Upload folder using huggingface_hub

README.md

@@ -0,0 +1,64 @@
+---
+library_name: transformers
+license: other
+base_model: NousResearch/Meta-Llama-3-8B-Instruct
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+model-index:
+- name: sft
+  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. -->
+
+# sft
+
+This model is a fine-tuned version of [NousResearch/Meta-Llama-3-8B-Instruct](https://huggingface.co/NousResearch/Meta-Llama-3-8B-Instruct) on the identity and the eightwords-20241112-alapaca datasets.
+It achieves the following results on the evaluation set:
+- Loss: 1.6951
+
+## 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: 1e-05
+- train_batch_size: 2
+- eval_batch_size: 1
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 16
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_ratio: 0.1
+- num_epochs: 32.0
+
+### Training results
+
+| Training Loss | Epoch   | Step | Validation Loss |
+|:-------------:|:-------:|:----:|:---------------:|
+| 0.7669        | 14.0474 | 2000 | 1.4163          |
+| 0.4249        | 28.0948 | 4000 | 1.6929          |
+
+
+### Framework versions
+
+- Transformers 4.44.2
+- Pytorch 2.5.0+cu121
+- Datasets 3.1.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 31.91571553994732,
+    "eval_loss": 1.6950665712356567,
+    "eval_runtime": 10.5237,
+    "eval_samples_per_second": 24.136,
+    "eval_steps_per_second": 24.136,
+    "total_flos": 9.547395325030564e+17,
+    "train_loss": 0.8426483250682203,
+    "train_runtime": 5512.317,
+    "train_samples_per_second": 13.218,
+    "train_steps_per_second": 0.824
+}

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "NousResearch/Meta-Llama-3-8B-Instruct",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "eos_token_id": 128009,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 8192,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 500000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.44.2",
+  "use_cache": false,
+  "vocab_size": 128256
+}

eval_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 31.91571553994732,
+    "eval_loss": 1.6950665712356567,
+    "eval_runtime": 10.5237,
+    "eval_samples_per_second": 24.136,
+    "eval_steps_per_second": 24.136
+}

generation_config.json

@@ -0,0 +1,12 @@
+{
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": [
+    128001,
+    128009
+  ],
+  "max_length": 4096,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.44.2"
+}

model-00001-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ecc0341ed2b146b4fe2c0f3df48a5f6bcaf29e0e030a8f079907f7c299b12f8
+size 4976698672

model-00002-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d526a06532449d4ea615e3da511bb21ee3236fc7275b627cfd368a37a1f31b6b
+size 4999802720

model-00003-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aefe54e6f89f6fe79b75ed4c338b2705fbab32bd191f9db3e48dbc56a830d6fe
+size 4915916176

model-00004-of-00004.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:753de5569f5a0596435aea32b491a677ec0ba7a3b304cd5acd3d36a1cdb637c0
+size 1168138808

model.safetensors.index.json

@@ -0,0 +1,298 @@
+{
+  "metadata": {
+    "total_size": 16060522496
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00004-of-00004.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00004.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.norm.weight": "model-00004-of-00004.safetensors"
+  }
+}

runs/Nov12_11-04-30_bda85f71ac37/events.out.tfevents.1731409873.bda85f71ac37.2905.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93916c0280f104f597f2e284742034b5e917cbe8251146f3528237c2713d3bd3
+size 5386

runs/Nov12_11-12-36_bda85f71ac37/events.out.tfevents.1731409975.bda85f71ac37.5524.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7093aebb8811fe2a726457c66224a0ce0fcc037489940acd74f3be0362e69c9e
+size 101821

runs/Nov12_11-12-36_bda85f71ac37/events.out.tfevents.1731415555.bda85f71ac37.5524.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8abc7283643fc435a8a368d3d3c09104b5f361a36440f14b0870f95de79d078a
+size 359

special_tokens_map.json

@@ -0,0 +1,17 @@
+{
+  "bos_token": {
+    "content": "<|begin_of_text|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|eot_id|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|eot_id|>"
+}

