "loss": 0.7537,
+      "step": 141500
+    },
+    {
+      "epoch": 2.0,
+      "eval_bleu": 42.7907,
+      "eval_gen_len": 13.7941,
+      "eval_loss": 0.8987888693809509,
+      "eval_runtime": 10044.1408,
+      "eval_samples_per_second": 14.121,
+      "eval_steps_per_second": 1.765,
+      "step": 141839
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.376596999279303e-05,
+      "loss": 0.7511,
+      "step": 142000
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.3742231616289084e-05,
+      "loss": 0.75,
+      "step": 142500
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.371849323978514e-05,
+      "loss": 0.7489,
+      "step": 143000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3694754863281195e-05,
+      "loss": 0.7359,
+      "step": 143500
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 4.367101648677725e-05,
+      "loss": 0.7416,
+      "step": 144000
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.364732558702632e-05,
+      "loss": 0.7421,
+      "step": 144500
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.362358721052237e-05,
+      "loss": 0.7448,
+      "step": 145000
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 4.359984883401843e-05,
+      "loss": 0.7314,
+      "step": 145500
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 4.357611045751448e-05,
+      "loss": 0.731,
+      "step": 146000
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 4.3552419557763544e-05,
+      "loss": 0.7307,
+      "step": 146500
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 4.35286811812596e-05,
+      "loss": 0.7411,
+      "step": 147000
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.3504942804755655e-05,
+      "loss": 0.7298,
+      "step": 147500
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 4.348120442825171e-05,
+      "loss": 0.7144,
+      "step": 148000
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 4.3457466051747766e-05,
+      "loss": 0.7146,
+      "step": 148500
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 4.343372767524382e-05,
+      "loss": 0.7157,
+      "step": 149000
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 4.341003677549288e-05,
+      "loss": 0.7116,
+      "step": 149500
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.338629839898894e-05,
+      "loss": 0.7105,
+      "step": 150000
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.336256002248499e-05,
+      "loss": 0.7096,
+      "step": 150500
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 4.3338916599487063e-05,
+      "loss": 0.7026,
+      "step": 151000
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 4.331517822298312e-05,
+      "loss": 0.706,
+      "step": 151500
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 4.3291439846479174e-05,
+      "loss": 0.6996,
+      "step": 152000
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 4.326770146997523e-05,
+      "loss": 0.6965,
+      "step": 152500
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 4.3243963093471285e-05,
+      "loss": 0.699,
+      "step": 153000
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 4.322022471696734e-05,
+      "loss": 0.6926,
+      "step": 153500
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 4.3196486340463396e-05,
+      "loss": 0.6925,
+      "step": 154000
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 4.3172747963959445e-05,
+      "loss": 0.6825,
+      "step": 154500
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 4.31490095874555e-05,
+      "loss": 0.6869,
+      "step": 155000
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 4.312527121095156e-05,
+      "loss": 0.691,
+      "step": 155500
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 4.310153283444762e-05,
+      "loss": 0.682,
+      "step": 156000
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 4.3077794457943673e-05,
+      "loss": 0.6822,
+      "step": 156500
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 4.305405608143972e-05,
+      "loss": 0.6666,
+      "step": 157000
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 4.303031770493578e-05,
+      "loss": 0.6749,
+      "step": 157500
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 4.300657932843183e-05,
+      "loss": 0.6903,
+      "step": 158000
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 4.29828884286809e-05,
+      "loss": 0.6816,
+      "step": 158500
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 4.2959150052176955e-05,
+      "loss": 0.6776,
+      "step": 159000
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 4.2935459152426015e-05,
+      "loss": 0.6787,
+      "step": 159500
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 4.2911768252675075e-05,
+      "loss": 0.6559,
+      "step": 160000
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 4.288807735292414e-05,
+      "loss": 0.668,
+      "step": 160500
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 4.28643389764202e-05,
+      "loss": 0.6755,
+      "step": 161000
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 4.284060059991625e-05,
+      "loss": 0.6751,
+      "step": 161500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 4.281686222341231e-05,
+      "loss": 0.6608,
+      "step": 162000
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 4.2793123846908364e-05,
+      "loss": 0.6562,
+      "step": 162500
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 4.276938547040442e-05,
+      "loss": 0.6672,
+      "step": 163000
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 4.2745647093900475e-05,
+      "loss": 0.6577,
+      "step": 163500
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 4.272190871739653e-05,
+      "loss": 0.6714,
+      "step": 164000
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 4.2698170340892586e-05,
+      "loss": 0.6706,
+      "step": 164500
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 4.2674479441141646e-05,
+      "loss": 0.6593,
+      "step": 165000
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 4.26507410646377e-05,
+      "loss": 0.6633,
+      "step": 165500
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 4.262705016488677e-05,
+      "loss": 0.6672,
+      "step": 166000
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 4.260331178838282e-05,
+      "loss": 0.6584,
+      "step": 166500
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 4.257957341187888e-05,
+      "loss": 0.6593,
+      "step": 167000
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 4.255583503537493e-05,
+      "loss": 0.6651,
+      "step": 167500
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 4.253209665887098e-05,
+      "loss": 0.6519,
+      "step": 168000
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 4.250835828236704e-05,
+      "loss": 0.6518,
+      "step": 168500
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 4.24846199058631e-05,
+      "loss": 0.6598,
+      "step": 169000
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 4.2460881529359156e-05,
+      "loss": 0.6445,
+      "step": 169500
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 4.2437143152855205e-05,
+      "loss": 0.6499,
+      "step": 170000
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 4.241340477635126e-05,
+      "loss": 0.6516,
+      "step": 170500
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 4.2389666399847316e-05,
+      "loss": 0.6553,
+      "step": 171000
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 4.236592802334337e-05,
+      "loss": 0.668,
+      "step": 171500
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 4.2342189646839427e-05,
+      "loss": 0.6447,
+      "step": 172000
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 4.231845127033548e-05,
+      "loss": 0.6474,
+      "step": 172500
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 4.229471289383154e-05,
+      "loss": 0.6594,
+      "step": 173000
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 4.227097451732759e-05,
+      "loss": 0.6516,
+      "step": 173500
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 4.224723614082365e-05,
+      "loss": 0.6404,
+      "step": 174000
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 4.222354524107271e-05,
+      "loss": 0.6526,
+      "step": 174500
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 4.2199806864568764e-05,
+      "loss": 0.6459,
+      "step": 175000
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 4.2176068488064826e-05,
+      "loss": 0.6519,
+      "step": 175500
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 4.2152377588313886e-05,
+      "loss": 0.6391,
+      "step": 176000
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 4.212863921180994e-05,
+      "loss": 0.6346,
+      "step": 176500
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 4.2104900835306e-05,
+      "loss": 0.6395,
+      "step": 177000
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 4.2081162458802046e-05,
+      "loss": 0.6409,
+      "step": 177500
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 4.20574240822981e-05,
+      "loss": 0.6285,
+      "step": 178000
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 4.2033685705794163e-05,
+      "loss": 0.6307,
+      "step": 178500
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 4.200999480604322e-05,
+      "loss": 0.6506,
+      "step": 179000
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 4.198625642953928e-05,
+      "loss": 0.6307,
+      "step": 179500
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 4.1962518053035334e-05,
+      "loss": 0.6446,
+      "step": 180000
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 4.193877967653139e-05,
+      "loss": 0.6422,
+      "step": 180500
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 4.1915088776780457e-05,
+      "loss": 0.6398,
+      "step": 181000
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 4.1891397877029516e-05,
+      "loss": 0.6333,
+      "step": 181500
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 4.186765950052557e-05,
+      "loss": 0.6363,
+      "step": 182000
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 4.184392112402162e-05,
+      "loss": 0.6319,
+      "step": 182500
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 4.1820182747517676e-05,
+      "loss": 0.6294,
+      "step": 183000
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 4.179649184776674e-05,
+      "loss": 0.6366,
+      "step": 183500
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 4.17727534712628e-05,
+      "loss": 0.6384,
+      "step": 184000
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 4.1749015094758854e-05,
+      "loss": 0.6365,
+      "step": 184500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 4.172527671825491e-05,
+      "loss": 0.6331,
+      "step": 185000
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 4.1701538341750965e-05,
+      "loss": 0.6327,
+      "step": 185500
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 4.167779996524701e-05,
+      "loss": 0.628,
+      "step": 186000
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 4.1654061588743076e-05,
+      "loss": 0.6309,
+      "step": 186500
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 4.163032321223913e-05,
+      "loss": 0.6159,
+      "step": 187000
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 4.1606584835735187e-05,
+      "loss": 0.6307,
+      "step": 187500
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 4.158284645923124e-05,
+      "loss": 0.6356,
+      "step": 188000
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 4.155910808272729e-05,
+      "loss": 0.6269,
+      "step": 188500
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 4.1535369706223346e-05,
+      "loss": 0.6368,
+      "step": 189000
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 4.15116313297194e-05,
+      "loss": 0.6282,
+      "step": 189500
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 4.1487892953215464e-05,
+      "loss": 0.6226,
+      "step": 190000
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 4.146415457671152e-05,
+      "loss": 0.6181,
+      "step": 190500
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 4.144041620020757e-05,
+      "loss": 0.6138,
+      "step": 191000
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 4.1416677823703623e-05,
+      "loss": 0.6184,
+      "step": 191500
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 4.139298692395269e-05,
+      "loss": 0.624,
+      "step": 192000
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 4.136924854744874e-05,
+      "loss": 0.6125,
+      "step": 192500
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 4.13455101709448e-05,
+      "loss": 0.6261,
+      "step": 193000
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 4.1321771794440857e-05,
+      "loss": 0.6189,
+      "step": 193500
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 4.129803341793691e-05,
+      "loss": 0.6149,
+      "step": 194000
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 4.127429504143297e-05,
+      "loss": 0.6296,
+      "step": 194500
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 4.1250556664929016e-05,
+      "loss": 0.6263,
+      "step": 195000
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 4.122681828842507e-05,
+      "loss": 0.6298,
+      "step": 195500
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 4.120312738867414e-05,
+      "loss": 0.6102,
+      "step": 196000
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 4.1179389012170194e-05,
+      "loss": 0.6248,
+      "step": 196500
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 4.115565063566625e-05,
+      "loss": 0.6121,
+      "step": 197000
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 4.1131912259162305e-05,
+      "loss": 0.6156,
+      "step": 197500
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 4.1108221359411365e-05,
+      "loss": 0.6076,
+      "step": 198000
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 4.1084577936413436e-05,
+      "loss": 0.6231,
+      "step": 198500
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 4.106083955990949e-05,
+      "loss": 0.6281,
+      "step": 199000
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 4.103710118340555e-05,
+      "loss": 0.6157,
+      "step": 199500
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 4.10133628069016e-05,
+      "loss": 0.6086,
+      "step": 200000
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 4.098962443039766e-05,
+      "loss": 0.6225,
+      "step": 200500
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 4.096588605389371e-05,
+      "loss": 0.6078,
+      "step": 201000
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 4.094214767738977e-05,
+      "loss": 0.6217,
+      "step": 201500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 4.0918409300885824e-05,
+      "loss": 0.6096,
+      "step": 202000
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 4.089467092438188e-05,
+      "loss": 0.6253,
+      "step": 202500
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 4.087102750138395e-05,
+      "loss": 0.6068,
+      "step": 203000
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 4.0847289124880006e-05,
+      "loss": 0.613,
+      "step": 203500
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 4.082355074837606e-05,
+      "loss": 0.6179,
+      "step": 204000
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 4.079981237187212e-05,
+      "loss": 0.6312,
+      "step": 204500
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 4.077612147212118e-05,
+      "loss": 0.601,
+      "step": 205000
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 4.075238309561723e-05,
+      "loss": 0.615,
+      "step": 205500
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 4.072864471911329e-05,
+      "loss": 0.6071,
+      "step": 206000
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 4.0704906342609344e-05,
+      "loss": 0.6135,
+      "step": 206500
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 4.06811679661054e-05,
+      "loss": 0.6073,
+      "step": 207000
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 4.065747706635446e-05,
+      "loss": 0.6073,
+      "step": 207500
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 4.0633738689850515e-05,
+      "loss": 0.6053,
+      "step": 208000
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 4.061000031334658e-05,
+      "loss": 0.6116,
+      "step": 208500
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 4.058626193684263e-05,
+      "loss": 0.606,
+      "step": 209000
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 4.056252356033868e-05,
+      "loss": 0.6183,
+      "step": 209500
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 4.0538785183834736e-05,
+      "loss": 0.601,
+      "step": 210000
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 4.051504680733079e-05,
+      "loss": 0.6041,
+      "step": 210500
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 4.049130843082685e-05,
+      "loss": 0.6073,
+      "step": 211000
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 4.04675700543229e-05,
+      "loss": 0.6106,
+      "step": 211500
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 4.044383167781896e-05,
+      "loss": 0.6105,
+      "step": 212000
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.0420093301315014e-05,
+      "loss": 0.5973,
+      "step": 212500
+    },
+    {
+      "epoch": 3.0,
+      "eval_bleu": 43.4697,
+      "eval_gen_len": 13.6556,
+      "eval_loss": 0.9142627120018005,
+      "eval_runtime": 9819.0968,
+      "eval_samples_per_second": 14.445,
+      "eval_steps_per_second": 1.806,
+      "step": 212758
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.039635492481107e-05,
+      "loss": 0.6014,
+      "step": 213000
+    },
+    {
+      "epoch": 3.01,
+      "learning_rate": 4.037266402506013e-05,
+      "loss": 0.6099,
+      "step": 213500
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 4.0348925648556185e-05,
+      "loss": 0.5951,
+      "step": 214000
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 4.032518727205224e-05,
+      "loss": 0.597,
+      "step": 214500
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 4.03014488955483e-05,
+      "loss": 0.596,
+      "step": 215000
+    },
+    {
+      "epoch": 3.04,
+      "learning_rate": 4.027771051904435e-05,
+      "loss": 0.5914,
+      "step": 215500
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 4.0253972142540406e-05,
+      "loss": 0.5979,
+      "step": 216000
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 4.023023376603646e-05,
+      "loss": 0.5855,
+      "step": 216500
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 4.020649538953252e-05,
+      "loss": 0.5887,
+      "step": 217000
+    },
+    {
+      "epoch": 3.07,
+      "learning_rate": 4.018275701302857e-05,
+      "loss": 0.5853,
+      "step": 217500
+    },
+    {
+      "epoch": 3.07,
+      "learning_rate": 4.015901863652463e-05,
+      "loss": 0.5921,
+      "step": 218000
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 4.013528026002068e-05,
+      "loss": 0.5873,
+      "step": 218500
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 4.0111589360269744e-05,
+      "loss": 0.5708,
+      "step": 219000