tokenizer_config.json

@@ -0,0 +1,2065 @@
+{
+  "added_tokens_decoder": {
+    "128000": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128001": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128002": {
+      "content": "<|reserved_special_token_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128003": {
+      "content": "<|reserved_special_token_1|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128004": {
+      "content": "<|reserved_special_token_2|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128005": {
+      "content": "<|reserved_special_token_3|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128006": {
+      "content": "<|start_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128007": {
+      "content": "<|end_header_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128008": {
+      "content": "<|reserved_special_token_4|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128009": {
+      "content": "<|eot_id|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128010": {
+      "content": "<|reserved_special_token_5|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128011": {
+      "content": "<|reserved_special_token_6|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128012": {
+      "content": "<|reserved_special_token_7|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128013": {
+      "content": "<|reserved_special_token_8|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128014": {
+      "content": "<|reserved_special_token_9|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128015": {
+      "content": "<|reserved_special_token_10|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128016": {
+      "content": "<|reserved_special_token_11|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128017": {
+      "content": "<|reserved_special_token_12|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128018": {
+      "content": "<|reserved_special_token_13|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128019": {
+      "content": "<|reserved_special_token_14|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128020": {
+      "content": "<|reserved_special_token_15|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128021": {
+      "content": "<|reserved_special_token_16|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128022": {
+      "content": "<|reserved_special_token_17|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128023": {
+      "content": "<|reserved_special_token_18|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128024": {
+      "content": "<|reserved_special_token_19|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128025": {
+      "content": "<|reserved_special_token_20|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128026": {
+      "content": "<|reserved_special_token_21|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128027": {
+      "content": "<|reserved_special_token_22|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128028": {
+      "content": "<|reserved_special_token_23|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128029": {
+      "content": "<|reserved_special_token_24|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128030": {
+      "content": "<|reserved_special_token_25|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128031": {
+      "content": "<|reserved_special_token_26|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128032": {
+      "content": "<|reserved_special_token_27|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128033": {
+      "content": "<|reserved_special_token_28|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128034": {
+      "content": "<|reserved_special_token_29|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128035": {
+      "content": "<|reserved_special_token_30|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128036": {
+      "content": "<|reserved_special_token_31|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128037": {
+      "content": "<|reserved_special_token_32|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128038": {
+      "content": "<|reserved_special_token_33|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128039": {
+      "content": "<|reserved_special_token_34|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128040": {
+      "content": "<|reserved_special_token_35|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128041": {
+      "content": "<|reserved_special_token_36|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128042": {
+      "content": "<|reserved_special_token_37|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128043": {
+      "content": "<|reserved_special_token_38|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128044": {
+      "content": "<|reserved_special_token_39|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128045": {
+      "content": "<|reserved_special_token_40|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128046": {
+      "content": "<|reserved_special_token_41|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128047": {
+      "content": "<|reserved_special_token_42|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128048": {
+      "content": "<|reserved_special_token_43|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128049": {
+      "content": "<|reserved_special_token_44|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128050": {
+      "content": "<|reserved_special_token_45|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128051": {
+      "content": "<|reserved_special_token_46|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128052": {
+      "content": "<|reserved_special_token_47|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128053": {
+      "content": "<|reserved_special_token_48|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128054": {
+      "content": "<|reserved_special_token_49|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128055": {
+      "content": "<|reserved_special_token_50|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128056": {
+      "content": "<|reserved_special_token_51|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128057": {
+      "content": "<|reserved_special_token_52|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128058": {
+      "content": "<|reserved_special_token_53|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128059": {
+      "content": "<|reserved_special_token_54|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128060": {
+      "content": "<|reserved_special_token_55|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128061": {
+      "content": "<|reserved_special_token_56|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128062": {
+      "content": "<|reserved_special_token_57|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128063": {
+      "content": "<|reserved_special_token_58|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128064": {
+      "content": "<|reserved_special_token_59|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128065": {
+      "content": "<|reserved_special_token_60|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128066": {
+      "content": "<|reserved_special_token_61|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128067": {
+      "content": "<|reserved_special_token_62|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128068": {
+      "content": "<|reserved_special_token_63|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128069": {
+      "content": "<|reserved_special_token_64|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128070": {
+      "content": "<|reserved_special_token_65|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128071": {
+      "content": "<|reserved_special_token_66|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128072": {
+      "content": "<|reserved_special_token_67|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128073": {
+      "content": "<|reserved_special_token_68|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128074": {
+      "content": "<|reserved_special_token_69|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128075": {
+      "content": "<|reserved_special_token_70|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128076": {
+      "content": "<|reserved_special_token_71|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128077": {
+      "content": "<|reserved_special_token_72|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128078": {
+      "content": "<|reserved_special_token_73|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128079": {
+      "content": "<|reserved_special_token_74|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128080": {
+      "content": "<|reserved_special_token_75|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128081": {
+      "content": "<|reserved_special_token_76|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128082": {
+      "content": "<|reserved_special_token_77|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128083": {
+      "content": "<|reserved_special_token_78|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128084": {
+      "content": "<|reserved_special_token_79|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128085": {
+      "content": "<|reserved_special_token_80|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128086": {
+      "content": "<|reserved_special_token_81|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128087": {
+      "content": "<|reserved_special_token_82|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128088": {
+      "content": "<|reserved_special_token_83|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128089": {
+      "content": "<|reserved_special_token_84|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128090": {
+      "content": "<|reserved_special_token_85|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128091": {
+      "content": "<|reserved_special_token_86|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128092": {
+      "content": "<|reserved_special_token_87|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128093": {
+      "content": "<|reserved_special_token_88|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128094": {
+      "content": "<|reserved_special_token_89|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128095": {
+      "content": "<|reserved_special_token_90|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128096": {
+      "content": "<|reserved_special_token_91|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128097": {
+      "content": "<|reserved_special_token_92|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128098": {
+      "content": "<|reserved_special_token_93|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128099": {
+      "content": "<|reserved_special_token_94|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128100": {
+      "content": "<|reserved_special_token_95|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128101": {
+      "content": "<|reserved_special_token_96|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128102": {
+      "content": "<|reserved_special_token_97|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128103": {
+      "content": "<|reserved_special_token_98|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128104": {
+      "content": "<|reserved_special_token_99|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128105": {
+      "content": "<|reserved_special_token_100|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128106": {
+      "content": "<|reserved_special_token_101|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128107": {
+      "content": "<|reserved_special_token_102|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128108": {
+      "content": "<|reserved_special_token_103|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128109": {
+      "content": "<|reserved_special_token_104|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128110": {
+      "content": "<|reserved_special_token_105|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128111": {
+      "content": "<|reserved_special_token_106|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128112": {
+      "content": "<|reserved_special_token_107|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128113": {
+      "content": "<|reserved_special_token_108|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128114": {
+      "content": "<|reserved_special_token_109|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128115": {
+      "content": "<|reserved_special_token_110|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128116": {
+      "content": "<|reserved_special_token_111|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128117": {
+      "content": "<|reserved_special_token_112|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128118": {
+      "content": "<|reserved_special_token_113|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128119": {
+      "content": "<|reserved_special_token_114|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128120": {
+      "content": "<|reserved_special_token_115|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128121": {
+      "content": "<|reserved_special_token_116|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128122": {
+      "content": "<|reserved_special_token_117|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128123": {
+      "content": "<|reserved_special_token_118|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128124": {
+      "content": "<|reserved_special_token_119|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128125": {
+      "content": "<|reserved_special_token_120|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128126": {
+      "content": "<|reserved_special_token_121|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128127": {
+      "content": "<|reserved_special_token_122|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128128": {
+      "content": "<|reserved_special_token_123|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128129": {
+      "content": "<|reserved_special_token_124|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128130": {
+      "content": "<|reserved_special_token_125|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128131": {
+      "content": "<|reserved_special_token_126|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128132": {
+      "content": "<|reserved_special_token_127|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128133": {