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 4.00878509837658e-05,
+      "loss": 0.5777,
+      "step": 219500
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 4.0064112607261855e-05,
+      "loss": 0.5772,
+      "step": 220000
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 4.004037423075791e-05,
+      "loss": 0.573,
+      "step": 220500
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 4.0016635854253966e-05,
+      "loss": 0.5734,
+      "step": 221000
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 3.999289747775002e-05,
+      "loss": 0.5682,
+      "step": 221500
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 3.996920657799909e-05,
+      "loss": 0.5628,
+      "step": 222000
+    },
+    {
+      "epoch": 3.14,
+      "learning_rate": 3.9945468201495136e-05,
+      "loss": 0.5687,
+      "step": 222500
+    },
+    {
+      "epoch": 3.14,
+      "learning_rate": 3.992172982499119e-05,
+      "loss": 0.5569,
+      "step": 223000
+    },
+    {
+      "epoch": 3.15,
+      "learning_rate": 3.989799144848725e-05,
+      "loss": 0.5571,
+      "step": 223500
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 3.98742530719833e-05,
+      "loss": 0.5619,
+      "step": 224000
+    },
+    {
+      "epoch": 3.17,
+      "learning_rate": 3.9850514695479365e-05,
+      "loss": 0.5635,
+      "step": 224500
+    },
+    {
+      "epoch": 3.17,
+      "learning_rate": 3.982677631897542e-05,
+      "loss": 0.5538,
+      "step": 225000
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 3.980303794247147e-05,
+      "loss": 0.5494,
+      "step": 225500
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 3.9779299565967525e-05,
+      "loss": 0.5532,
+      "step": 226000
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 3.975556118946358e-05,
+      "loss": 0.5576,
+      "step": 226500
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 3.973187028971264e-05,
+      "loss": 0.5499,
+      "step": 227000
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 3.970822686671472e-05,
+      "loss": 0.5434,
+      "step": 227500
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 3.968448849021077e-05,
+      "loss": 0.5421,
+      "step": 228000
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 3.966075011370682e-05,
+      "loss": 0.5377,
+      "step": 228500
+    },
+    {
+      "epoch": 3.23,
+      "learning_rate": 3.963701173720288e-05,
+      "loss": 0.5614,
+      "step": 229000
+    },
+    {
+      "epoch": 3.24,
+      "learning_rate": 3.961327336069894e-05,
+      "loss": 0.5458,
+      "step": 229500
+    },
+    {
+      "epoch": 3.24,
+      "learning_rate": 3.9589534984194995e-05,
+      "loss": 0.5483,
+      "step": 230000
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 3.9565796607691044e-05,
+      "loss": 0.5423,
+      "step": 230500
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 3.95420582311871e-05,
+      "loss": 0.5346,
+      "step": 231000
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 3.9518319854683155e-05,
+      "loss": 0.5334,
+      "step": 231500
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 3.9494628954932215e-05,
+      "loss": 0.5451,
+      "step": 232000
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 3.947089057842828e-05,
+      "loss": 0.5378,
+      "step": 232500
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 3.944715220192433e-05,
+      "loss": 0.5323,
+      "step": 233000
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 3.942346130217339e-05,
+      "loss": 0.5346,
+      "step": 233500
+    },
+    {
+      "epoch": 3.3,
+      "learning_rate": 3.939972292566945e-05,
+      "loss": 0.5334,
+      "step": 234000
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 3.9375984549165504e-05,
+      "loss": 0.5292,
+      "step": 234500
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 3.935224617266155e-05,
+      "loss": 0.5442,
+      "step": 235000
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 3.9328507796157615e-05,
+      "loss": 0.5423,
+      "step": 235500
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 3.930476941965367e-05,
+      "loss": 0.53,
+      "step": 236000
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 3.9281031043149725e-05,
+      "loss": 0.5359,
+      "step": 236500
+    },
+    {
+      "epoch": 3.34,
+      "learning_rate": 3.9257340143398785e-05,
+      "loss": 0.5432,
+      "step": 237000
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 3.923360176689484e-05,
+      "loss": 0.5325,
+      "step": 237500
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 3.9209863390390896e-05,
+      "loss": 0.53,
+      "step": 238000
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 3.918612501388695e-05,
+      "loss": 0.5347,
+      "step": 238500
+    },
+    {
+      "epoch": 3.37,
+      "learning_rate": 3.916238663738301e-05,
+      "loss": 0.5268,
+      "step": 239000
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 3.913864826087906e-05,
+      "loss": 0.5233,
+      "step": 239500
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 3.911490988437512e-05,
+      "loss": 0.5281,
+      "step": 240000
+    },
+    {
+      "epoch": 3.39,
+      "learning_rate": 3.9091171507871174e-05,
+      "loss": 0.5191,
+      "step": 240500
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 3.906743313136723e-05,
+      "loss": 0.5226,
+      "step": 241000
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 3.904369475486328e-05,
+      "loss": 0.525,
+      "step": 241500
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 3.901995637835934e-05,
+      "loss": 0.5252,
+      "step": 242000
+    },
+    {
+      "epoch": 3.42,
+      "learning_rate": 3.8996218001855396e-05,
+      "loss": 0.5406,
+      "step": 242500
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 3.897257457885746e-05,
+      "loss": 0.5168,
+      "step": 243000
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 3.8948836202353515e-05,
+      "loss": 0.5218,
+      "step": 243500
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 3.892509782584958e-05,
+      "loss": 0.5304,
+      "step": 244000
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 3.890135944934563e-05,
+      "loss": 0.5217,
+      "step": 244500
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 3.887762107284169e-05,
+      "loss": 0.5143,
+      "step": 245000
+    },
+    {
+      "epoch": 3.46,
+      "learning_rate": 3.885388269633774e-05,
+      "loss": 0.5326,
+      "step": 245500
+    },
+    {
+      "epoch": 3.47,
+      "learning_rate": 3.883014431983379e-05,
+      "loss": 0.5152,
+      "step": 246000
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 3.880640594332985e-05,
+      "loss": 0.5288,
+      "step": 246500
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 3.8782667566825904e-05,
+      "loss": 0.517,
+      "step": 247000
+    },
+    {
+      "epoch": 3.49,
+      "learning_rate": 3.8758929190321966e-05,
+      "loss": 0.5092,
+      "step": 247500
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 3.8735238290571026e-05,
+      "loss": 0.5147,
+      "step": 248000
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 3.871149991406708e-05,
+      "loss": 0.5175,
+      "step": 248500
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 3.868780901431614e-05,
+      "loss": 0.4983,
+      "step": 249000
+    },
+    {
+      "epoch": 3.52,
+      "learning_rate": 3.866411811456521e-05,
+      "loss": 0.5145,
+      "step": 249500
+    },
+    {
+      "epoch": 3.53,
+      "learning_rate": 3.8640379738061264e-05,
+      "loss": 0.5226,
+      "step": 250000
+    },
+    {
+      "epoch": 3.53,
+      "learning_rate": 3.861664136155731e-05,
+      "loss": 0.5148,
+      "step": 250500
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 3.859290298505337e-05,
+      "loss": 0.5172,
+      "step": 251000
+    },
+    {
+      "epoch": 3.55,
+      "learning_rate": 3.856916460854942e-05,
+      "loss": 0.5245,
+      "step": 251500
+    },
+    {
+      "epoch": 3.55,
+      "learning_rate": 3.854547370879849e-05,
+      "loss": 0.5164,
+      "step": 252000
+    },
+    {
+      "epoch": 3.56,
+      "learning_rate": 3.8521735332294545e-05,
+      "loss": 0.5087,
+      "step": 252500
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 3.84979969557906e-05,
+      "loss": 0.517,
+      "step": 253000
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 3.8474258579286656e-05,
+      "loss": 0.5089,
+      "step": 253500
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 3.8450520202782705e-05,
+      "loss": 0.5045,
+      "step": 254000
+    },
+    {
+      "epoch": 3.59,
+      "learning_rate": 3.842682930303177e-05,
+      "loss": 0.5169,
+      "step": 254500
+    },
+    {
+      "epoch": 3.6,
+      "learning_rate": 3.840309092652783e-05,
+      "loss": 0.5204,
+      "step": 255000
+    },
+    {
+      "epoch": 3.6,
+      "learning_rate": 3.837935255002388e-05,
+      "loss": 0.5064,
+      "step": 255500
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 3.835561417351994e-05,
+      "loss": 0.5074,
+      "step": 256000
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 3.8331875797015994e-05,
+      "loss": 0.5147,
+      "step": 256500
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 3.830813742051205e-05,
+      "loss": 0.5031,
+      "step": 257000
+    },
+    {
+      "epoch": 3.63,
+      "learning_rate": 3.8284399044008104e-05,
+      "loss": 0.5046,
+      "step": 257500
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 3.826066066750415e-05,
+      "loss": 0.5023,
+      "step": 258000
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 3.8236922291000215e-05,
+      "loss": 0.5053,
+      "step": 258500
+    },
+    {
+      "epoch": 3.65,
+      "learning_rate": 3.821318391449627e-05,
+      "loss": 0.5145,
+      "step": 259000
+    },
+    {
+      "epoch": 3.66,
+      "learning_rate": 3.8189445537992326e-05,
+      "loss": 0.5037,
+      "step": 259500
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 3.816570716148838e-05,
+      "loss": 0.5164,
+      "step": 260000
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 3.814196878498443e-05,
+      "loss": 0.5089,
+      "step": 260500
+    },
+    {
+      "epoch": 3.68,
+      "learning_rate": 3.8118230408480486e-05,
+      "loss": 0.499,
+      "step": 261000
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 3.809453950872955e-05,
+      "loss": 0.5004,
+      "step": 261500
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 3.807084860897861e-05,
+      "loss": 0.4955,
+      "step": 262000
+    },
+    {
+      "epoch": 3.7,
+      "learning_rate": 3.804715770922768e-05,
+      "loss": 0.5021,
+      "step": 262500
+    },
+    {
+      "epoch": 3.71,
+      "learning_rate": 3.802341933272373e-05,
+      "loss": 0.5064,
+      "step": 263000
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 3.799968095621979e-05,
+      "loss": 0.4947,
+      "step": 263500
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 3.7975942579715846e-05,
+      "loss": 0.5033,
+      "step": 264000
+    },
+    {
+      "epoch": 3.73,
+      "learning_rate": 3.79522042032119e-05,
+      "loss": 0.5022,
+      "step": 264500
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 3.792846582670796e-05,
+      "loss": 0.4956,
+      "step": 265000
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 3.790477492695702e-05,
+      "loss": 0.5102,
+      "step": 265500
+    },
+    {
+      "epoch": 3.75,
+      "learning_rate": 3.788103655045307e-05,
+      "loss": 0.509,
+      "step": 266000
+    },
+    {
+      "epoch": 3.76,
+      "learning_rate": 3.785729817394913e-05,
+      "loss": 0.5048,
+      "step": 266500
+    },
+    {
+      "epoch": 3.76,
+      "learning_rate": 3.783355979744518e-05,
+      "loss": 0.4954,
+      "step": 267000
+    },
+    {
+      "epoch": 3.77,
+      "learning_rate": 3.780982142094124e-05,
+      "loss": 0.5009,
+      "step": 267500
+    },
+    {
+      "epoch": 3.78,
+      "learning_rate": 3.7786083044437294e-05,
+      "loss": 0.4909,
+      "step": 268000
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 3.776234466793335e-05,
+      "loss": 0.5018,
+      "step": 268500
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 3.77386062914294e-05,
+      "loss": 0.4903,
+      "step": 269000
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 3.7714867914925454e-05,
+      "loss": 0.504,
+      "step": 269500
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 3.7691129538421516e-05,
+      "loss": 0.5088,
+      "step": 270000
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 3.7667438638670576e-05,
+      "loss": 0.5009,
+      "step": 270500
+    },
+    {
+      "epoch": 3.82,
+      "learning_rate": 3.7643747738919636e-05,
+      "loss": 0.4924,
+      "step": 271000
+    },
+    {
+      "epoch": 3.83,
+      "learning_rate": 3.76200568391687e-05,
+      "loss": 0.4999,
+      "step": 271500
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 3.759631846266476e-05,
+      "loss": 0.4997,
+      "step": 272000
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 3.7572580086160813e-05,
+      "loss": 0.4968,
+      "step": 272500
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 3.754884170965687e-05,
+      "loss": 0.4905,
+      "step": 273000
+    },
+    {
+      "epoch": 3.86,
+      "learning_rate": 3.7525103333152924e-05,
+      "loss": 0.5055,
+      "step": 273500
+    },
+    {
+      "epoch": 3.86,
+      "learning_rate": 3.750136495664898e-05,
+      "loss": 0.4918,
+      "step": 274000
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 3.747762658014503e-05,
+      "loss": 0.4947,
+      "step": 274500
+    },
+    {
+      "epoch": 3.88,
+      "learning_rate": 3.745388820364109e-05,
+      "loss": 0.5016,
+      "step": 275000
+    },
+    {
+      "epoch": 3.88,
+      "learning_rate": 3.7430149827137146e-05,
+      "loss": 0.5084,
+      "step": 275500
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 3.74064114506332e-05,
+      "loss": 0.4882,
+      "step": 276000
+    },
+    {
+      "epoch": 3.9,
+      "learning_rate": 3.738272055088226e-05,
+      "loss": 0.4948,
+      "step": 276500
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 3.735898217437832e-05,
+      "loss": 0.4971,
+      "step": 277000
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 3.733524379787437e-05,
+      "loss": 0.4867,
+      "step": 277500
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 3.731150542137043e-05,
+      "loss": 0.4917,
+      "step": 278000
+    },
+    {
+      "epoch": 3.93,
+      "learning_rate": 3.7287767044866484e-05,
+      "loss": 0.4853,
+      "step": 278500
+    },
+    {
+      "epoch": 3.93,
+      "learning_rate": 3.726402866836254e-05,
+      "loss": 0.4948,
+      "step": 279000
+    },
+    {
+      "epoch": 3.94,
+      "learning_rate": 3.7240290291858594e-05,
+      "loss": 0.4935,
+      "step": 279500
+    },
+    {
+      "epoch": 3.95,
+      "learning_rate": 3.721655191535465e-05,
+      "loss": 0.4906,
+      "step": 280000
+    },
+    {
+      "epoch": 3.96,
+      "learning_rate": 3.71928135388507e-05,
+      "loss": 0.4992,
+      "step": 280500
+    },
+    {
+      "epoch": 3.96,
+      "learning_rate": 3.716907516234676e-05,
+      "loss": 0.4814,
+      "step": 281000
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 3.7145336785842816e-05,
+      "loss": 0.4903,
+      "step": 281500
+    },
+    {
+      "epoch": 3.98,
+      "learning_rate": 3.712159840933887e-05,
+      "loss": 0.4869,
+      "step": 282000
+    },
+    {
+      "epoch": 3.98,
+      "learning_rate": 3.709786003283493e-05,
+      "loss": 0.493,
+      "step": 282500
+    },
+    {
+      "epoch": 3.99,
+      "learning_rate": 3.7074121656330976e-05,
+      "loss": 0.4909,
+      "step": 283000
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.705038327982703e-05,
+      "loss": 0.4873,
+      "step": 283500
+    },
+    {
+      "epoch": 4.0,
+      "eval_bleu": 43.6153,
+      "eval_gen_len": 13.6544,
+      "eval_loss": 0.9757916927337646,
+      "eval_runtime": 9810.7798,
+      "eval_samples_per_second": 14.457,
+      "eval_steps_per_second": 1.807,
+      "step": 283678
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.70266923800761e-05,
+      "loss": 0.4799,
+      "step": 284000
+    },
+    {
+      "epoch": 4.01,
+      "learning_rate": 3.700300148032516e-05,
+      "loss": 0.4853,
+      "step": 284500
+    },
+    {
+      "epoch": 4.02,
+      "learning_rate": 3.6979263103821214e-05,
+      "loss": 0.4872,
+      "step": 285000
+    },
+    {
+      "epoch": 4.03,
+      "learning_rate": 3.695552472731727e-05,
+      "loss": 0.4809,
+      "step": 285500
+    },
+    {
+      "epoch": 4.03,
+      "learning_rate": 3.6931786350813324e-05,
+      "loss": 0.4787,
+      "step": 286000
+    },
+    {
+      "epoch": 4.04,
+      "learning_rate": 3.690804797430938e-05,
+      "loss": 0.4797,
+      "step": 286500
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 3.6884309597805435e-05,
+      "loss": 0.4811,
+      "step": 287000
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 3.686057122130149e-05,
+      "loss": 0.4728,
+      "step": 287500
+    },
+    {
+      "epoch": 4.06,
+      "learning_rate": 3.6836832844797546e-05,
+      "loss": 0.4756,
+      "step": 288000
+    },
+    {
+      "epoch": 4.07,
+      "learning_rate": 3.68130944682936e-05,
+      "loss": 0.4728,
+      "step": 288500
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 3.678935609178966e-05,
+      "loss": 0.4844,
+      "step": 289000
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 3.676561771528571e-05,
+      "loss": 0.4677,
+      "step": 289500
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 3.674192681553478e-05,
+      "loss": 0.4631,
+      "step": 290000
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 3.6718188439030835e-05,
+      "loss": 0.4626,
+      "step": 290500
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 3.6694450062526884e-05,
+      "loss": 0.4705,
+      "step": 291000
+    },
+    {
+      "epoch": 4.11,
+      "learning_rate": 3.667071168602294e-05,
+      "loss": 0.4582,
+      "step": 291500
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 3.6646973309518994e-05,
+      "loss": 0.4668,
+      "step": 292000
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 3.662323493301505e-05,
+      "loss": 0.4597,
+      "step": 292500
+    },
+    {
+      "epoch": 4.13,
+      "learning_rate": 3.6599496556511105e-05,
+      "loss": 0.4518,
+      "step": 293000
+    },
+    {
+      "epoch": 4.14,
+      "learning_rate": 3.657575818000716e-05,
+      "loss": 0.4586,
+      "step": 293500
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 3.6552019803503216e-05,
+      "loss": 0.4413,
+      "step": 294000
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 3.6528328903752276e-05,
+      "loss": 0.4528,
+      "step": 294500
+    },
+    {
+      "epoch": 4.16,
+      "learning_rate": 3.650459052724833e-05,
+      "loss": 0.4512,
+      "step": 295000
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 3.64808996274974e-05,
+      "loss": 0.4555,
+      "step": 295500
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 3.6457161250993454e-05,
+      "loss": 0.4474,