+      "content": "<|reserved_special_token_128|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128134": {
+      "content": "<|reserved_special_token_129|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128135": {
+      "content": "<|reserved_special_token_130|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128136": {
+      "content": "<|reserved_special_token_131|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128137": {
+      "content": "<|reserved_special_token_132|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128138": {
+      "content": "<|reserved_special_token_133|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128139": {
+      "content": "<|reserved_special_token_134|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128140": {
+      "content": "<|reserved_special_token_135|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128141": {
+      "content": "<|reserved_special_token_136|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128142": {
+      "content": "<|reserved_special_token_137|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128143": {
+      "content": "<|reserved_special_token_138|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128144": {
+      "content": "<|reserved_special_token_139|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128145": {
+      "content": "<|reserved_special_token_140|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128146": {
+      "content": "<|reserved_special_token_141|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128147": {
+      "content": "<|reserved_special_token_142|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128148": {
+      "content": "<|reserved_special_token_143|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128149": {
+      "content": "<|reserved_special_token_144|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128150": {
+      "content": "<|reserved_special_token_145|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128151": {
+      "content": "<|reserved_special_token_146|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128152": {
+      "content": "<|reserved_special_token_147|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128153": {
+      "content": "<|reserved_special_token_148|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128154": {
+      "content": "<|reserved_special_token_149|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128155": {
+      "content": "<|reserved_special_token_150|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128156": {
+      "content": "<|reserved_special_token_151|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128157": {
+      "content": "<|reserved_special_token_152|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128158": {
+      "content": "<|reserved_special_token_153|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128159": {
+      "content": "<|reserved_special_token_154|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128160": {
+      "content": "<|reserved_special_token_155|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128161": {
+      "content": "<|reserved_special_token_156|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128162": {
+      "content": "<|reserved_special_token_157|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128163": {
+      "content": "<|reserved_special_token_158|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128164": {
+      "content": "<|reserved_special_token_159|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128165": {
+      "content": "<|reserved_special_token_160|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128166": {
+      "content": "<|reserved_special_token_161|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128167": {
+      "content": "<|reserved_special_token_162|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128168": {
+      "content": "<|reserved_special_token_163|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128169": {
+      "content": "<|reserved_special_token_164|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128170": {
+      "content": "<|reserved_special_token_165|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128171": {
+      "content": "<|reserved_special_token_166|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128172": {
+      "content": "<|reserved_special_token_167|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128173": {
+      "content": "<|reserved_special_token_168|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128174": {
+      "content": "<|reserved_special_token_169|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128175": {
+      "content": "<|reserved_special_token_170|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128176": {
+      "content": "<|reserved_special_token_171|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128177": {
+      "content": "<|reserved_special_token_172|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128178": {
+      "content": "<|reserved_special_token_173|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128179": {
+      "content": "<|reserved_special_token_174|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128180": {
+      "content": "<|reserved_special_token_175|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128181": {
+      "content": "<|reserved_special_token_176|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128182": {
+      "content": "<|reserved_special_token_177|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128183": {
+      "content": "<|reserved_special_token_178|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128184": {
+      "content": "<|reserved_special_token_179|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128185": {
+      "content": "<|reserved_special_token_180|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128186": {
+      "content": "<|reserved_special_token_181|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128187": {
+      "content": "<|reserved_special_token_182|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128188": {
+      "content": "<|reserved_special_token_183|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128189": {
+      "content": "<|reserved_special_token_184|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128190": {
+      "content": "<|reserved_special_token_185|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128191": {
+      "content": "<|reserved_special_token_186|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128192": {
+      "content": "<|reserved_special_token_187|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128193": {
+      "content": "<|reserved_special_token_188|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128194": {
+      "content": "<|reserved_special_token_189|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128195": {
+      "content": "<|reserved_special_token_190|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128196": {
+      "content": "<|reserved_special_token_191|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128197": {
+      "content": "<|reserved_special_token_192|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128198": {
+      "content": "<|reserved_special_token_193|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128199": {
+      "content": "<|reserved_special_token_194|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128200": {
+      "content": "<|reserved_special_token_195|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128201": {
+      "content": "<|reserved_special_token_196|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128202": {
+      "content": "<|reserved_special_token_197|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128203": {
+      "content": "<|reserved_special_token_198|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128204": {
+      "content": "<|reserved_special_token_199|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128205": {
+      "content": "<|reserved_special_token_200|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128206": {
+      "content": "<|reserved_special_token_201|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128207": {
+      "content": "<|reserved_special_token_202|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128208": {
+      "content": "<|reserved_special_token_203|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128209": {
+      "content": "<|reserved_special_token_204|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128210": {
+      "content": "<|reserved_special_token_205|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128211": {
+      "content": "<|reserved_special_token_206|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128212": {
+      "content": "<|reserved_special_token_207|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128213": {
+      "content": "<|reserved_special_token_208|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128214": {
+      "content": "<|reserved_special_token_209|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128215": {
+      "content": "<|reserved_special_token_210|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128216": {
+      "content": "<|reserved_special_token_211|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128217": {
+      "content": "<|reserved_special_token_212|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128218": {
+      "content": "<|reserved_special_token_213|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128219": {
+      "content": "<|reserved_special_token_214|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128220": {
+      "content": "<|reserved_special_token_215|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128221": {
+      "content": "<|reserved_special_token_216|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128222": {
+      "content": "<|reserved_special_token_217|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128223": {
+      "content": "<|reserved_special_token_218|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128224": {
+      "content": "<|reserved_special_token_219|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128225": {
+      "content": "<|reserved_special_token_220|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128226": {
+      "content": "<|reserved_special_token_221|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128227": {
+      "content": "<|reserved_special_token_222|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128228": {
+      "content": "<|reserved_special_token_223|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128229": {
+      "content": "<|reserved_special_token_224|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128230": {
+      "content": "<|reserved_special_token_225|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128231": {
+      "content": "<|reserved_special_token_226|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128232": {
+      "content": "<|reserved_special_token_227|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128233": {
+      "content": "<|reserved_special_token_228|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128234": {
+      "content": "<|reserved_special_token_229|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128235": {
+      "content": "<|reserved_special_token_230|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128236": {
+      "content": "<|reserved_special_token_231|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128237": {
+      "content": "<|reserved_special_token_232|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128238": {
+      "content": "<|reserved_special_token_233|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128239": {
+      "content": "<|reserved_special_token_234|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128240": {
+      "content": "<|reserved_special_token_235|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128241": {
+      "content": "<|reserved_special_token_236|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128242": {
+      "content": "<|reserved_special_token_237|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128243": {
+      "content": "<|reserved_special_token_238|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128244": {
+      "content": "<|reserved_special_token_239|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128245": {
+      "content": "<|reserved_special_token_240|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128246": {
+      "content": "<|reserved_special_token_241|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128247": {
+      "content": "<|reserved_special_token_242|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128248": {
+      "content": "<|reserved_special_token_243|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128249": {
+      "content": "<|reserved_special_token_244|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128250": {
+      "content": "<|reserved_special_token_245|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128251": {
+      "content": "<|reserved_special_token_246|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128252": {
+      "content": "<|reserved_special_token_247|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128253": {
+      "content": "<|reserved_special_token_248|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128254": {
+      "content": "<|reserved_special_token_249|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "128255": {
+      "content": "<|reserved_special_token_250|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|begin_of_text|>",
+  "chat_template": "{% set loop_messages = messages %}{% for message in loop_messages %}{% set content = '<|start_header_id|>' + message['role'] + '<|end_header_id|>\n\n'+ message['content'] | trim + '<|eot_id|>' %}{% if loop.index0 == 0 %}{% set content = bos_token + content %}{% endif %}{{ content }}{% endfor %}{% if add_generation_prompt %}{{ '<|start_header_id|>assistant<|end_header_id|>\n\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|eot_id|>",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|eot_id|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "PreTrainedTokenizerFast"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 31.91571553994732,
+    "total_flos": 9.547395325030564e+17,
+    "train_loss": 0.8426483250682203,
+    "train_runtime": 5512.317,
+    "train_samples_per_second": 13.218,
+    "train_steps_per_second": 0.824
+}