+      "step": 296000
+    },
+    {
+      "epoch": 4.18,
+      "learning_rate": 3.643342287448951e-05,
+      "loss": 0.4426,
+      "step": 296500
+    },
+    {
+      "epoch": 4.19,
+      "learning_rate": 3.6409684497985565e-05,
+      "loss": 0.4459,
+      "step": 297000
+    },
+    {
+      "epoch": 4.19,
+      "learning_rate": 3.6385993598234625e-05,
+      "loss": 0.4473,
+      "step": 297500
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 3.636225522173068e-05,
+      "loss": 0.4426,
+      "step": 298000
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 3.6338516845226736e-05,
+      "loss": 0.4413,
+      "step": 298500
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 3.631477846872279e-05,
+      "loss": 0.4335,
+      "step": 299000
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 3.629104009221885e-05,
+      "loss": 0.4422,
+      "step": 299500
+    },
+    {
+      "epoch": 4.23,
+      "learning_rate": 3.626734919246791e-05,
+      "loss": 0.4566,
+      "step": 300000
+    },
+    {
+      "epoch": 4.24,
+      "learning_rate": 3.624361081596396e-05,
+      "loss": 0.4375,
+      "step": 300500
+    },
+    {
+      "epoch": 4.24,
+      "learning_rate": 3.621987243946002e-05,
+      "loss": 0.4433,
+      "step": 301000
+    },
+    {
+      "epoch": 4.25,
+      "learning_rate": 3.6196181539709084e-05,
+      "loss": 0.4311,
+      "step": 301500
+    },
+    {
+      "epoch": 4.26,
+      "learning_rate": 3.617244316320514e-05,
+      "loss": 0.4282,
+      "step": 302000
+    },
+    {
+      "epoch": 4.27,
+      "learning_rate": 3.6148704786701195e-05,
+      "loss": 0.4302,
+      "step": 302500
+    },
+    {
+      "epoch": 4.27,
+      "learning_rate": 3.612496641019725e-05,
+      "loss": 0.446,
+      "step": 303000
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 3.61012280336933e-05,
+      "loss": 0.4282,
+      "step": 303500
+    },
+    {
+      "epoch": 4.29,
+      "learning_rate": 3.607748965718936e-05,
+      "loss": 0.4335,
+      "step": 304000
+    },
+    {
+      "epoch": 4.29,
+      "learning_rate": 3.605375128068542e-05,
+      "loss": 0.433,
+      "step": 304500
+    },
+    {
+      "epoch": 4.3,
+      "learning_rate": 3.603001290418147e-05,
+      "loss": 0.4168,
+      "step": 305000
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 3.600627452767753e-05,
+      "loss": 0.4391,
+      "step": 305500
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 3.598253615117358e-05,
+      "loss": 0.433,
+      "step": 306000
+    },
+    {
+      "epoch": 4.32,
+      "learning_rate": 3.595879777466963e-05,
+      "loss": 0.4391,
+      "step": 306500
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 3.593505939816569e-05,
+      "loss": 0.4276,
+      "step": 307000
+    },
+    {
+      "epoch": 4.34,
+      "learning_rate": 3.591132102166174e-05,
+      "loss": 0.4343,
+      "step": 307500
+    },
+    {
+      "epoch": 4.34,
+      "learning_rate": 3.5887582645157805e-05,
+      "loss": 0.4407,
+      "step": 308000
+    },
+    {
+      "epoch": 4.35,
+      "learning_rate": 3.5863891745406865e-05,
+      "loss": 0.4258,
+      "step": 308500
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 3.584015336890292e-05,
+      "loss": 0.4285,
+      "step": 309000
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 3.581641499239897e-05,
+      "loss": 0.4316,
+      "step": 309500
+    },
+    {
+      "epoch": 4.37,
+      "learning_rate": 3.5792676615895025e-05,
+      "loss": 0.4249,
+      "step": 310000
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 3.576898571614409e-05,
+      "loss": 0.4276,
+      "step": 310500
+    },
+    {
+      "epoch": 4.39,
+      "learning_rate": 3.574524733964015e-05,
+      "loss": 0.4286,
+      "step": 311000
+    },
+    {
+      "epoch": 4.39,
+      "learning_rate": 3.57215089631362e-05,
+      "loss": 0.4195,
+      "step": 311500
+    },
+    {
+      "epoch": 4.4,
+      "learning_rate": 3.569777058663226e-05,
+      "loss": 0.419,
+      "step": 312000
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 3.567407968688132e-05,
+      "loss": 0.422,
+      "step": 312500
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 3.5650388787130385e-05,
+      "loss": 0.4299,
+      "step": 313000
+    },
+    {
+      "epoch": 4.42,
+      "learning_rate": 3.562665041062644e-05,
+      "loss": 0.4341,
+      "step": 313500
+    },
+    {
+      "epoch": 4.43,
+      "learning_rate": 3.5602912034122496e-05,
+      "loss": 0.4122,
+      "step": 314000
+    },
+    {
+      "epoch": 4.43,
+      "learning_rate": 3.5579173657618544e-05,
+      "loss": 0.4251,
+      "step": 314500
+    },
+    {
+      "epoch": 4.44,
+      "learning_rate": 3.55554352811146e-05,
+      "loss": 0.4298,
+      "step": 315000
+    },
+    {
+      "epoch": 4.45,
+      "learning_rate": 3.553169690461066e-05,
+      "loss": 0.4183,
+      "step": 315500
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 3.550795852810672e-05,
+      "loss": 0.4199,
+      "step": 316000
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 3.548426762835578e-05,
+      "loss": 0.4281,
+      "step": 316500
+    },
+    {
+      "epoch": 4.47,
+      "learning_rate": 3.546052925185183e-05,
+      "loss": 0.4158,
+      "step": 317000
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 3.543679087534789e-05,
+      "loss": 0.4267,
+      "step": 317500
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 3.541305249884394e-05,
+      "loss": 0.4193,
+      "step": 318000
+    },
+    {
+      "epoch": 4.49,
+      "learning_rate": 3.538931412234e-05,
+      "loss": 0.4106,
+      "step": 318500
+    },
+    {
+      "epoch": 4.5,
+      "learning_rate": 3.5365575745836055e-05,
+      "loss": 0.4186,
+      "step": 319000
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 3.534183736933211e-05,
+      "loss": 0.4141,
+      "step": 319500
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 3.5318098992828166e-05,
+      "loss": 0.398,
+      "step": 320000
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 3.529436061632422e-05,
+      "loss": 0.4241,
+      "step": 320500
+    },
+    {
+      "epoch": 4.53,
+      "learning_rate": 3.527062223982027e-05,
+      "loss": 0.422,
+      "step": 321000
+    },
+    {
+      "epoch": 4.53,
+      "learning_rate": 3.5246883863316325e-05,
+      "loss": 0.4167,
+      "step": 321500
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 3.522319296356539e-05,
+      "loss": 0.4214,
+      "step": 322000
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 3.519950206381445e-05,
+      "loss": 0.4171,
+      "step": 322500
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 3.517576368731051e-05,
+      "loss": 0.4211,
+      "step": 323000
+    },
+    {
+      "epoch": 4.56,
+      "learning_rate": 3.5152072787559574e-05,
+      "loss": 0.41,
+      "step": 323500
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 3.512833441105563e-05,
+      "loss": 0.4243,
+      "step": 324000
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 3.5104596034551685e-05,
+      "loss": 0.4066,
+      "step": 324500
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 3.508085765804774e-05,
+      "loss": 0.4055,
+      "step": 325000
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 3.5057119281543796e-05,
+      "loss": 0.419,
+      "step": 325500
+    },
+    {
+      "epoch": 4.6,
+      "learning_rate": 3.5033380905039845e-05,
+      "loss": 0.4222,
+      "step": 326000
+    },
+    {
+      "epoch": 4.6,
+      "learning_rate": 3.50096425285359e-05,
+      "loss": 0.4092,
+      "step": 326500
+    },
+    {
+      "epoch": 4.61,
+      "learning_rate": 3.498595162878497e-05,
+      "loss": 0.4094,
+      "step": 327000
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 3.496221325228102e-05,
+      "loss": 0.4102,
+      "step": 327500
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 3.493847487577708e-05,
+      "loss": 0.4096,
+      "step": 328000
+    },
+    {
+      "epoch": 4.63,
+      "learning_rate": 3.4914736499273133e-05,
+      "loss": 0.4052,
+      "step": 328500
+    },
+    {
+      "epoch": 4.64,
+      "learning_rate": 3.489099812276919e-05,
+      "loss": 0.4076,
+      "step": 329000
+    },
+    {
+      "epoch": 4.65,
+      "learning_rate": 3.486725974626524e-05,
+      "loss": 0.414,
+      "step": 329500
+    },
+    {
+      "epoch": 4.65,
+      "learning_rate": 3.48435213697613e-05,
+      "loss": 0.4084,
+      "step": 330000
+    },
+    {
+      "epoch": 4.66,
+      "learning_rate": 3.4819782993257355e-05,
+      "loss": 0.405,
+      "step": 330500
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 3.479604461675341e-05,
+      "loss": 0.4167,
+      "step": 331000
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 3.4772306240249466e-05,
+      "loss": 0.4113,
+      "step": 331500
+    },
+    {
+      "epoch": 4.68,
+      "learning_rate": 3.4748615340498526e-05,
+      "loss": 0.403,
+      "step": 332000
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 3.472492444074759e-05,
+      "loss": 0.4066,
+      "step": 332500
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 3.470118606424365e-05,
+      "loss": 0.3965,
+      "step": 333000
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 3.467749516449271e-05,
+      "loss": 0.4052,
+      "step": 333500
+    },
+    {
+      "epoch": 4.71,
+      "learning_rate": 3.4653756787988764e-05,
+      "loss": 0.4093,
+      "step": 334000
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 3.463001841148481e-05,
+      "loss": 0.4015,
+      "step": 334500
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 3.4606280034980875e-05,
+      "loss": 0.4096,
+      "step": 335000
+    },
+    {
+      "epoch": 4.73,
+      "learning_rate": 3.458254165847693e-05,
+      "loss": 0.4041,
+      "step": 335500
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 3.4558803281972986e-05,
+      "loss": 0.4046,
+      "step": 336000
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 3.453506490546904e-05,
+      "loss": 0.4112,
+      "step": 336500
+    },
+    {
+      "epoch": 4.75,
+      "learning_rate": 3.4511326528965097e-05,
+      "loss": 0.4107,
+      "step": 337000
+    },
+    {
+      "epoch": 4.76,
+      "learning_rate": 3.4487588152461145e-05,
+      "loss": 0.406,
+      "step": 337500
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 3.44638497759572e-05,
+      "loss": 0.4008,
+      "step": 338000
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 3.444011139945326e-05,
+      "loss": 0.4088,
+      "step": 338500
+    },
+    {
+      "epoch": 4.78,
+      "learning_rate": 3.441637302294932e-05,
+      "loss": 0.4006,
+      "step": 339000
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 3.4392634646445374e-05,
+      "loss": 0.3962,
+      "step": 339500
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 3.436889626994142e-05,
+      "loss": 0.4048,
+      "step": 340000
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 3.434515789343748e-05,
+      "loss": 0.4076,
+      "step": 340500
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 3.4321419516933533e-05,
+      "loss": 0.4108,
+      "step": 341000
+    },
+    {
+      "epoch": 4.82,
+      "learning_rate": 3.429768114042959e-05,
+      "loss": 0.4037,
+      "step": 341500
+    },
+    {
+      "epoch": 4.82,
+      "learning_rate": 3.4273942763925644e-05,
+      "loss": 0.4019,
+      "step": 342000
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 3.42502043874217e-05,
+      "loss": 0.3984,
+      "step": 342500
+    },
+    {
+      "epoch": 4.84,
+      "learning_rate": 3.4226466010917755e-05,
+      "loss": 0.4065,
+      "step": 343000
+    },
+    {
+      "epoch": 4.84,
+      "learning_rate": 3.4202775111166815e-05,
+      "loss": 0.3995,
+      "step": 343500
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 3.417903673466287e-05,
+      "loss": 0.3979,
+      "step": 344000
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 3.4155298358158926e-05,
+      "loss": 0.4088,
+      "step": 344500
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 3.413155998165499e-05,
+      "loss": 0.3951,
+      "step": 345000
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 3.4107821605151044e-05,
+      "loss": 0.4011,
+      "step": 345500
+    },
+    {
+      "epoch": 4.88,
+      "learning_rate": 3.408408322864709e-05,
+      "loss": 0.4073,
+      "step": 346000
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 3.406034485214315e-05,
+      "loss": 0.4042,
+      "step": 346500
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 3.4036606475639203e-05,
+      "loss": 0.3924,
+      "step": 347000
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 3.401286809913526e-05,
+      "loss": 0.3928,
+      "step": 347500
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 3.3989177199384326e-05,
+      "loss": 0.4044,
+      "step": 348000
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 3.396543882288038e-05,
+      "loss": 0.3934,
+      "step": 348500
+    },
+    {
+      "epoch": 4.92,
+      "learning_rate": 3.394170044637644e-05,
+      "loss": 0.3946,
+      "step": 349000
+    },
+    {
+      "epoch": 4.93,
+      "learning_rate": 3.391796206987249e-05,
+      "loss": 0.3907,
+      "step": 349500
+    },
+    {
+      "epoch": 4.94,
+      "learning_rate": 3.389427117012155e-05,
+      "loss": 0.3947,
+      "step": 350000
+    },
+    {
+      "epoch": 4.94,
+      "learning_rate": 3.387062774712362e-05,
+      "loss": 0.3962,
+      "step": 350500
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 3.384688937061968e-05,
+      "loss": 0.3961,
+      "step": 351000
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 3.3823150994115734e-05,
+      "loss": 0.4004,
+      "step": 351500
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 3.379941261761179e-05,
+      "loss": 0.3917,
+      "step": 352000
+    },
+    {
+      "epoch": 4.97,
+      "learning_rate": 3.3775721717860856e-05,
+      "loss": 0.3939,
+      "step": 352500
+    },
+    {
+      "epoch": 4.98,
+      "learning_rate": 3.3752030818109916e-05,
+      "loss": 0.3942,
+      "step": 353000
+    },
+    {
+      "epoch": 4.98,
+      "learning_rate": 3.372829244160597e-05,
+      "loss": 0.3975,
+      "step": 353500
+    },
+    {
+      "epoch": 4.99,
+      "learning_rate": 3.370455406510202e-05,
+      "loss": 0.4001,
+      "step": 354000
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.3680815688598076e-05,
+      "loss": 0.388,
+      "step": 354500
+    },
+    {
+      "epoch": 5.0,
+      "eval_bleu": 43.3763,
+      "eval_gen_len": 13.6067,
+      "eval_loss": 1.0399531126022339,
+      "eval_runtime": 9774.1481,
+      "eval_samples_per_second": 14.512,
+      "eval_steps_per_second": 1.814,
+      "step": 354597
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 1063785,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 1.2295340767110496e+19,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/checkpoint-354597/training_args.bin b/checkpoint-354597/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..dbbe22d1f032c0b0fd9eedfe9ae519ce9ccd36a7
--- /dev/null
+++ b/checkpoint-354597/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0adf1a980c6128833811b7e6eb546e117ffd3efb8c21dc7de95b5e76a5b21b8d
+size 4728
diff --git a/config.json b/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..c7d123d026fa1d1679aeb1220d9ae316d7c13a0b
--- /dev/null
+++ b/config.json
@@ -0,0 +1,59 @@
+{
+  "_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
+  "_num_labels": 3,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
+  "add_bias_logits": false,
+  "add_final_layer_norm": true,
+  "architectures": [
+    "MBartForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 0,
+  "classif_dropout": 0.0,
+  "classifier_dropout": 0.0,
+  "d_model": 1024,
+  "decoder_attention_heads": 16,
+  "decoder_ffn_dim": 4096,
+  "decoder_layerdrop": 0.0,
+  "decoder_layers": 12,
+  "decoder_start_token_id": 2,
+  "dropout": 0.1,
+  "early_stopping": true,
+  "encoder_attention_heads": 16,
+  "encoder_ffn_dim": 4096,
+  "encoder_layerdrop": 0.0,
+  "encoder_layers": 12,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 250014,
+  "forced_eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "id2label": {
+    "0": "LABEL_0",
+    "1": "LABEL_1",
+    "2": "LABEL_2"
+  },
+  "init_std": 0.02,
+  "is_encoder_decoder": true,
+  "label2id": {
+    "LABEL_0": 0,
+    "LABEL_1": 1,
+    "LABEL_2": 2
+  },
+  "max_length": 200,
+  "max_position_embeddings": 1024,
+  "model_type": "mbart",
+  "normalize_before": true,
+  "normalize_embedding": true,
+  "num_beams": 5,
+  "num_hidden_layers": 12,
+  "output_past": true,
+  "pad_token_id": 1,
+  "scale_embedding": true,
+  "static_position_embeddings": false,
+  "tokenizer_class": "MBart50Tokenizer",
+  "torch_dtype": "float32",
+  "transformers_version": "4.35.2",
+  "use_cache": true,
+  "vocab_size": 250054
+}
diff --git a/eval_results.json b/eval_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..ac8a216131ebad07b67ed29f6f7ef2a553235bb1
--- /dev/null
+++ b/eval_results.json
@@ -0,0 +1,10 @@
+{
+    "epoch": 5.0,
+    "eval_bleu": 42.7907,
+    "eval_gen_len": 13.7941,
+    "eval_loss": 0.8987888693809509,
+    "eval_runtime": 10014.5277,
+    "eval_samples": 141838,
+    "eval_samples_per_second": 14.163,
+    "eval_steps_per_second": 1.77
+}
\ No newline at end of file
diff --git a/generated_predictions.txt b/generated_predictions.txt
new file mode 100644
index 0000000000000000000000000000000000000000..c9475ad1d3b72d3df6a9a67394248a969c572b67
--- /dev/null
+++ b/generated_predictions.txt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:68f04834d8e5f3ea90c084718fffc042ddbf3c150288148528a406d46d48a385
+size 14465840
diff --git a/generation_config.json b/generation_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..7e47c12e3900189593d4b56d0d776b58a7a55627
--- /dev/null
+++ b/generation_config.json
@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 0,
+  "decoder_start_token_id": 2,
+  "early_stopping": true,
+  "eos_token_id": 2,
+  "forced_bos_token_id": 250014,
+  "forced_eos_token_id": 2,
+  "max_length": 200,
+  "num_beams": 5,
+  "pad_token_id": 1,
+  "transformers_version": "4.35.2"
+}
diff --git a/model.safetensors b/model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..ce0d737acda6b9afef4df04505c044a395af9684
--- /dev/null
+++ b/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:787d8f1a526506bcb42c60d3888520b5032131b63fc207871e65cb6cc98e53c8
+size 2444578688
diff --git a/predict_results.json b/predict_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..67dd5e4d2bb74d2adf10dc5439a1eab32c343346
--- /dev/null
+++ b/predict_results.json
@@ -0,0 +1,9 @@
+{
+    "predict_bleu": 42.8094,
+    "predict_gen_len": 13.7545,
+    "predict_loss": 0.89827561378479,
+    "predict_runtime": 9974.8511,
+    "predict_samples": 141838,
+    "predict_samples_per_second": 14.22,
+    "predict_steps_per_second": 1.777
+}
\ No newline at end of file
diff --git a/sentencepiece.bpe.model b/sentencepiece.bpe.model
new file mode 100644
index 0000000000000000000000000000000000000000..7a3f40a75f870bc1f21700cd414dc2acc431583c
--- /dev/null
+++ b/sentencepiece.bpe.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