trainer_log.jsonl

@@ -0,0 +1,457 @@
+{"current_steps": 10, "total_steps": 4544, "loss": 2.4857, "lr": 2.197802197802198e-07, "epoch": 0.07023705004389816, "percentage": 0.22, "elapsed_time": "0:00:17", "remaining_time": "2:10:53"}
+{"current_steps": 20, "total_steps": 4544, "loss": 2.5113, "lr": 4.395604395604396e-07, "epoch": 0.14047410008779632, "percentage": 0.44, "elapsed_time": "0:00:33", "remaining_time": "2:06:55"}
+{"current_steps": 30, "total_steps": 4544, "loss": 2.4922, "lr": 6.593406593406594e-07, "epoch": 0.21071115013169447, "percentage": 0.66, "elapsed_time": "0:00:50", "remaining_time": "2:05:31"}
+{"current_steps": 40, "total_steps": 4544, "loss": 2.4394, "lr": 8.791208791208792e-07, "epoch": 0.28094820017559263, "percentage": 0.88, "elapsed_time": "0:01:06", "remaining_time": "2:04:32"}
+{"current_steps": 50, "total_steps": 4544, "loss": 2.4773, "lr": 1.098901098901099e-06, "epoch": 0.35118525021949076, "percentage": 1.1, "elapsed_time": "0:01:28", "remaining_time": "2:12:03"}
+{"current_steps": 60, "total_steps": 4544, "loss": 2.3524, "lr": 1.3186813186813187e-06, "epoch": 0.42142230026338895, "percentage": 1.32, "elapsed_time": "0:01:58", "remaining_time": "2:27:51"}
+{"current_steps": 70, "total_steps": 4544, "loss": 2.42, "lr": 1.5384615384615387e-06, "epoch": 0.4916593503072871, "percentage": 1.54, "elapsed_time": "0:02:17", "remaining_time": "2:26:54"}
+{"current_steps": 80, "total_steps": 4544, "loss": 2.3288, "lr": 1.7582417582417585e-06, "epoch": 0.5618964003511853, "percentage": 1.76, "elapsed_time": "0:02:35", "remaining_time": "2:24:25"}
+{"current_steps": 90, "total_steps": 4544, "loss": 2.2519, "lr": 1.9780219780219782e-06, "epoch": 0.6321334503950834, "percentage": 1.98, "elapsed_time": "0:02:53", "remaining_time": "2:22:48"}
+{"current_steps": 100, "total_steps": 4544, "loss": 2.2021, "lr": 2.197802197802198e-06, "epoch": 0.7023705004389815, "percentage": 2.2, "elapsed_time": "0:03:10", "remaining_time": "2:21:08"}
+{"current_steps": 110, "total_steps": 4544, "loss": 2.2151, "lr": 2.4175824175824177e-06, "epoch": 0.7726075504828798, "percentage": 2.42, "elapsed_time": "0:03:28", "remaining_time": "2:19:59"}
+{"current_steps": 120, "total_steps": 4544, "loss": 2.1102, "lr": 2.6373626373626375e-06, "epoch": 0.8428446005267779, "percentage": 2.64, "elapsed_time": "0:03:47", "remaining_time": "2:19:42"}
+{"current_steps": 130, "total_steps": 4544, "loss": 2.1034, "lr": 2.8571428571428573e-06, "epoch": 0.913081650570676, "percentage": 2.86, "elapsed_time": "0:04:04", "remaining_time": "2:18:27"}
+{"current_steps": 140, "total_steps": 4544, "loss": 2.0737, "lr": 3.0769230769230774e-06, "epoch": 0.9833187006145742, "percentage": 3.08, "elapsed_time": "0:04:21", "remaining_time": "2:17:01"}
+{"current_steps": 150, "total_steps": 4544, "loss": 2.0345, "lr": 3.2967032967032968e-06, "epoch": 1.0535557506584723, "percentage": 3.3, "elapsed_time": "0:04:38", "remaining_time": "2:15:52"}
+{"current_steps": 160, "total_steps": 4544, "loss": 1.9568, "lr": 3.516483516483517e-06, "epoch": 1.1237928007023705, "percentage": 3.52, "elapsed_time": "0:04:54", "remaining_time": "2:14:30"}
+{"current_steps": 170, "total_steps": 4544, "loss": 1.9494, "lr": 3.7362637362637367e-06, "epoch": 1.1940298507462686, "percentage": 3.74, "elapsed_time": "0:05:09", "remaining_time": "2:12:53"}
+{"current_steps": 180, "total_steps": 4544, "loss": 1.8932, "lr": 3.9560439560439565e-06, "epoch": 1.2642669007901668, "percentage": 3.96, "elapsed_time": "0:05:25", "remaining_time": "2:11:30"}
+{"current_steps": 190, "total_steps": 4544, "loss": 1.9219, "lr": 4.175824175824177e-06, "epoch": 1.334503950834065, "percentage": 4.18, "elapsed_time": "0:05:40", "remaining_time": "2:10:11"}
+{"current_steps": 200, "total_steps": 4544, "loss": 1.9751, "lr": 4.395604395604396e-06, "epoch": 1.404741000877963, "percentage": 4.4, "elapsed_time": "0:05:56", "remaining_time": "2:09:05"}
+{"current_steps": 210, "total_steps": 4544, "loss": 1.8729, "lr": 4.615384615384616e-06, "epoch": 1.4749780509218613, "percentage": 4.62, "elapsed_time": "0:06:13", "remaining_time": "2:08:26"}
+{"current_steps": 220, "total_steps": 4544, "loss": 1.885, "lr": 4.8351648351648355e-06, "epoch": 1.5452151009657595, "percentage": 4.84, "elapsed_time": "0:06:28", "remaining_time": "2:07:14"}
+{"current_steps": 230, "total_steps": 4544, "loss": 1.8549, "lr": 5.054945054945055e-06, "epoch": 1.6154521510096576, "percentage": 5.06, "elapsed_time": "0:06:44", "remaining_time": "2:06:26"}
+{"current_steps": 240, "total_steps": 4544, "loss": 1.8736, "lr": 5.274725274725275e-06, "epoch": 1.6856892010535558, "percentage": 5.28, "elapsed_time": "0:06:59", "remaining_time": "2:05:22"}
+{"current_steps": 250, "total_steps": 4544, "loss": 1.8388, "lr": 5.494505494505495e-06, "epoch": 1.755926251097454, "percentage": 5.5, "elapsed_time": "0:07:15", "remaining_time": "2:04:39"}
+{"current_steps": 260, "total_steps": 4544, "loss": 1.8676, "lr": 5.7142857142857145e-06, "epoch": 1.826163301141352, "percentage": 5.72, "elapsed_time": "0:07:30", "remaining_time": "2:03:40"}
+{"current_steps": 270, "total_steps": 4544, "loss": 1.8181, "lr": 5.934065934065935e-06, "epoch": 1.89640035118525, "percentage": 5.94, "elapsed_time": "0:07:45", "remaining_time": "2:02:46"}
+{"current_steps": 280, "total_steps": 4544, "loss": 1.8007, "lr": 6.153846153846155e-06, "epoch": 1.9666374012291485, "percentage": 6.16, "elapsed_time": "0:08:00", "remaining_time": "2:01:53"}
+{"current_steps": 290, "total_steps": 4544, "loss": 1.7727, "lr": 6.373626373626373e-06, "epoch": 2.0368744512730466, "percentage": 6.38, "elapsed_time": "0:08:16", "remaining_time": "2:01:17"}
+{"current_steps": 300, "total_steps": 4544, "loss": 1.7038, "lr": 6.5934065934065935e-06, "epoch": 2.1071115013169446, "percentage": 6.6, "elapsed_time": "0:08:31", "remaining_time": "2:00:37"}
+{"current_steps": 310, "total_steps": 4544, "loss": 1.7398, "lr": 6.813186813186814e-06, "epoch": 2.177348551360843, "percentage": 6.82, "elapsed_time": "0:08:46", "remaining_time": "1:59:45"}
+{"current_steps": 320, "total_steps": 4544, "loss": 1.7468, "lr": 7.032967032967034e-06, "epoch": 2.247585601404741, "percentage": 7.04, "elapsed_time": "0:09:01", "remaining_time": "1:59:07"}
+{"current_steps": 330, "total_steps": 4544, "loss": 1.7326, "lr": 7.252747252747253e-06, "epoch": 2.317822651448639, "percentage": 7.26, "elapsed_time": "0:09:16", "remaining_time": "1:58:20"}
+{"current_steps": 340, "total_steps": 4544, "loss": 1.7779, "lr": 7.472527472527473e-06, "epoch": 2.388059701492537, "percentage": 7.48, "elapsed_time": "0:09:30", "remaining_time": "1:57:35"}
+{"current_steps": 350, "total_steps": 4544, "loss": 1.6381, "lr": 7.692307692307694e-06, "epoch": 2.4582967515364356, "percentage": 7.7, "elapsed_time": "0:09:46", "remaining_time": "1:57:07"}
+{"current_steps": 360, "total_steps": 4544, "loss": 1.6605, "lr": 7.912087912087913e-06, "epoch": 2.5285338015803336, "percentage": 7.92, "elapsed_time": "0:10:00", "remaining_time": "1:56:19"}
+{"current_steps": 370, "total_steps": 4544, "loss": 1.6683, "lr": 8.131868131868132e-06, "epoch": 2.5987708516242316, "percentage": 8.14, "elapsed_time": "0:10:14", "remaining_time": "1:55:33"}
+{"current_steps": 380, "total_steps": 4544, "loss": 1.6819, "lr": 8.351648351648353e-06, "epoch": 2.66900790166813, "percentage": 8.36, "elapsed_time": "0:10:29", "remaining_time": "1:54:52"}
+{"current_steps": 390, "total_steps": 4544, "loss": 1.7072, "lr": 8.571428571428571e-06, "epoch": 2.739244951712028, "percentage": 8.58, "elapsed_time": "0:10:43", "remaining_time": "1:54:16"}
+{"current_steps": 400, "total_steps": 4544, "loss": 1.6747, "lr": 8.791208791208792e-06, "epoch": 2.809482001755926, "percentage": 8.8, "elapsed_time": "0:10:58", "remaining_time": "1:53:41"}
+{"current_steps": 410, "total_steps": 4544, "loss": 1.6596, "lr": 9.010989010989011e-06, "epoch": 2.8797190517998246, "percentage": 9.02, "elapsed_time": "0:11:12", "remaining_time": "1:53:01"}