+size 5069051
diff --git a/special_tokens_map.json b/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..92619141640d5fcbb4429807de2248352b0dca79
--- /dev/null
+++ b/special_tokens_map.json
@@ -0,0 +1,69 @@
+{
+  "additional_special_tokens": [
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "bos_token": "<s>",
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": {
+    "content": "<mask>",
+    "lstrip": true,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "unk_token": "<unk>"
+}
diff --git a/tokenizer.json b/tokenizer.json
new file mode 100644
index 0000000000000000000000000000000000000000..d7c3ce71bb70639c3fb46702de9c8356f8e2f956
--- /dev/null
+++ b/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d91c41f87c8dbce15b820b41232d0dcd26ba285c22362400d3dd771a711417d
+size 17110107
diff --git a/tokenizer_config.json b/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..733cf8031772d40c50da15c3fe56fe63f05c2a13
--- /dev/null
+++ b/tokenizer_config.json
@@ -0,0 +1,528 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<pad>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250001": {
+      "content": "ar_AR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250002": {
+      "content": "cs_CZ",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250003": {
+      "content": "de_DE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250004": {
+      "content": "en_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250005": {
+      "content": "es_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250006": {
+      "content": "et_EE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250007": {
+      "content": "fi_FI",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250008": {
+      "content": "fr_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250009": {
+      "content": "gu_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250010": {
+      "content": "hi_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250011": {
+      "content": "it_IT",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250012": {
+      "content": "ja_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250013": {
+      "content": "kk_KZ",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250014": {
+      "content": "ko_KR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250015": {
+      "content": "lt_LT",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250016": {
+      "content": "lv_LV",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250017": {
+      "content": "my_MM",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250018": {
+      "content": "ne_NP",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250019": {
+      "content": "nl_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250020": {
+      "content": "ro_RO",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250021": {
+      "content": "ru_RU",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250022": {
+      "content": "si_LK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250023": {
+      "content": "tr_TR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250024": {
+      "content": "vi_VN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250025": {
+      "content": "zh_CN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250026": {
+      "content": "af_ZA",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250027": {
+      "content": "az_AZ",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250028": {
+      "content": "bn_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250029": {
+      "content": "fa_IR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250030": {
+      "content": "he_IL",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250031": {
+      "content": "hr_HR",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250032": {
+      "content": "id_ID",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250033": {
+      "content": "ka_GE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250034": {
+      "content": "km_KH",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250035": {
+      "content": "mk_MK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250036": {
+      "content": "ml_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250037": {
+      "content": "mn_MN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250038": {
+      "content": "mr_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250039": {
+      "content": "pl_PL",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250040": {
+      "content": "ps_AF",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250041": {
+      "content": "pt_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250042": {
+      "content": "sv_SE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250043": {
+      "content": "sw_KE",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250044": {
+      "content": "ta_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250045": {
+      "content": "te_IN",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250046": {
+      "content": "th_TH",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250047": {
+      "content": "tl_XX",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250048": {
+      "content": "uk_UA",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250049": {
+      "content": "ur_PK",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250050": {
+      "content": "xh_ZA",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250051": {
+      "content": "gl_ES",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250052": {
+      "content": "sl_SI",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "250053": {
+      "content": "<mask>",
+      "lstrip": true,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "ar_AR",
+    "cs_CZ",
+    "de_DE",
+    "en_XX",
+    "es_XX",
+    "et_EE",
+    "fi_FI",
+    "fr_XX",
+    "gu_IN",
+    "hi_IN",
+    "it_IT",
+    "ja_XX",
+    "kk_KZ",
+    "ko_KR",
+    "lt_LT",
+    "lv_LV",
+    "my_MM",
+    "ne_NP",
+    "nl_XX",
+    "ro_RO",
+    "ru_RU",
+    "si_LK",
+    "tr_TR",
+    "vi_VN",
+    "zh_CN",
+    "af_ZA",
+    "az_AZ",
+    "bn_IN",
+    "fa_IR",
+    "he_IL",
+    "hr_HR",
+    "id_ID",
+    "ka_GE",
+    "km_KH",
+    "mk_MK",
+    "ml_IN",
+    "mn_MN",
+    "mr_IN",
+    "pl_PL",
+    "ps_AF",
+    "pt_XX",
+    "sv_SE",
+    "sw_KE",
+    "ta_IN",
+    "te_IN",
+    "th_TH",
+    "tl_XX",
+    "uk_UA",
+    "ur_PK",
+    "xh_ZA",
+    "gl_ES",
+    "sl_SI"
+  ],
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "<s>",
+  "eos_token": "</s>",
+  "language_codes": "ML50",
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "sep_token": "</s>",
+  "sp_model_kwargs": {},
+  "src_lang": "zh_CN",
+  "tgt_lang": "ko_KR",
+  "tokenizer_class": "MBart50Tokenizer",
+  "unk_token": "<unk>"
+}
diff --git a/train_results.json b/train_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..277bbf411f95bba9478db424cb65aac48c75e824
--- /dev/null
+++ b/train_results.json
@@ -0,0 +1,8 @@
+{
+    "epoch": 5.0,
+    "train_loss": 0.7189551201696451,
+    "train_runtime": 324742.1955,
+    "train_samples": 1134707,
+    "train_samples_per_second": 52.413,
+    "train_steps_per_second": 3.276
+}
\ No newline at end of file
diff --git a/trainer_state.json b/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..246bebb1a2a12e9518ac68a6b3d3ce78e42c052f
--- /dev/null
+++ b/trainer_state.json
@@ -0,0 +1,4332 @@
+{
+  "best_metric": 0.8987888693809509,
+  "best_model_checkpoint": "./zhko_mbartLarge_100p_run1/checkpoint-141839",
+  "epoch": 4.9999929497528885,
+  "eval_steps": 500,
+  "global_step": 354597,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.335965407031397e-06,
+      "loss": 2.879,
+      "step": 500
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 4.6860312088738486e-06,
+      "loss": 2.0241,
+      "step": 1000
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 7.0360970107162996e-06,
+      "loss": 1.8819,
+      "step": 1500
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.386162812558751e-06,
+      "loss": 1.8009,
+      "step": 2000
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.1736228614401204e-05,
+      "loss": 1.7072,
+      "step": 2500
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 1.4086294416243657e-05,
+      "loss": 1.682,
+      "step": 3000
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 1.643636021808611e-05,
+      "loss": 1.6188,
+      "step": 3500
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 1.878642601992856e-05,
+      "loss": 1.5802,
+      "step": 4000
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 2.1136491821771008e-05,
+      "loss": 1.5547,
+      "step": 4500
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 2.348655762361346e-05,
+      "loss": 1.5422,
+      "step": 5000
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2.5831923293852228e-05,
+      "loss": 1.5194,
+      "step": 5500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 2.8181989095694684e-05,
+      "loss": 1.4815,
+      "step": 6000
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 3.053205489753713e-05,
+      "loss": 1.4509,
+      "step": 6500
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 3.288212069937958e-05,
+      "loss": 1.458,
+      "step": 7000
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 3.5227486369618354e-05,
+      "loss": 1.4415,
+      "step": 7500
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 3.757285203985712e-05,
+      "loss": 1.4284,
+      "step": 8000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 3.991821771009588e-05,
+      "loss": 1.4297,
+      "step": 8500
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.2268283511938334e-05,
+      "loss": 1.4307,
+      "step": 9000
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 4.461834931378079e-05,
+      "loss": 1.3971,
+      "step": 9500
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 4.696371498401955e-05,
+      "loss": 1.4106,
+      "step": 10000
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 4.9313780785862004e-05,
+      "loss": 1.3999,
+      "step": 10500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.998319322943521e-05,
+      "loss": 1.3937,
+      "step": 11000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.995945485293127e-05,
+      "loss": 1.3773,
+      "step": 11500
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 4.9935716476427316e-05,
+      "loss": 1.3621,
+      "step": 12000
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.991197809992337e-05,
+      "loss": 1.3552,
+      "step": 12500
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 4.988823972341943e-05,
+      "loss": 1.3312,
+      "step": 13000
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 4.986450134691549e-05,
+      "loss": 1.3331,
+      "step": 13500
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.984081044716455e-05,
+      "loss": 1.3259,
+      "step": 14000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 4.9817072070660604e-05,
+      "loss": 1.2983,
+      "step": 14500
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 4.979333369415666e-05,
+      "loss": 1.2828,
+      "step": 15000
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 4.976959531765271e-05,
+      "loss": 1.2793,
+      "step": 15500
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.9745856941148764e-05,
+      "loss": 1.2968,
+      "step": 16000
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 4.9722118564644826e-05,
+      "loss": 1.2774,
+      "step": 16500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4.969838018814088e-05,
+      "loss": 1.2507,
+      "step": 17000
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.967464181163694e-05,
+      "loss": 1.2521,
+      "step": 17500
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 4.965090343513299e-05,
+      "loss": 1.2337,
+      "step": 18000
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 4.962716505862904e-05,
+      "loss": 1.2238,
+      "step": 18500
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.96034266821251e-05,
+      "loss": 1.2111,
+      "step": 19000
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 4.957968830562115e-05,
+      "loss": 1.2336,
+      "step": 19500
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 4.955599740587022e-05,
+      "loss": 1.1975,
+      "step": 20000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 4.9532259029366274e-05,
+      "loss": 1.2092,
+      "step": 20500
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.950852065286233e-05,
+      "loss": 1.2079,
+      "step": 21000
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 4.948482975311139e-05,
+      "loss": 1.1831,
+      "step": 21500
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.946109137660745e-05,
+      "loss": 1.1965,
+      "step": 22000
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.94373530001035e-05,
+      "loss": 1.1928,
+      "step": 22500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 4.9413614623599556e-05,
+      "loss": 1.1779,
+      "step": 23000
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 4.938987624709561e-05,
+      "loss": 1.1759,
+      "step": 23500
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 4.936613787059167e-05,
+      "loss": 1.1871,
+      "step": 24000
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.934239949408772e-05,
+      "loss": 1.1782,
+      "step": 24500
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 4.931866111758378e-05,
+      "loss": 1.1619,
+      "step": 25000
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 4.9294922741079834e-05,
+      "loss": 1.1615,
+      "step": 25500
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.927118436457589e-05,
+      "loss": 1.1476,
+      "step": 26000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 4.924749346482495e-05,
+      "loss": 1.1507,
+      "step": 26500
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 4.922380256507401e-05,
+      "loss": 1.1489,
+      "step": 27000
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.9200064188570064e-05,
+      "loss": 1.144,
+      "step": 27500
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 4.917632581206613e-05,
+      "loss": 1.1223,
+      "step": 28000
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 4.915258743556218e-05,
+      "loss": 1.1319,
+      "step": 28500
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 4.912884905905824e-05,
+      "loss": 1.1235,
+      "step": 29000
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.9105110682554286e-05,
+      "loss": 1.1408,
+      "step": 29500
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 4.908141978280335e-05,
+      "loss": 1.1299,
+      "step": 30000
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 4.905772888305242e-05,
+      "loss": 1.1172,
+      "step": 30500
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.903399050654847e-05,
+      "loss": 1.1316,
+      "step": 31000
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 4.9010299606797535e-05,
+      "loss": 1.1154,
+      "step": 31500
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 4.8986561230293584e-05,
+      "loss": 1.1162,
+      "step": 32000
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 4.896282285378964e-05,
+      "loss": 1.101,
+      "step": 32500
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.89390844772857e-05,
+      "loss": 1.1153,
+      "step": 33000
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.891534610078176e-05,
+      "loss": 1.1023,
+      "step": 33500
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 4.889160772427781e-05,
+      "loss": 1.0935,
+      "step": 34000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.886786934777387e-05,
+      "loss": 1.0898,
+      "step": 34500
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 4.884413097126992e-05,
+      "loss": 1.0846,
+      "step": 35000
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 4.882039259476597e-05,
+      "loss": 1.0963,
+      "step": 35500
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.879665421826203e-05,
+      "loss": 1.0764,
+      "step": 36000
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 4.877291584175809e-05,
+      "loss": 1.0649,
+      "step": 36500
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 4.8749177465254145e-05,
+      "loss": 1.0935,
+      "step": 37000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.8725439088750194e-05,
+      "loss": 1.0739,
+      "step": 37500
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.870170071224625e-05,
+      "loss": 1.0779,
+      "step": 38000
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.8677962335742305e-05,
+      "loss": 1.083,
+      "step": 38500
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.865422395923836e-05,
+      "loss": 1.0729,
+      "step": 39000
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.8630485582734416e-05,
+      "loss": 1.0587,
+      "step": 39500
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.860679468298348e-05,
+      "loss": 1.078,
+      "step": 40000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.858305630647954e-05,
+      "loss": 1.0802,
+      "step": 40500
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.855931792997559e-05,
+      "loss": 1.0538,
+      "step": 41000
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.853557955347164e-05,
+      "loss": 1.0562,
+      "step": 41500
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.85118411769677e-05,
+      "loss": 1.0516,
+      "step": 42000
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 4.848810280046375e-05,
+      "loss": 1.0624,
+      "step": 42500
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.8464364423959815e-05,
+      "loss": 1.0587,
+      "step": 43000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.8440626047455864e-05,
+      "loss": 1.0544,
+      "step": 43500
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 4.841688767095192e-05,
+      "loss": 1.0467,
+      "step": 44000
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.8393291724707e-05,
+      "loss": 1.0613,
+      "step": 44500
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 4.836955334820306e-05,
+      "loss": 1.0365,
+      "step": 45000
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 4.834581497169911e-05,
+      "loss": 1.0424,
+      "step": 45500
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.832207659519516e-05,
+      "loss": 1.0442,
+      "step": 46000
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.829833821869122e-05,
+      "loss": 1.0354,
+      "step": 46500
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 4.8274789749199304e-05,
+      "loss": 1.0543,
+      "step": 47000
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 4.825105137269536e-05,
+      "loss": 1.0476,
+      "step": 47500
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.8227312996191415e-05,
+      "loss": 1.0427,
+      "step": 48000
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.820357461968748e-05,
+      "loss": 1.0174,
+      "step": 48500
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.8179836243183526e-05,
+      "loss": 1.0325,
+      "step": 49000
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 4.815609786667958e-05,
+      "loss": 1.0126,
+      "step": 49500
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.813240696692865e-05,
+      "loss": 1.0335,
+      "step": 50000
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.81086685904247e-05,
+      "loss": 1.0219,
+      "step": 50500
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 4.808493021392075e-05,
+      "loss": 1.0291,
+      "step": 51000
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.8061191837416815e-05,
+      "loss": 1.0251,
+      "step": 51500
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.803745346091287e-05,
+      "loss": 1.01,
+      "step": 52000
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 4.8013715084408925e-05,
+      "loss": 1.0224,
+      "step": 52500
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.7989976707904974e-05,
+      "loss": 1.0163,
+      "step": 53000
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 4.796623833140103e-05,
+      "loss": 1.0214,
+      "step": 53500
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 4.7942499954897085e-05,
+      "loss": 1.0187,
+      "step": 54000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 4.791876157839314e-05,
+      "loss": 1.0091,
+      "step": 54500
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.78950232018892e-05,
+      "loss": 1.0256,
+      "step": 55000
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 4.787128482538525e-05,
+      "loss": 1.0025,
+      "step": 55500
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 4.784754644888131e-05,
+      "loss": 1.0013,
+      "step": 56000
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.782380807237736e-05,
+      "loss": 1.0098,
+      "step": 56500
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 4.780006969587342e-05,
+      "loss": 1.017,
+      "step": 57000
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 4.777637879612248e-05,
+      "loss": 1.0114,
+      "step": 57500
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.775264041961854e-05,
+      "loss": 0.9977,
+      "step": 58000
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 4.7728902043114596e-05,
+      "loss": 1.0107,
+      "step": 58500
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 4.7705163666610644e-05,
+      "loss": 0.9869,
+      "step": 59000
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 4.76814252901067e-05,
+      "loss": 1.008,
+      "step": 59500
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.7657686913602755e-05,
+      "loss": 1.0025,
+      "step": 60000
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 4.763394853709881e-05,
+      "loss": 0.9964,
+      "step": 60500
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 4.7610210160594866e-05,
+      "loss": 0.9921,
+      "step": 61000
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.758647178409092e-05,
+      "loss": 0.9938,
+      "step": 61500
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 4.756273340758698e-05,
+      "loss": 0.9934,
+      "step": 62000
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 4.753899503108303e-05,
+      "loss": 1.0249,
+      "step": 62500
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.751530413133209e-05,
+      "loss": 0.9832,
+      "step": 63000
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.749156575482815e-05,
+      "loss": 1.001,
+      "step": 63500
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 4.74678273783242e-05,
+      "loss": 0.9755,
+      "step": 64000
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 4.7444089001820266e-05,
+      "loss": 0.995,
+      "step": 64500
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.742035062531632e-05,
+      "loss": 0.9824,
+      "step": 65000
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 4.739661224881237e-05,
+      "loss": 0.9759,
+      "step": 65500
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.7372873872308425e-05,
+      "loss": 0.98,
+      "step": 66000
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.734913549580448e-05,
+      "loss": 0.9905,
+      "step": 66500
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 4.7325397119300536e-05,
+      "loss": 0.9811,
+      "step": 67000
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 4.730165874279659e-05,
+      "loss": 0.9873,
+      "step": 67500
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 4.727792036629265e-05,
+      "loss": 0.9804,
+      "step": 68000
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.7254229466541714e-05,
+      "loss": 0.9695,
+      "step": 68500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 4.723049109003776e-05,
+      "loss": 0.9751,
+      "step": 69000
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.720675271353382e-05,
+      "loss": 0.9735,
+      "step": 69500
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.7183061813782885e-05,
+      "loss": 0.9832,
+      "step": 70000
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 4.7159370914031945e-05,
+      "loss": 0.9719,
+      "step": 70500
+    },
+    {
+      "epoch": 1.0,
+      "eval_bleu": 40.8492,
+      "eval_gen_len": 13.8028,
+      "eval_loss": 0.9435123801231384,
+      "eval_runtime": 10127.6159,
+      "eval_samples_per_second": 14.005,
+      "eval_steps_per_second": 1.751,
+      "step": 70919
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 4.7135632537528e-05,
+      "loss": 0.9687,
+      "step": 71000
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 4.7111894161024056e-05,
+      "loss": 0.9639,
+      "step": 71500
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.708815578452011e-05,
+      "loss": 0.9669,
+      "step": 72000
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 4.7064417408016166e-05,
+      "loss": 0.956,
+      "step": 72500
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 4.704067903151222e-05,
+      "loss": 0.9563,
+      "step": 73000
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 4.701694065500828e-05,
+      "loss": 0.9564,
+      "step": 73500
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 4.699320227850433e-05,
+      "loss": 0.9555,
+      "step": 74000
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 4.696946390200039e-05,
+      "loss": 0.9408,
+      "step": 74500
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.6945725525496444e-05,
+      "loss": 0.9347,
+      "step": 75000
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 4.69219871489925e-05,
+      "loss": 0.9368,
+      "step": 75500
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 4.6898296249241566e-05,
+      "loss": 0.9401,
+      "step": 76000
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 4.6874557872737615e-05,
+      "loss": 0.9383,
+      "step": 76500
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.685081949623367e-05,
+      "loss": 0.9232,
+      "step": 77000
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 4.6827081119729726e-05,
+      "loss": 0.9046,
+      "step": 77500
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 4.680334274322578e-05,
+      "loss": 0.9236,
+      "step": 78000
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.6779604366721837e-05,
+      "loss": 0.9091,
+      "step": 78500
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 4.675586599021789e-05,
+      "loss": 0.9036,
+      "step": 79000
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 4.673212761371395e-05,
+      "loss": 0.909,
+      "step": 79500
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 4.670838923721e-05,
+      "loss": 0.9003,
+      "step": 80000
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.668469833745906e-05,
+      "loss": 0.8957,
+      "step": 80500
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 4.666095996095512e-05,
+      "loss": 0.8928,
+      "step": 81000
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 4.66373640147102e-05,
+      "loss": 0.8842,
+      "step": 81500
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.6613625638206256e-05,
+      "loss": 0.885,
+      "step": 82000
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 4.658988726170231e-05,
+      "loss": 0.8821,
+      "step": 82500
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 4.656614888519836e-05,
+      "loss": 0.8769,
+      "step": 83000
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 4.6542410508694416e-05,
+      "loss": 0.8681,
+      "step": 83500
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 4.651867213219048e-05,
+      "loss": 0.8767,
+      "step": 84000
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 4.6494933755686534e-05,
+      "loss": 0.8702,
+      "step": 84500
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 4.647119537918259e-05,
+      "loss": 0.8753,
+      "step": 85000
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 4.644745700267864e-05,
+      "loss": 0.8587,
+      "step": 85500
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 4.642371862617469e-05,
+      "loss": 0.8459,
+      "step": 86000
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 4.639998024967075e-05,
+      "loss": 0.853,
+      "step": 86500
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 4.6376241873166804e-05,
+      "loss": 0.8811,
+      "step": 87000
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 4.635255097341587e-05,
+      "loss": 0.8618,
+      "step": 87500
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 4.6328812596911926e-05,
+      "loss": 0.8493,
+      "step": 88000
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 4.630507422040798e-05,
+      "loss": 0.8603,
+      "step": 88500
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 4.628133584390403e-05,
+      "loss": 0.8333,
+      "step": 89000
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 4.6257597467400086e-05,
+      "loss": 0.8386,
+      "step": 89500
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 4.623385909089614e-05,
+      "loss": 0.8424,
+      "step": 90000
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 4.6210120714392204e-05,
+      "loss": 0.8519,
+      "step": 90500
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 4.618638233788826e-05,
+      "loss": 0.8297,
+      "step": 91000
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 4.616264396138431e-05,
+      "loss": 0.8406,
+      "step": 91500
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 4.613890558488036e-05,
+      "loss": 0.8392,
+      "step": 92000
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 4.611516720837642e-05,
+      "loss": 0.8269,
+      "step": 92500
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 4.6091476308625486e-05,
+      "loss": 0.8392,
+      "step": 93000
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 4.6067785408874545e-05,
+      "loss": 0.8474,
+      "step": 93500
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 4.60440470323706e-05,
+      "loss": 0.8329,
+      "step": 94000
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 4.6020308655866656e-05,
+      "loss": 0.8327,
+      "step": 94500
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 4.599657027936271e-05,
+      "loss": 0.8435,
+      "step": 95000
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 4.597283190285877e-05,
+      "loss": 0.8343,
+      "step": 95500
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 4.5949141003107834e-05,
+      "loss": 0.8316,
+      "step": 96000
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 4.592540262660389e-05,
+      "loss": 0.8339,
+      "step": 96500
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 4.590166425009994e-05,
+      "loss": 0.8199,
+      "step": 97000
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 4.5877925873595994e-05,
+      "loss": 0.8162,
+      "step": 97500
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 4.585418749709205e-05,
+      "loss": 0.8239,
+      "step": 98000
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 4.5830496597341116e-05,
+      "loss": 0.8197,
+      "step": 98500
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 4.5806805697590176e-05,
+      "loss": 0.8164,
+      "step": 99000
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 4.578306732108623e-05,
+      "loss": 0.818,
+      "step": 99500
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 4.575932894458229e-05,
+      "loss": 0.8189,
+      "step": 100000
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 4.573559056807834e-05,
+      "loss": 0.8303,
+      "step": 100500
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 4.57118521915744e-05,
+      "loss": 0.8174,
+      "step": 101000
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 4.568811381507045e-05,
+      "loss": 0.814,
+      "step": 101500
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 4.566437543856651e-05,
+      "loss": 0.8226,
+      "step": 102000
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 4.5640637062062564e-05,
+      "loss": 0.818,
+      "step": 102500
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 4.561689868555862e-05,
+      "loss": 0.8054,
+      "step": 103000
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 4.5593160309054675e-05,
+      "loss": 0.8191,
+      "step": 103500
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 4.556946940930374e-05,
+      "loss": 0.8134,
+      "step": 104000
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 4.554573103279979e-05,
+      "loss": 0.8077,
+      "step": 104500
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 4.5521992656295846e-05,
+      "loss": 0.8081,
+      "step": 105000
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 4.54982542797919e-05,
+      "loss": 0.7985,
+      "step": 105500
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 4.547451590328796e-05,
+      "loss": 0.8064,
+      "step": 106000
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 4.545077752678401e-05,
+      "loss": 0.8008,
+      "step": 106500
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 4.542703915028007e-05,
+      "loss": 0.7934,
+      "step": 107000
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 4.540330077377612e-05,
+      "loss": 0.7915,
+      "step": 107500
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 4.537956239727218e-05,
+      "loss": 0.8087,
+      "step": 108000
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 4.5355824020768234e-05,
+      "loss": 0.7941,
+      "step": 108500
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 4.533208564426429e-05,
+      "loss": 0.8032,
+      "step": 109000
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 4.5308347267760345e-05,
+      "loss": 0.8061,
+      "step": 109500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 4.5284608891256394e-05,
+      "loss": 0.7974,
+      "step": 110000
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 4.526091799150546e-05,
+      "loss": 0.7915,
+      "step": 110500
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 4.523722709175453e-05,
+      "loss": 0.7998,
+      "step": 111000
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 4.5213488715250576e-05,
+      "loss": 0.8014,
+      "step": 111500
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 4.518975033874663e-05,
+      "loss": 0.7848,
+      "step": 112000
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.516601196224269e-05,
+      "loss": 0.7914,
+      "step": 112500
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 4.5142321062491754e-05,
+      "loss": 0.795,
+      "step": 113000
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 4.5118630162740814e-05,
+      "loss": 0.7913,
+      "step": 113500
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.509489178623687e-05,
+      "loss": 0.7936,
+      "step": 114000
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 4.5071153409732924e-05,
+      "loss": 0.7932,
+      "step": 114500
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 4.504741503322898e-05,
+      "loss": 0.7835,
+      "step": 115000
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 4.502367665672504e-05,
+      "loss": 0.7942,
+      "step": 115500
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 4.499993828022109e-05,
+      "loss": 0.7771,
+      "step": 116000
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 4.4976199903717146e-05,
+      "loss": 0.7805,
+      "step": 116500
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 4.4952509003966206e-05,
+      "loss": 0.7952,
+      "step": 117000
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 4.492877062746226e-05,
+      "loss": 0.7799,
+      "step": 117500
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 4.490503225095832e-05,
+      "loss": 0.7932,
+      "step": 118000
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 4.488129387445438e-05,
+      "loss": 0.7919,
+      "step": 118500
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 4.4857555497950435e-05,
+      "loss": 0.7776,
+      "step": 119000
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 4.4833864598199495e-05,
+      "loss": 0.7731,
+      "step": 119500
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 4.481012622169555e-05,
+      "loss": 0.7778,
+      "step": 120000
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 4.47863878451916e-05,
+      "loss": 0.7638,
+      "step": 120500
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 4.476264946868766e-05,
+      "loss": 0.7819,
+      "step": 121000
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 4.473891109218372e-05,
+      "loss": 0.7717,
+      "step": 121500
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 4.471517271567977e-05,
+      "loss": 0.7803,
+      "step": 122000
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 4.469143433917583e-05,
+      "loss": 0.7807,
+      "step": 122500
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 4.4667695962671876e-05,
+      "loss": 0.7651,
+      "step": 123000
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 4.464395758616793e-05,
+      "loss": 0.7811,
+      "step": 123500
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 4.4620266686417e-05,
+      "loss": 0.7794,
+      "step": 124000
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 4.4596528309913054e-05,
+      "loss": 0.7763,
+      "step": 124500
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 4.457278993340911e-05,
+      "loss": 0.7757,
+      "step": 125000
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 4.4549051556905165e-05,
+      "loss": 0.7683,
+      "step": 125500
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 4.4525360657154225e-05,
+      "loss": 0.7764,
+      "step": 126000
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 4.450162228065028e-05,
+      "loss": 0.7661,
+      "step": 126500
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 4.4477883904146336e-05,
+      "loss": 0.7645,
+      "step": 127000
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 4.445414552764239e-05,
+      "loss": 0.7703,
+      "step": 127500
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 4.443040715113845e-05,
+      "loss": 0.7791,
+      "step": 128000
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 4.44066687746345e-05,
+      "loss": 0.7657,
+      "step": 128500
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 4.438293039813056e-05,
+      "loss": 0.7651,
+      "step": 129000
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 4.435919202162661e-05,
+      "loss": 0.7803,
+      "step": 129500
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 4.433545364512267e-05,
+      "loss": 0.7504,
+      "step": 130000
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 4.4311762745371735e-05,
+      "loss": 0.7785,
+      "step": 130500
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 4.4288024368867784e-05,
+      "loss": 0.7592,
+      "step": 131000
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 4.426428599236384e-05,
+      "loss": 0.7714,
+      "step": 131500
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 4.4240547615859895e-05,
+      "loss": 0.7652,
+      "step": 132000
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 4.421680923935595e-05,
+      "loss": 0.7608,
+      "step": 132500
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.4193070862852006e-05,
+      "loss": 0.7688,
+      "step": 133000
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 4.416933248634806e-05,
+      "loss": 0.79,
+      "step": 133500
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 4.414559410984412e-05,
+      "loss": 0.7526,
+      "step": 134000
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 4.412190321009318e-05,
+      "loss": 0.765,
+      "step": 134500
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 4.409816483358923e-05,
+      "loss": 0.7554,
+      "step": 135000
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 4.407442645708529e-05,
+      "loss": 0.7636,
+      "step": 135500
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 4.4050735557334354e-05,
+      "loss": 0.7582,
+      "step": 136000
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 4.402699718083041e-05,
+      "loss": 0.7504,
+      "step": 136500
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 4.400330628107947e-05,
+      "loss": 0.7591,
+      "step": 137000
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 4.3979567904575525e-05,
+      "loss": 0.7629,
+      "step": 137500
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 4.395582952807158e-05,
+      "loss": 0.7543,
+      "step": 138000
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 4.3932091151567636e-05,
+      "loss": 0.7634,
+      "step": 138500
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 4.390835277506369e-05,
+      "loss": 0.7553,
+      "step": 139000
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 4.388461439855975e-05,
+      "loss": 0.7548,
+      "step": 139500
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 4.38608760220558e-05,
+      "loss": 0.7554,
+      "step": 140000
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 4.383713764555186e-05,
+      "loss": 0.7539,
+      "step": 140500
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 4.3813399269047914e-05,
+      "loss": 0.7562,
+      "step": 141000
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.378966089254396e-05,
+      "loss": 0.7537,
+      "step": 141500
+    },
+    {
+      "epoch": 2.0,
+      "eval_bleu": 42.7907,
+      "eval_gen_len": 13.7941,
+      "eval_loss": 0.8987888693809509,
+      "eval_runtime": 10044.1408,
+      "eval_samples_per_second": 14.121,
+      "eval_steps_per_second": 1.765,
+      "step": 141839
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 4.376596999279303e-05,
+      "loss": 0.7511,
+      "step": 142000
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 4.3742231616289084e-05,
+      "loss": 0.75,
+      "step": 142500
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.371849323978514e-05,
+      "loss": 0.7489,
+      "step": 143000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 4.3694754863281195e-05,
+      "loss": 0.7359,
+      "step": 143500
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 4.367101648677725e-05,
+      "loss": 0.7416,
+      "step": 144000
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.364732558702632e-05,
+      "loss": 0.7421,
+      "step": 144500
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 4.362358721052237e-05,
+      "loss": 0.7448,
+      "step": 145000
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 4.359984883401843e-05,
+      "loss": 0.7314,
+      "step": 145500
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 4.357611045751448e-05,
+      "loss": 0.731,
+      "step": 146000
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 4.3552419557763544e-05,
+      "loss": 0.7307,
+      "step": 146500
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 4.35286811812596e-05,
+      "loss": 0.7411,
+      "step": 147000
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 4.3504942804755655e-05,
+      "loss": 0.7298,
+      "step": 147500
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 4.348120442825171e-05,
+      "loss": 0.7144,
+      "step": 148000
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 4.3457466051747766e-05,
+      "loss": 0.7146,
+      "step": 148500
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 4.343372767524382e-05,
+      "loss": 0.7157,
+      "step": 149000
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 4.341003677549288e-05,
+      "loss": 0.7116,
+      "step": 149500
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.338629839898894e-05,
+      "loss": 0.7105,
+      "step": 150000
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 4.336256002248499e-05,
+      "loss": 0.7096,
+      "step": 150500
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 4.3338916599487063e-05,
+      "loss": 0.7026,
+      "step": 151000
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 4.331517822298312e-05,
+      "loss": 0.706,
+      "step": 151500
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 4.3291439846479174e-05,
+      "loss": 0.6996,
+      "step": 152000
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 4.326770146997523e-05,
+      "loss": 0.6965,
+      "step": 152500
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 4.3243963093471285e-05,
+      "loss": 0.699,
+      "step": 153000
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 4.322022471696734e-05,
+      "loss": 0.6926,
+      "step": 153500
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 4.3196486340463396e-05,
+      "loss": 0.6925,
+      "step": 154000
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 4.3172747963959445e-05,
+      "loss": 0.6825,
+      "step": 154500
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 4.31490095874555e-05,
+      "loss": 0.6869,
+      "step": 155000
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 4.312527121095156e-05,
+      "loss": 0.691,
+      "step": 155500
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 4.310153283444762e-05,
+      "loss": 0.682,
+      "step": 156000
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 4.3077794457943673e-05,
+      "loss": 0.6822,
+      "step": 156500
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 4.305405608143972e-05,
+      "loss": 0.6666,
+      "step": 157000
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 4.303031770493578e-05,
+      "loss": 0.6749,
+      "step": 157500
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 4.300657932843183e-05,
+      "loss": 0.6903,
+      "step": 158000
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 4.29828884286809e-05,
+      "loss": 0.6816,
+      "step": 158500
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 4.2959150052176955e-05,
+      "loss": 0.6776,
+      "step": 159000
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 4.2935459152426015e-05,
+      "loss": 0.6787,
+      "step": 159500
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 4.2911768252675075e-05,
+      "loss": 0.6559,
+      "step": 160000
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 4.288807735292414e-05,
+      "loss": 0.668,
+      "step": 160500
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 4.28643389764202e-05,
+      "loss": 0.6755,
+      "step": 161000
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 4.284060059991625e-05,
+      "loss": 0.6751,
+      "step": 161500
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 4.281686222341231e-05,
+      "loss": 0.6608,
+      "step": 162000
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 4.2793123846908364e-05,
+      "loss": 0.6562,
+      "step": 162500
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 4.276938547040442e-05,
+      "loss": 0.6672,
+      "step": 163000
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 4.2745647093900475e-05,
+      "loss": 0.6577,
+      "step": 163500
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 4.272190871739653e-05,
+      "loss": 0.6714,
+      "step": 164000
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 4.2698170340892586e-05,
+      "loss": 0.6706,
+      "step": 164500
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 4.2674479441141646e-05,
+      "loss": 0.6593,
+      "step": 165000
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 4.26507410646377e-05,
+      "loss": 0.6633,
+      "step": 165500
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 4.262705016488677e-05,
+      "loss": 0.6672,
+      "step": 166000
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 4.260331178838282e-05,
+      "loss": 0.6584,
+      "step": 166500
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 4.257957341187888e-05,
+      "loss": 0.6593,
+      "step": 167000
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 4.255583503537493e-05,
+      "loss": 0.6651,
+      "step": 167500
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 4.253209665887098e-05,
+      "loss": 0.6519,
+      "step": 168000
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 4.250835828236704e-05,
+      "loss": 0.6518,
+      "step": 168500
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 4.24846199058631e-05,
+      "loss": 0.6598,
+      "step": 169000
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 4.2460881529359156e-05,
+      "loss": 0.6445,
+      "step": 169500
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 4.2437143152855205e-05,
+      "loss": 0.6499,
+      "step": 170000
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 4.241340477635126e-05,
+      "loss": 0.6516,
+      "step": 170500
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 4.2389666399847316e-05,
+      "loss": 0.6553,
+      "step": 171000
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 4.236592802334337e-05,
+      "loss": 0.668,
+      "step": 171500
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 4.2342189646839427e-05,
+      "loss": 0.6447,
+      "step": 172000
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 4.231845127033548e-05,
+      "loss": 0.6474,
+      "step": 172500
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 4.229471289383154e-05,
+      "loss": 0.6594,
+      "step": 173000
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 4.227097451732759e-05,
+      "loss": 0.6516,
+      "step": 173500
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 4.224723614082365e-05,
+      "loss": 0.6404,
+      "step": 174000
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 4.222354524107271e-05,
+      "loss": 0.6526,
+      "step": 174500
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 4.2199806864568764e-05,
+      "loss": 0.6459,
+      "step": 175000
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 4.2176068488064826e-05,
+      "loss": 0.6519,
+      "step": 175500
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 4.2152377588313886e-05,
+      "loss": 0.6391,
+      "step": 176000
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 4.212863921180994e-05,
+      "loss": 0.6346,
+      "step": 176500
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 4.2104900835306e-05,
+      "loss": 0.6395,
+      "step": 177000
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 4.2081162458802046e-05,
+      "loss": 0.6409,
+      "step": 177500
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 4.20574240822981e-05,
+      "loss": 0.6285,
+      "step": 178000
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 4.2033685705794163e-05,
+      "loss": 0.6307,
+      "step": 178500
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 4.200999480604322e-05,
+      "loss": 0.6506,
+      "step": 179000
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 4.198625642953928e-05,
+      "loss": 0.6307,
+      "step": 179500
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 4.1962518053035334e-05,
+      "loss": 0.6446,
+      "step": 180000
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 4.193877967653139e-05,
+      "loss": 0.6422,
+      "step": 180500
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 4.1915088776780457e-05,
+      "loss": 0.6398,
+      "step": 181000
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 4.1891397877029516e-05,
+      "loss": 0.6333,
+      "step": 181500
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 4.186765950052557e-05,
+      "loss": 0.6363,
+      "step": 182000
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 4.184392112402162e-05,
+      "loss": 0.6319,
+      "step": 182500
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 4.1820182747517676e-05,
+      "loss": 0.6294,
+      "step": 183000
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 4.179649184776674e-05,
+      "loss": 0.6366,
+      "step": 183500
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 4.17727534712628e-05,
+      "loss": 0.6384,
+      "step": 184000
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 4.1749015094758854e-05,
+      "loss": 0.6365,
+      "step": 184500
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 4.172527671825491e-05,
+      "loss": 0.6331,
+      "step": 185000
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 4.1701538341750965e-05,
+      "loss": 0.6327,
+      "step": 185500
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 4.167779996524701e-05,
+      "loss": 0.628,
+      "step": 186000
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 4.1654061588743076e-05,
+      "loss": 0.6309,
+      "step": 186500
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 4.163032321223913e-05,
+      "loss": 0.6159,
+      "step": 187000
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 4.1606584835735187e-05,
+      "loss": 0.6307,
+      "step": 187500
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 4.158284645923124e-05,
+      "loss": 0.6356,
+      "step": 188000
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 4.155910808272729e-05,
+      "loss": 0.6269,
+      "step": 188500
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 4.1535369706223346e-05,
+      "loss": 0.6368,
+      "step": 189000
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 4.15116313297194e-05,
+      "loss": 0.6282,
+      "step": 189500
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 4.1487892953215464e-05,
+      "loss": 0.6226,
+      "step": 190000
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 4.146415457671152e-05,
+      "loss": 0.6181,
+      "step": 190500
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 4.144041620020757e-05,
+      "loss": 0.6138,
+      "step": 191000
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 4.1416677823703623e-05,
+      "loss": 0.6184,
+      "step": 191500
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 4.139298692395269e-05,
+      "loss": 0.624,
+      "step": 192000
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 4.136924854744874e-05,
+      "loss": 0.6125,
+      "step": 192500
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 4.13455101709448e-05,
+      "loss": 0.6261,
+      "step": 193000
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 4.1321771794440857e-05,
+      "loss": 0.6189,
+      "step": 193500
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 4.129803341793691e-05,
+      "loss": 0.6149,
+      "step": 194000
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 4.127429504143297e-05,
+      "loss": 0.6296,
+      "step": 194500
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 4.1250556664929016e-05,
+      "loss": 0.6263,
+      "step": 195000
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 4.122681828842507e-05,
+      "loss": 0.6298,
+      "step": 195500
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 4.120312738867414e-05,
+      "loss": 0.6102,
+      "step": 196000
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 4.1179389012170194e-05,
+      "loss": 0.6248,
+      "step": 196500
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 4.115565063566625e-05,
+      "loss": 0.6121,
+      "step": 197000
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 4.1131912259162305e-05,
+      "loss": 0.6156,
+      "step": 197500
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 4.1108221359411365e-05,
+      "loss": 0.6076,
+      "step": 198000
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 4.1084577936413436e-05,
+      "loss": 0.6231,
+      "step": 198500
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 4.106083955990949e-05,
+      "loss": 0.6281,
+      "step": 199000
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 4.103710118340555e-05,
+      "loss": 0.6157,
+      "step": 199500
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 4.10133628069016e-05,
+      "loss": 0.6086,
+      "step": 200000
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 4.098962443039766e-05,
+      "loss": 0.6225,
+      "step": 200500
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 4.096588605389371e-05,
+      "loss": 0.6078,
+      "step": 201000
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 4.094214767738977e-05,
+      "loss": 0.6217,
+      "step": 201500
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 4.0918409300885824e-05,
+      "loss": 0.6096,
+      "step": 202000
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 4.089467092438188e-05,
+      "loss": 0.6253,
+      "step": 202500
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 4.087102750138395e-05,
+      "loss": 0.6068,
+      "step": 203000
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 4.0847289124880006e-05,
+      "loss": 0.613,
+      "step": 203500
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 4.082355074837606e-05,
+      "loss": 0.6179,
+      "step": 204000
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 4.079981237187212e-05,
+      "loss": 0.6312,
+      "step": 204500
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 4.077612147212118e-05,
+      "loss": 0.601,
+      "step": 205000
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 4.075238309561723e-05,
+      "loss": 0.615,
+      "step": 205500
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 4.072864471911329e-05,
+      "loss": 0.6071,
+      "step": 206000
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 4.0704906342609344e-05,
+      "loss": 0.6135,
+      "step": 206500
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 4.06811679661054e-05,
+      "loss": 0.6073,
+      "step": 207000
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 4.065747706635446e-05,
+      "loss": 0.6073,
+      "step": 207500
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 4.0633738689850515e-05,
+      "loss": 0.6053,
+      "step": 208000
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 4.061000031334658e-05,
+      "loss": 0.6116,
+      "step": 208500
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 4.058626193684263e-05,
+      "loss": 0.606,
+      "step": 209000
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 4.056252356033868e-05,
+      "loss": 0.6183,
+      "step": 209500
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 4.0538785183834736e-05,
+      "loss": 0.601,
+      "step": 210000
+    },
+    {
+      "epoch": 2.97,
+      "learning_rate": 4.051504680733079e-05,
+      "loss": 0.6041,
+      "step": 210500
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 4.049130843082685e-05,
+      "loss": 0.6073,
+      "step": 211000
+    },
+    {
+      "epoch": 2.98,
+      "learning_rate": 4.04675700543229e-05,
+      "loss": 0.6106,
+      "step": 211500
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 4.044383167781896e-05,
+      "loss": 0.6105,
+      "step": 212000
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.0420093301315014e-05,
+      "loss": 0.5973,
+      "step": 212500
+    },
+    {
+      "epoch": 3.0,
+      "eval_bleu": 43.4697,
+      "eval_gen_len": 13.6556,
+      "eval_loss": 0.9142627120018005,
+      "eval_runtime": 9819.0968,
+      "eval_samples_per_second": 14.445,
+      "eval_steps_per_second": 1.806,
+      "step": 212758
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 4.039635492481107e-05,
+      "loss": 0.6014,
+      "step": 213000
+    },
+    {
+      "epoch": 3.01,
+      "learning_rate": 4.037266402506013e-05,
+      "loss": 0.6099,
+      "step": 213500
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 4.0348925648556185e-05,
+      "loss": 0.5951,
+      "step": 214000
+    },
+    {
+      "epoch": 3.02,
+      "learning_rate": 4.032518727205224e-05,
+      "loss": 0.597,
+      "step": 214500
+    },
+    {
+      "epoch": 3.03,
+      "learning_rate": 4.03014488955483e-05,
+      "loss": 0.596,
+      "step": 215000
+    },
+    {
+      "epoch": 3.04,
+      "learning_rate": 4.027771051904435e-05,
+      "loss": 0.5914,
+      "step": 215500
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 4.0253972142540406e-05,
+      "loss": 0.5979,
+      "step": 216000
+    },
+    {
+      "epoch": 3.05,
+      "learning_rate": 4.023023376603646e-05,
+      "loss": 0.5855,
+      "step": 216500
+    },
+    {
+      "epoch": 3.06,
+      "learning_rate": 4.020649538953252e-05,
+      "loss": 0.5887,
+      "step": 217000
+    },
+    {
+      "epoch": 3.07,
+      "learning_rate": 4.018275701302857e-05,
+      "loss": 0.5853,
+      "step": 217500
+    },
+    {
+      "epoch": 3.07,
+      "learning_rate": 4.015901863652463e-05,
+      "loss": 0.5921,
+      "step": 218000
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 4.013528026002068e-05,
+      "loss": 0.5873,
+      "step": 218500
+    },
+    {
+      "epoch": 3.09,
+      "learning_rate": 4.0111589360269744e-05,
+      "loss": 0.5708,
+      "step": 219000
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 4.00878509837658e-05,
+      "loss": 0.5777,
+      "step": 219500
+    },
+    {
+      "epoch": 3.1,
+      "learning_rate": 4.0064112607261855e-05,
+      "loss": 0.5772,
+      "step": 220000
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 4.004037423075791e-05,
+      "loss": 0.573,
+      "step": 220500
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 4.0016635854253966e-05,
+      "loss": 0.5734,
+      "step": 221000
+    },
+    {
+      "epoch": 3.12,
+      "learning_rate": 3.999289747775002e-05,
+      "loss": 0.5682,
+      "step": 221500
+    },
+    {
+      "epoch": 3.13,
+      "learning_rate": 3.996920657799909e-05,
+      "loss": 0.5628,
+      "step": 222000
+    },
+    {
+      "epoch": 3.14,
+      "learning_rate": 3.9945468201495136e-05,
+      "loss": 0.5687,
+      "step": 222500
+    },
+    {
+      "epoch": 3.14,
+      "learning_rate": 3.992172982499119e-05,
+      "loss": 0.5569,
+      "step": 223000
+    },
+    {
+      "epoch": 3.15,
+      "learning_rate": 3.989799144848725e-05,
+      "loss": 0.5571,
+      "step": 223500
+    },
+    {
+      "epoch": 3.16,
+      "learning_rate": 3.98742530719833e-05,
+      "loss": 0.5619,
+      "step": 224000
+    },
+    {
+      "epoch": 3.17,
+      "learning_rate": 3.9850514695479365e-05,
+      "loss": 0.5635,
+      "step": 224500
+    },
+    {
+      "epoch": 3.17,
+      "learning_rate": 3.982677631897542e-05,
+      "loss": 0.5538,
+      "step": 225000
+    },
+    {
+      "epoch": 3.18,
+      "learning_rate": 3.980303794247147e-05,
+      "loss": 0.5494,
+      "step": 225500
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 3.9779299565967525e-05,
+      "loss": 0.5532,
+      "step": 226000
+    },
+    {
+      "epoch": 3.19,
+      "learning_rate": 3.975556118946358e-05,
+      "loss": 0.5576,
+      "step": 226500
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 3.973187028971264e-05,
+      "loss": 0.5499,
+      "step": 227000
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 3.970822686671472e-05,
+      "loss": 0.5434,
+      "step": 227500
+    },
+    {
+      "epoch": 3.21,
+      "learning_rate": 3.968448849021077e-05,
+      "loss": 0.5421,
+      "step": 228000
+    },
+    {
+      "epoch": 3.22,
+      "learning_rate": 3.966075011370682e-05,
+      "loss": 0.5377,
+      "step": 228500
+    },
+    {
+      "epoch": 3.23,
+      "learning_rate": 3.963701173720288e-05,
+      "loss": 0.5614,
+      "step": 229000
+    },
+    {
+      "epoch": 3.24,
+      "learning_rate": 3.961327336069894e-05,
+      "loss": 0.5458,
+      "step": 229500
+    },
+    {
+      "epoch": 3.24,
+      "learning_rate": 3.9589534984194995e-05,
+      "loss": 0.5483,
+      "step": 230000
+    },
+    {
+      "epoch": 3.25,
+      "learning_rate": 3.9565796607691044e-05,
+      "loss": 0.5423,
+      "step": 230500
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 3.95420582311871e-05,
+      "loss": 0.5346,
+      "step": 231000
+    },
+    {
+      "epoch": 3.26,
+      "learning_rate": 3.9518319854683155e-05,
+      "loss": 0.5334,
+      "step": 231500
+    },
+    {
+      "epoch": 3.27,
+      "learning_rate": 3.9494628954932215e-05,
+      "loss": 0.5451,
+      "step": 232000
+    },
+    {
+      "epoch": 3.28,
+      "learning_rate": 3.947089057842828e-05,
+      "loss": 0.5378,
+      "step": 232500
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 3.944715220192433e-05,
+      "loss": 0.5323,
+      "step": 233000
+    },
+    {
+      "epoch": 3.29,
+      "learning_rate": 3.942346130217339e-05,
+      "loss": 0.5346,
+      "step": 233500
+    },
+    {
+      "epoch": 3.3,
+      "learning_rate": 3.939972292566945e-05,
+      "loss": 0.5334,
+      "step": 234000
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 3.9375984549165504e-05,
+      "loss": 0.5292,
+      "step": 234500
+    },
+    {
+      "epoch": 3.31,
+      "learning_rate": 3.935224617266155e-05,
+      "loss": 0.5442,
+      "step": 235000
+    },
+    {
+      "epoch": 3.32,
+      "learning_rate": 3.9328507796157615e-05,
+      "loss": 0.5423,
+      "step": 235500
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 3.930476941965367e-05,
+      "loss": 0.53,
+      "step": 236000
+    },
+    {
+      "epoch": 3.33,
+      "learning_rate": 3.9281031043149725e-05,
+      "loss": 0.5359,
+      "step": 236500
+    },
+    {
+      "epoch": 3.34,
+      "learning_rate": 3.9257340143398785e-05,
+      "loss": 0.5432,
+      "step": 237000
+    },
+    {
+      "epoch": 3.35,
+      "learning_rate": 3.923360176689484e-05,
+      "loss": 0.5325,
+      "step": 237500
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 3.9209863390390896e-05,
+      "loss": 0.53,
+      "step": 238000
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 3.918612501388695e-05,
+      "loss": 0.5347,
+      "step": 238500
+    },
+    {
+      "epoch": 3.37,
+      "learning_rate": 3.916238663738301e-05,
+      "loss": 0.5268,
+      "step": 239000
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 3.913864826087906e-05,
+      "loss": 0.5233,
+      "step": 239500
+    },
+    {
+      "epoch": 3.38,
+      "learning_rate": 3.911490988437512e-05,
+      "loss": 0.5281,
+      "step": 240000
+    },
+    {
+      "epoch": 3.39,
+      "learning_rate": 3.9091171507871174e-05,
+      "loss": 0.5191,
+      "step": 240500
+    },
+    {
+      "epoch": 3.4,
+      "learning_rate": 3.906743313136723e-05,
+      "loss": 0.5226,
+      "step": 241000
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 3.904369475486328e-05,
+      "loss": 0.525,
+      "step": 241500
+    },
+    {
+      "epoch": 3.41,
+      "learning_rate": 3.901995637835934e-05,
+      "loss": 0.5252,
+      "step": 242000
+    },
+    {
+      "epoch": 3.42,
+      "learning_rate": 3.8996218001855396e-05,
+      "loss": 0.5406,
+      "step": 242500
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 3.897257457885746e-05,
+      "loss": 0.5168,
+      "step": 243000
+    },
+    {
+      "epoch": 3.43,
+      "learning_rate": 3.8948836202353515e-05,
+      "loss": 0.5218,
+      "step": 243500
+    },
+    {
+      "epoch": 3.44,
+      "learning_rate": 3.892509782584958e-05,
+      "loss": 0.5304,
+      "step": 244000
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 3.890135944934563e-05,
+      "loss": 0.5217,
+      "step": 244500
+    },
+    {
+      "epoch": 3.45,
+      "learning_rate": 3.887762107284169e-05,
+      "loss": 0.5143,
+      "step": 245000
+    },
+    {
+      "epoch": 3.46,
+      "learning_rate": 3.885388269633774e-05,
+      "loss": 0.5326,
+      "step": 245500
+    },
+    {
+      "epoch": 3.47,
+      "learning_rate": 3.883014431983379e-05,
+      "loss": 0.5152,
+      "step": 246000
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 3.880640594332985e-05,
+      "loss": 0.5288,
+      "step": 246500
+    },
+    {
+      "epoch": 3.48,
+      "learning_rate": 3.8782667566825904e-05,
+      "loss": 0.517,
+      "step": 247000
+    },
+    {
+      "epoch": 3.49,
+      "learning_rate": 3.8758929190321966e-05,
+      "loss": 0.5092,
+      "step": 247500
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 3.8735238290571026e-05,
+      "loss": 0.5147,
+      "step": 248000
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 3.871149991406708e-05,
+      "loss": 0.5175,
+      "step": 248500
+    },
+    {
+      "epoch": 3.51,
+      "learning_rate": 3.868780901431614e-05,
+      "loss": 0.4983,
+      "step": 249000
+    },
+    {
+      "epoch": 3.52,
+      "learning_rate": 3.866411811456521e-05,
+      "loss": 0.5145,
+      "step": 249500
+    },
+    {
+      "epoch": 3.53,
+      "learning_rate": 3.8640379738061264e-05,
+      "loss": 0.5226,
+      "step": 250000
+    },
+    {
+      "epoch": 3.53,
+      "learning_rate": 3.861664136155731e-05,
+      "loss": 0.5148,
+      "step": 250500
+    },
+    {
+      "epoch": 3.54,
+      "learning_rate": 3.859290298505337e-05,
+      "loss": 0.5172,
+      "step": 251000
+    },
+    {
+      "epoch": 3.55,
+      "learning_rate": 3.856916460854942e-05,
+      "loss": 0.5245,
+      "step": 251500
+    },
+    {
+      "epoch": 3.55,
+      "learning_rate": 3.854547370879849e-05,
+      "loss": 0.5164,
+      "step": 252000
+    },
+    {
+      "epoch": 3.56,
+      "learning_rate": 3.8521735332294545e-05,
+      "loss": 0.5087,
+      "step": 252500
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 3.84979969557906e-05,
+      "loss": 0.517,
+      "step": 253000
+    },
+    {
+      "epoch": 3.57,
+      "learning_rate": 3.8474258579286656e-05,
+      "loss": 0.5089,
+      "step": 253500
+    },
+    {
+      "epoch": 3.58,
+      "learning_rate": 3.8450520202782705e-05,
+      "loss": 0.5045,
+      "step": 254000
+    },
+    {
+      "epoch": 3.59,
+      "learning_rate": 3.842682930303177e-05,
+      "loss": 0.5169,
+      "step": 254500
+    },
+    {
+      "epoch": 3.6,
+      "learning_rate": 3.840309092652783e-05,
+      "loss": 0.5204,
+      "step": 255000
+    },
+    {
+      "epoch": 3.6,
+      "learning_rate": 3.837935255002388e-05,
+      "loss": 0.5064,
+      "step": 255500
+    },
+    {
+      "epoch": 3.61,
+      "learning_rate": 3.835561417351994e-05,
+      "loss": 0.5074,
+      "step": 256000
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 3.8331875797015994e-05,
+      "loss": 0.5147,
+      "step": 256500
+    },
+    {
+      "epoch": 3.62,
+      "learning_rate": 3.830813742051205e-05,
+      "loss": 0.5031,
+      "step": 257000
+    },
+    {
+      "epoch": 3.63,
+      "learning_rate": 3.8284399044008104e-05,
+      "loss": 0.5046,
+      "step": 257500
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 3.826066066750415e-05,
+      "loss": 0.5023,
+      "step": 258000
+    },
+    {
+      "epoch": 3.64,
+      "learning_rate": 3.8236922291000215e-05,
+      "loss": 0.5053,
+      "step": 258500
+    },
+    {
+      "epoch": 3.65,
+      "learning_rate": 3.821318391449627e-05,
+      "loss": 0.5145,
+      "step": 259000
+    },
+    {
+      "epoch": 3.66,
+      "learning_rate": 3.8189445537992326e-05,
+      "loss": 0.5037,
+      "step": 259500
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 3.816570716148838e-05,
+      "loss": 0.5164,
+      "step": 260000
+    },
+    {
+      "epoch": 3.67,
+      "learning_rate": 3.814196878498443e-05,
+      "loss": 0.5089,
+      "step": 260500
+    },
+    {
+      "epoch": 3.68,
+      "learning_rate": 3.8118230408480486e-05,
+      "loss": 0.499,
+      "step": 261000
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 3.809453950872955e-05,
+      "loss": 0.5004,
+      "step": 261500
+    },
+    {
+      "epoch": 3.69,
+      "learning_rate": 3.807084860897861e-05,
+      "loss": 0.4955,
+      "step": 262000
+    },
+    {
+      "epoch": 3.7,
+      "learning_rate": 3.804715770922768e-05,
+      "loss": 0.5021,
+      "step": 262500
+    },
+    {
+      "epoch": 3.71,
+      "learning_rate": 3.802341933272373e-05,
+      "loss": 0.5064,
+      "step": 263000
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 3.799968095621979e-05,
+      "loss": 0.4947,
+      "step": 263500
+    },
+    {
+      "epoch": 3.72,
+      "learning_rate": 3.7975942579715846e-05,
+      "loss": 0.5033,
+      "step": 264000
+    },
+    {
+      "epoch": 3.73,
+      "learning_rate": 3.79522042032119e-05,
+      "loss": 0.5022,
+      "step": 264500
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 3.792846582670796e-05,
+      "loss": 0.4956,
+      "step": 265000
+    },
+    {
+      "epoch": 3.74,
+      "learning_rate": 3.790477492695702e-05,
+      "loss": 0.5102,
+      "step": 265500
+    },
+    {
+      "epoch": 3.75,
+      "learning_rate": 3.788103655045307e-05,
+      "loss": 0.509,
+      "step": 266000
+    },
+    {
+      "epoch": 3.76,
+      "learning_rate": 3.785729817394913e-05,
+      "loss": 0.5048,
+      "step": 266500
+    },
+    {
+      "epoch": 3.76,
+      "learning_rate": 3.783355979744518e-05,
+      "loss": 0.4954,
+      "step": 267000
+    },
+    {
+      "epoch": 3.77,
+      "learning_rate": 3.780982142094124e-05,
+      "loss": 0.5009,
+      "step": 267500
+    },
+    {
+      "epoch": 3.78,
+      "learning_rate": 3.7786083044437294e-05,
+      "loss": 0.4909,
+      "step": 268000
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 3.776234466793335e-05,
+      "loss": 0.5018,
+      "step": 268500
+    },
+    {
+      "epoch": 3.79,
+      "learning_rate": 3.77386062914294e-05,
+      "loss": 0.4903,
+      "step": 269000
+    },
+    {
+      "epoch": 3.8,
+      "learning_rate": 3.7714867914925454e-05,
+      "loss": 0.504,
+      "step": 269500
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 3.7691129538421516e-05,
+      "loss": 0.5088,
+      "step": 270000
+    },
+    {
+      "epoch": 3.81,
+      "learning_rate": 3.7667438638670576e-05,
+      "loss": 0.5009,
+      "step": 270500
+    },
+    {
+      "epoch": 3.82,
+      "learning_rate": 3.7643747738919636e-05,
+      "loss": 0.4924,
+      "step": 271000
+    },
+    {
+      "epoch": 3.83,
+      "learning_rate": 3.76200568391687e-05,
+      "loss": 0.4999,
+      "step": 271500
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 3.759631846266476e-05,
+      "loss": 0.4997,
+      "step": 272000
+    },
+    {
+      "epoch": 3.84,
+      "learning_rate": 3.7572580086160813e-05,
+      "loss": 0.4968,
+      "step": 272500
+    },
+    {
+      "epoch": 3.85,
+      "learning_rate": 3.754884170965687e-05,
+      "loss": 0.4905,
+      "step": 273000
+    },
+    {
+      "epoch": 3.86,
+      "learning_rate": 3.7525103333152924e-05,
+      "loss": 0.5055,
+      "step": 273500
+    },
+    {
+      "epoch": 3.86,
+      "learning_rate": 3.750136495664898e-05,
+      "loss": 0.4918,
+      "step": 274000
+    },
+    {
+      "epoch": 3.87,
+      "learning_rate": 3.747762658014503e-05,
+      "loss": 0.4947,
+      "step": 274500
+    },
+    {
+      "epoch": 3.88,
+      "learning_rate": 3.745388820364109e-05,
+      "loss": 0.5016,
+      "step": 275000
+    },
+    {
+      "epoch": 3.88,
+      "learning_rate": 3.7430149827137146e-05,
+      "loss": 0.5084,
+      "step": 275500
+    },
+    {
+      "epoch": 3.89,
+      "learning_rate": 3.74064114506332e-05,
+      "loss": 0.4882,
+      "step": 276000
+    },
+    {
+      "epoch": 3.9,
+      "learning_rate": 3.738272055088226e-05,
+      "loss": 0.4948,
+      "step": 276500
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 3.735898217437832e-05,
+      "loss": 0.4971,
+      "step": 277000
+    },
+    {
+      "epoch": 3.91,
+      "learning_rate": 3.733524379787437e-05,
+      "loss": 0.4867,
+      "step": 277500
+    },
+    {
+      "epoch": 3.92,
+      "learning_rate": 3.731150542137043e-05,
+      "loss": 0.4917,
+      "step": 278000
+    },
+    {
+      "epoch": 3.93,
+      "learning_rate": 3.7287767044866484e-05,
+      "loss": 0.4853,
+      "step": 278500
+    },
+    {
+      "epoch": 3.93,
+      "learning_rate": 3.726402866836254e-05,
+      "loss": 0.4948,
+      "step": 279000
+    },
+    {
+      "epoch": 3.94,
+      "learning_rate": 3.7240290291858594e-05,
+      "loss": 0.4935,
+      "step": 279500
+    },
+    {
+      "epoch": 3.95,
+      "learning_rate": 3.721655191535465e-05,
+      "loss": 0.4906,
+      "step": 280000
+    },
+    {
+      "epoch": 3.96,
+      "learning_rate": 3.71928135388507e-05,
+      "loss": 0.4992,
+      "step": 280500
+    },
+    {
+      "epoch": 3.96,
+      "learning_rate": 3.716907516234676e-05,
+      "loss": 0.4814,
+      "step": 281000
+    },
+    {
+      "epoch": 3.97,
+      "learning_rate": 3.7145336785842816e-05,
+      "loss": 0.4903,
+      "step": 281500
+    },
+    {
+      "epoch": 3.98,
+      "learning_rate": 3.712159840933887e-05,
+      "loss": 0.4869,
+      "step": 282000
+    },
+    {
+      "epoch": 3.98,
+      "learning_rate": 3.709786003283493e-05,
+      "loss": 0.493,
+      "step": 282500
+    },
+    {
+      "epoch": 3.99,
+      "learning_rate": 3.7074121656330976e-05,
+      "loss": 0.4909,
+      "step": 283000
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.705038327982703e-05,
+      "loss": 0.4873,
+      "step": 283500
+    },
+    {
+      "epoch": 4.0,
+      "eval_bleu": 43.6153,
+      "eval_gen_len": 13.6544,
+      "eval_loss": 0.9757916927337646,
+      "eval_runtime": 9810.7798,
+      "eval_samples_per_second": 14.457,
+      "eval_steps_per_second": 1.807,
+      "step": 283678
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 3.70266923800761e-05,
+      "loss": 0.4799,
+      "step": 284000
+    },
+    {
+      "epoch": 4.01,
+      "learning_rate": 3.700300148032516e-05,
+      "loss": 0.4853,
+      "step": 284500
+    },
+    {
+      "epoch": 4.02,
+      "learning_rate": 3.6979263103821214e-05,
+      "loss": 0.4872,
+      "step": 285000
+    },
+    {
+      "epoch": 4.03,
+      "learning_rate": 3.695552472731727e-05,
+      "loss": 0.4809,
+      "step": 285500
+    },
+    {
+      "epoch": 4.03,
+      "learning_rate": 3.6931786350813324e-05,
+      "loss": 0.4787,
+      "step": 286000
+    },
+    {
+      "epoch": 4.04,
+      "learning_rate": 3.690804797430938e-05,
+      "loss": 0.4797,
+      "step": 286500
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 3.6884309597805435e-05,
+      "loss": 0.4811,
+      "step": 287000
+    },
+    {
+      "epoch": 4.05,
+      "learning_rate": 3.686057122130149e-05,
+      "loss": 0.4728,
+      "step": 287500
+    },
+    {
+      "epoch": 4.06,
+      "learning_rate": 3.6836832844797546e-05,
+      "loss": 0.4756,
+      "step": 288000
+    },
+    {
+      "epoch": 4.07,
+      "learning_rate": 3.68130944682936e-05,
+      "loss": 0.4728,
+      "step": 288500
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 3.678935609178966e-05,
+      "loss": 0.4844,
+      "step": 289000
+    },
+    {
+      "epoch": 4.08,
+      "learning_rate": 3.676561771528571e-05,
+      "loss": 0.4677,
+      "step": 289500
+    },
+    {
+      "epoch": 4.09,
+      "learning_rate": 3.674192681553478e-05,
+      "loss": 0.4631,
+      "step": 290000
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 3.6718188439030835e-05,
+      "loss": 0.4626,
+      "step": 290500
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 3.6694450062526884e-05,
+      "loss": 0.4705,
+      "step": 291000
+    },
+    {
+      "epoch": 4.11,
+      "learning_rate": 3.667071168602294e-05,
+      "loss": 0.4582,
+      "step": 291500
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 3.6646973309518994e-05,
+      "loss": 0.4668,
+      "step": 292000
+    },
+    {
+      "epoch": 4.12,
+      "learning_rate": 3.662323493301505e-05,
+      "loss": 0.4597,
+      "step": 292500
+    },
+    {
+      "epoch": 4.13,
+      "learning_rate": 3.6599496556511105e-05,
+      "loss": 0.4518,
+      "step": 293000
+    },
+    {
+      "epoch": 4.14,
+      "learning_rate": 3.657575818000716e-05,
+      "loss": 0.4586,
+      "step": 293500
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 3.6552019803503216e-05,
+      "loss": 0.4413,
+      "step": 294000
+    },
+    {
+      "epoch": 4.15,
+      "learning_rate": 3.6528328903752276e-05,
+      "loss": 0.4528,
+      "step": 294500
+    },
+    {
+      "epoch": 4.16,
+      "learning_rate": 3.650459052724833e-05,
+      "loss": 0.4512,
+      "step": 295000
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 3.64808996274974e-05,
+      "loss": 0.4555,
+      "step": 295500
+    },
+    {
+      "epoch": 4.17,
+      "learning_rate": 3.6457161250993454e-05,
+      "loss": 0.4474,
+      "step": 296000
+    },
+    {
+      "epoch": 4.18,
+      "learning_rate": 3.643342287448951e-05,
+      "loss": 0.4426,
+      "step": 296500
+    },
+    {
+      "epoch": 4.19,
+      "learning_rate": 3.6409684497985565e-05,
+      "loss": 0.4459,
+      "step": 297000
+    },
+    {
+      "epoch": 4.19,
+      "learning_rate": 3.6385993598234625e-05,
+      "loss": 0.4473,
+      "step": 297500
+    },
+    {
+      "epoch": 4.2,
+      "learning_rate": 3.636225522173068e-05,
+      "loss": 0.4426,
+      "step": 298000
+    },
+    {
+      "epoch": 4.21,
+      "learning_rate": 3.6338516845226736e-05,
+      "loss": 0.4413,
+      "step": 298500
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 3.631477846872279e-05,
+      "loss": 0.4335,
+      "step": 299000
+    },
+    {
+      "epoch": 4.22,
+      "learning_rate": 3.629104009221885e-05,
+      "loss": 0.4422,
+      "step": 299500
+    },
+    {
+      "epoch": 4.23,
+      "learning_rate": 3.626734919246791e-05,
+      "loss": 0.4566,
+      "step": 300000
+    },
+    {
+      "epoch": 4.24,
+      "learning_rate": 3.624361081596396e-05,
+      "loss": 0.4375,
+      "step": 300500
+    },
+    {
+      "epoch": 4.24,
+      "learning_rate": 3.621987243946002e-05,
+      "loss": 0.4433,
+      "step": 301000
+    },
+    {
+      "epoch": 4.25,
+      "learning_rate": 3.6196181539709084e-05,
+      "loss": 0.4311,
+      "step": 301500
+    },
+    {
+      "epoch": 4.26,
+      "learning_rate": 3.617244316320514e-05,
+      "loss": 0.4282,
+      "step": 302000
+    },
+    {
+      "epoch": 4.27,
+      "learning_rate": 3.6148704786701195e-05,
+      "loss": 0.4302,
+      "step": 302500
+    },
+    {
+      "epoch": 4.27,
+      "learning_rate": 3.612496641019725e-05,
+      "loss": 0.446,
+      "step": 303000
+    },
+    {
+      "epoch": 4.28,
+      "learning_rate": 3.61012280336933e-05,
+      "loss": 0.4282,
+      "step": 303500
+    },
+    {
+      "epoch": 4.29,
+      "learning_rate": 3.607748965718936e-05,
+      "loss": 0.4335,
+      "step": 304000
+    },
+    {
+      "epoch": 4.29,
+      "learning_rate": 3.605375128068542e-05,
+      "loss": 0.433,
+      "step": 304500
+    },
+    {
+      "epoch": 4.3,
+      "learning_rate": 3.603001290418147e-05,
+      "loss": 0.4168,
+      "step": 305000
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 3.600627452767753e-05,
+      "loss": 0.4391,
+      "step": 305500
+    },
+    {
+      "epoch": 4.31,
+      "learning_rate": 3.598253615117358e-05,
+      "loss": 0.433,
+      "step": 306000
+    },
+    {
+      "epoch": 4.32,
+      "learning_rate": 3.595879777466963e-05,
+      "loss": 0.4391,
+      "step": 306500
+    },
+    {
+      "epoch": 4.33,
+      "learning_rate": 3.593505939816569e-05,
+      "loss": 0.4276,
+      "step": 307000
+    },
+    {
+      "epoch": 4.34,
+      "learning_rate": 3.591132102166174e-05,
+      "loss": 0.4343,
+      "step": 307500
+    },
+    {
+      "epoch": 4.34,
+      "learning_rate": 3.5887582645157805e-05,
+      "loss": 0.4407,
+      "step": 308000
+    },
+    {
+      "epoch": 4.35,
+      "learning_rate": 3.5863891745406865e-05,
+      "loss": 0.4258,
+      "step": 308500
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 3.584015336890292e-05,
+      "loss": 0.4285,
+      "step": 309000
+    },
+    {
+      "epoch": 4.36,
+      "learning_rate": 3.581641499239897e-05,
+      "loss": 0.4316,
+      "step": 309500
+    },
+    {
+      "epoch": 4.37,
+      "learning_rate": 3.5792676615895025e-05,
+      "loss": 0.4249,
+      "step": 310000
+    },
+    {
+      "epoch": 4.38,
+      "learning_rate": 3.576898571614409e-05,
+      "loss": 0.4276,
+      "step": 310500
+    },
+    {
+      "epoch": 4.39,
+      "learning_rate": 3.574524733964015e-05,
+      "loss": 0.4286,
+      "step": 311000
+    },
+    {
+      "epoch": 4.39,
+      "learning_rate": 3.57215089631362e-05,
+      "loss": 0.4195,
+      "step": 311500
+    },
+    {
+      "epoch": 4.4,
+      "learning_rate": 3.569777058663226e-05,
+      "loss": 0.419,
+      "step": 312000
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 3.567407968688132e-05,
+      "loss": 0.422,
+      "step": 312500
+    },
+    {
+      "epoch": 4.41,
+      "learning_rate": 3.5650388787130385e-05,
+      "loss": 0.4299,
+      "step": 313000
+    },
+    {
+      "epoch": 4.42,
+      "learning_rate": 3.562665041062644e-05,
+      "loss": 0.4341,
+      "step": 313500
+    },
+    {
+      "epoch": 4.43,
+      "learning_rate": 3.5602912034122496e-05,
+      "loss": 0.4122,
+      "step": 314000
+    },
+    {
+      "epoch": 4.43,
+      "learning_rate": 3.5579173657618544e-05,
+      "loss": 0.4251,
+      "step": 314500
+    },
+    {
+      "epoch": 4.44,
+      "learning_rate": 3.55554352811146e-05,
+      "loss": 0.4298,
+      "step": 315000
+    },
+    {
+      "epoch": 4.45,
+      "learning_rate": 3.553169690461066e-05,
+      "loss": 0.4183,
+      "step": 315500
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 3.550795852810672e-05,
+      "loss": 0.4199,
+      "step": 316000
+    },
+    {
+      "epoch": 4.46,
+      "learning_rate": 3.548426762835578e-05,
+      "loss": 0.4281,
+      "step": 316500
+    },
+    {
+      "epoch": 4.47,
+      "learning_rate": 3.546052925185183e-05,
+      "loss": 0.4158,
+      "step": 317000
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 3.543679087534789e-05,
+      "loss": 0.4267,
+      "step": 317500
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 3.541305249884394e-05,
+      "loss": 0.4193,
+      "step": 318000
+    },
+    {
+      "epoch": 4.49,
+      "learning_rate": 3.538931412234e-05,
+      "loss": 0.4106,
+      "step": 318500
+    },
+    {
+      "epoch": 4.5,
+      "learning_rate": 3.5365575745836055e-05,
+      "loss": 0.4186,
+      "step": 319000
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 3.534183736933211e-05,
+      "loss": 0.4141,
+      "step": 319500
+    },
+    {
+      "epoch": 4.51,
+      "learning_rate": 3.5318098992828166e-05,
+      "loss": 0.398,
+      "step": 320000
+    },
+    {
+      "epoch": 4.52,
+      "learning_rate": 3.529436061632422e-05,
+      "loss": 0.4241,
+      "step": 320500
+    },
+    {
+      "epoch": 4.53,
+      "learning_rate": 3.527062223982027e-05,
+      "loss": 0.422,
+      "step": 321000
+    },
+    {
+      "epoch": 4.53,
+      "learning_rate": 3.5246883863316325e-05,
+      "loss": 0.4167,
+      "step": 321500
+    },
+    {
+      "epoch": 4.54,
+      "learning_rate": 3.522319296356539e-05,
+      "loss": 0.4214,
+      "step": 322000
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 3.519950206381445e-05,
+      "loss": 0.4171,
+      "step": 322500
+    },
+    {
+      "epoch": 4.55,
+      "learning_rate": 3.517576368731051e-05,
+      "loss": 0.4211,
+      "step": 323000
+    },
+    {
+      "epoch": 4.56,
+      "learning_rate": 3.5152072787559574e-05,
+      "loss": 0.41,
+      "step": 323500
+    },
+    {
+      "epoch": 4.57,
+      "learning_rate": 3.512833441105563e-05,
+      "loss": 0.4243,
+      "step": 324000
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 3.5104596034551685e-05,
+      "loss": 0.4066,
+      "step": 324500
+    },
+    {
+      "epoch": 4.58,
+      "learning_rate": 3.508085765804774e-05,
+      "loss": 0.4055,
+      "step": 325000
+    },
+    {
+      "epoch": 4.59,
+      "learning_rate": 3.5057119281543796e-05,
+      "loss": 0.419,
+      "step": 325500
+    },
+    {
+      "epoch": 4.6,
+      "learning_rate": 3.5033380905039845e-05,
+      "loss": 0.4222,
+      "step": 326000
+    },
+    {
+      "epoch": 4.6,
+      "learning_rate": 3.50096425285359e-05,
+      "loss": 0.4092,
+      "step": 326500
+    },
+    {
+      "epoch": 4.61,
+      "learning_rate": 3.498595162878497e-05,
+      "loss": 0.4094,
+      "step": 327000
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 3.496221325228102e-05,
+      "loss": 0.4102,
+      "step": 327500
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 3.493847487577708e-05,
+      "loss": 0.4096,
+      "step": 328000
+    },
+    {
+      "epoch": 4.63,
+      "learning_rate": 3.4914736499273133e-05,
+      "loss": 0.4052,
+      "step": 328500
+    },
+    {
+      "epoch": 4.64,
+      "learning_rate": 3.489099812276919e-05,
+      "loss": 0.4076,
+      "step": 329000
+    },
+    {
+      "epoch": 4.65,
+      "learning_rate": 3.486725974626524e-05,
+      "loss": 0.414,
+      "step": 329500
+    },
+    {
+      "epoch": 4.65,
+      "learning_rate": 3.48435213697613e-05,
+      "loss": 0.4084,
+      "step": 330000
+    },
+    {
+      "epoch": 4.66,
+      "learning_rate": 3.4819782993257355e-05,
+      "loss": 0.405,
+      "step": 330500
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 3.479604461675341e-05,
+      "loss": 0.4167,
+      "step": 331000
+    },
+    {
+      "epoch": 4.67,
+      "learning_rate": 3.4772306240249466e-05,
+      "loss": 0.4113,
+      "step": 331500
+    },
+    {
+      "epoch": 4.68,
+      "learning_rate": 3.4748615340498526e-05,
+      "loss": 0.403,
+      "step": 332000
+    },
+    {
+      "epoch": 4.69,
+      "learning_rate": 3.472492444074759e-05,
+      "loss": 0.4066,
+      "step": 332500
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 3.470118606424365e-05,
+      "loss": 0.3965,
+      "step": 333000
+    },
+    {
+      "epoch": 4.7,
+      "learning_rate": 3.467749516449271e-05,
+      "loss": 0.4052,
+      "step": 333500
+    },
+    {
+      "epoch": 4.71,
+      "learning_rate": 3.4653756787988764e-05,
+      "loss": 0.4093,
+      "step": 334000
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 3.463001841148481e-05,
+      "loss": 0.4015,
+      "step": 334500
+    },
+    {
+      "epoch": 4.72,
+      "learning_rate": 3.4606280034980875e-05,
+      "loss": 0.4096,
+      "step": 335000
+    },
+    {
+      "epoch": 4.73,
+      "learning_rate": 3.458254165847693e-05,
+      "loss": 0.4041,
+      "step": 335500
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 3.4558803281972986e-05,
+      "loss": 0.4046,
+      "step": 336000
+    },
+    {
+      "epoch": 4.74,
+      "learning_rate": 3.453506490546904e-05,
+      "loss": 0.4112,
+      "step": 336500
+    },
+    {
+      "epoch": 4.75,
+      "learning_rate": 3.4511326528965097e-05,
+      "loss": 0.4107,
+      "step": 337000
+    },
+    {
+      "epoch": 4.76,
+      "learning_rate": 3.4487588152461145e-05,
+      "loss": 0.406,
+      "step": 337500
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 3.44638497759572e-05,
+      "loss": 0.4008,
+      "step": 338000
+    },
+    {
+      "epoch": 4.77,
+      "learning_rate": 3.444011139945326e-05,
+      "loss": 0.4088,
+      "step": 338500
+    },
+    {
+      "epoch": 4.78,
+      "learning_rate": 3.441637302294932e-05,
+      "loss": 0.4006,
+      "step": 339000
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 3.4392634646445374e-05,
+      "loss": 0.3962,
+      "step": 339500
+    },
+    {
+      "epoch": 4.79,
+      "learning_rate": 3.436889626994142e-05,
+      "loss": 0.4048,
+      "step": 340000
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 3.434515789343748e-05,
+      "loss": 0.4076,
+      "step": 340500
+    },
+    {
+      "epoch": 4.81,
+      "learning_rate": 3.4321419516933533e-05,
+      "loss": 0.4108,
+      "step": 341000
+    },
+    {
+      "epoch": 4.82,
+      "learning_rate": 3.429768114042959e-05,
+      "loss": 0.4037,
+      "step": 341500
+    },
+    {
+      "epoch": 4.82,
+      "learning_rate": 3.4273942763925644e-05,
+      "loss": 0.4019,
+      "step": 342000
+    },
+    {
+      "epoch": 4.83,
+      "learning_rate": 3.42502043874217e-05,
+      "loss": 0.3984,
+      "step": 342500
+    },
+    {
+      "epoch": 4.84,
+      "learning_rate": 3.4226466010917755e-05,
+      "loss": 0.4065,
+      "step": 343000
+    },
+    {
+      "epoch": 4.84,
+      "learning_rate": 3.4202775111166815e-05,
+      "loss": 0.3995,
+      "step": 343500
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 3.417903673466287e-05,
+      "loss": 0.3979,
+      "step": 344000
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 3.4155298358158926e-05,
+      "loss": 0.4088,
+      "step": 344500
+    },
+    {
+      "epoch": 4.86,
+      "learning_rate": 3.413155998165499e-05,
+      "loss": 0.3951,
+      "step": 345000
+    },
+    {
+      "epoch": 4.87,
+      "learning_rate": 3.4107821605151044e-05,
+      "loss": 0.4011,
+      "step": 345500
+    },
+    {
+      "epoch": 4.88,
+      "learning_rate": 3.408408322864709e-05,
+      "loss": 0.4073,
+      "step": 346000
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 3.406034485214315e-05,
+      "loss": 0.4042,
+      "step": 346500
+    },
+    {
+      "epoch": 4.89,
+      "learning_rate": 3.4036606475639203e-05,
+      "loss": 0.3924,
+      "step": 347000
+    },
+    {
+      "epoch": 4.9,
+      "learning_rate": 3.401286809913526e-05,
+      "loss": 0.3928,
+      "step": 347500
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 3.3989177199384326e-05,
+      "loss": 0.4044,
+      "step": 348000
+    },
+    {
+      "epoch": 4.91,
+      "learning_rate": 3.396543882288038e-05,
+      "loss": 0.3934,
+      "step": 348500
+    },
+    {
+      "epoch": 4.92,
+      "learning_rate": 3.394170044637644e-05,
+      "loss": 0.3946,
+      "step": 349000
+    },
+    {
+      "epoch": 4.93,
+      "learning_rate": 3.391796206987249e-05,
+      "loss": 0.3907,
+      "step": 349500
+    },
+    {
+      "epoch": 4.94,
+      "learning_rate": 3.389427117012155e-05,
+      "loss": 0.3947,
+      "step": 350000
+    },
+    {
+      "epoch": 4.94,
+      "learning_rate": 3.387062774712362e-05,
+      "loss": 0.3962,
+      "step": 350500
+    },
+    {
+      "epoch": 4.95,
+      "learning_rate": 3.384688937061968e-05,
+      "loss": 0.3961,
+      "step": 351000
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 3.3823150994115734e-05,
+      "loss": 0.4004,
+      "step": 351500
+    },
+    {
+      "epoch": 4.96,
+      "learning_rate": 3.379941261761179e-05,
+      "loss": 0.3917,
+      "step": 352000
+    },
+    {
+      "epoch": 4.97,
+      "learning_rate": 3.3775721717860856e-05,
+      "loss": 0.3939,
+      "step": 352500
+    },
+    {
+      "epoch": 4.98,
+      "learning_rate": 3.3752030818109916e-05,
+      "loss": 0.3942,
+      "step": 353000
+    },
+    {
+      "epoch": 4.98,
+      "learning_rate": 3.372829244160597e-05,
+      "loss": 0.3975,
+      "step": 353500
+    },
+    {
+      "epoch": 4.99,
+      "learning_rate": 3.370455406510202e-05,
+      "loss": 0.4001,
+      "step": 354000
+    },
+    {
+      "epoch": 5.0,
+      "learning_rate": 3.3680815688598076e-05,
+      "loss": 0.388,
+      "step": 354500
+    },
+    {
+      "epoch": 5.0,
+      "eval_bleu": 43.3763,
+      "eval_gen_len": 13.6067,
+      "eval_loss": 1.0399531126022339,
+      "eval_runtime": 9774.1481,
+      "eval_samples_per_second": 14.512,
+      "eval_steps_per_second": 1.814,
+      "step": 354597
+    },
+    {
+      "epoch": 5.0,
+      "step": 354597,
+      "total_flos": 1.2295340767110496e+19,
+      "train_loss": 0.7189551201696451,
+      "train_runtime": 324742.1955,
+      "train_samples_per_second": 52.413,
+      "train_steps_per_second": 3.276
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 1063785,
+  "num_train_epochs": 15,
+  "save_steps": 500,
+  "total_flos": 1.2295340767110496e+19,
+  "trial_name": null,
+  "trial_params": null
+}
diff --git a/training_args.bin b/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..dbbe22d1f032c0b0fd9eedfe9ae519ce9ccd36a7
--- /dev/null
+++ b/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0adf1a980c6128833811b7e6eb546e117ffd3efb8c21dc7de95b5e76a5b21b8d
+size 4728