+{"current_steps": 420, "total_steps": 4544, "loss": 1.65, "lr": 9.230769230769232e-06, "epoch": 2.9499561018437226, "percentage": 9.24, "elapsed_time": "0:11:26", "remaining_time": "1:52:19"}
+{"current_steps": 430, "total_steps": 4544, "loss": 1.6037, "lr": 9.450549450549452e-06, "epoch": 3.0201931518876206, "percentage": 9.46, "elapsed_time": "0:11:40", "remaining_time": "1:51:45"}
+{"current_steps": 440, "total_steps": 4544, "loss": 1.6015, "lr": 9.670329670329671e-06, "epoch": 3.090430201931519, "percentage": 9.68, "elapsed_time": "0:11:54", "remaining_time": "1:51:08"}
+{"current_steps": 450, "total_steps": 4544, "loss": 1.5708, "lr": 9.890109890109892e-06, "epoch": 3.160667251975417, "percentage": 9.9, "elapsed_time": "0:12:09", "remaining_time": "1:50:36"}
+{"current_steps": 460, "total_steps": 4544, "loss": 1.5778, "lr": 9.99996310691202e-06, "epoch": 3.230904302019315, "percentage": 10.12, "elapsed_time": "0:12:22", "remaining_time": "1:49:56"}
+{"current_steps": 470, "total_steps": 4544, "loss": 1.5528, "lr": 9.999667965474806e-06, "epoch": 3.3011413520632136, "percentage": 10.34, "elapsed_time": "0:12:36", "remaining_time": "1:49:17"}
+{"current_steps": 480, "total_steps": 4544, "loss": 1.5539, "lr": 9.9990777000222e-06, "epoch": 3.3713784021071116, "percentage": 10.56, "elapsed_time": "0:12:49", "remaining_time": "1:48:38"}
+{"current_steps": 490, "total_steps": 4544, "loss": 1.5289, "lr": 9.998192345396817e-06, "epoch": 3.4416154521510096, "percentage": 10.78, "elapsed_time": "0:13:04", "remaining_time": "1:48:09"}
+{"current_steps": 500, "total_steps": 4544, "loss": 1.5335, "lr": 9.997011953860014e-06, "epoch": 3.511852502194908, "percentage": 11.0, "elapsed_time": "0:13:18", "remaining_time": "1:47:35"}
+{"current_steps": 510, "total_steps": 4544, "loss": 1.5565, "lr": 9.99553659508879e-06, "epoch": 3.582089552238806, "percentage": 11.22, "elapsed_time": "0:13:31", "remaining_time": "1:46:58"}
+{"current_steps": 520, "total_steps": 4544, "loss": 1.5414, "lr": 9.993766356171694e-06, "epoch": 3.652326602282704, "percentage": 11.44, "elapsed_time": "0:13:44", "remaining_time": "1:46:22"}
+{"current_steps": 530, "total_steps": 4544, "loss": 1.5271, "lr": 9.991701341603667e-06, "epoch": 3.722563652326602, "percentage": 11.66, "elapsed_time": "0:13:58", "remaining_time": "1:45:49"}
+{"current_steps": 540, "total_steps": 4544, "loss": 1.4782, "lr": 9.989341673279881e-06, "epoch": 3.7928007023705006, "percentage": 11.88, "elapsed_time": "0:14:11", "remaining_time": "1:45:14"}
+{"current_steps": 550, "total_steps": 4544, "loss": 1.5165, "lr": 9.986687490488545e-06, "epoch": 3.8630377524143986, "percentage": 12.1, "elapsed_time": "0:14:25", "remaining_time": "1:44:43"}
+{"current_steps": 560, "total_steps": 4544, "loss": 1.507, "lr": 9.98373894990268e-06, "epoch": 3.9332748024582966, "percentage": 12.32, "elapsed_time": "0:14:38", "remaining_time": "1:44:07"}
+{"current_steps": 570, "total_steps": 4544, "loss": 1.5207, "lr": 9.980496225570869e-06, "epoch": 4.003511852502195, "percentage": 12.54, "elapsed_time": "0:14:51", "remaining_time": "1:43:36"}
+{"current_steps": 580, "total_steps": 4544, "loss": 1.4324, "lr": 9.976959508906992e-06, "epoch": 4.073748902546093, "percentage": 12.76, "elapsed_time": "0:15:04", "remaining_time": "1:43:01"}
+{"current_steps": 590, "total_steps": 4544, "loss": 1.4171, "lr": 9.973129008678915e-06, "epoch": 4.143985952589992, "percentage": 12.98, "elapsed_time": "0:15:17", "remaining_time": "1:42:26"}
+{"current_steps": 600, "total_steps": 4544, "loss": 1.4378, "lr": 9.969004950996175e-06, "epoch": 4.214223002633889, "percentage": 13.2, "elapsed_time": "0:15:30", "remaining_time": "1:41:58"}
+{"current_steps": 610, "total_steps": 4544, "loss": 1.3886, "lr": 9.964587579296631e-06, "epoch": 4.284460052677788, "percentage": 13.42, "elapsed_time": "0:15:43", "remaining_time": "1:41:24"}
+{"current_steps": 620, "total_steps": 4544, "loss": 1.4154, "lr": 9.959877154332095e-06, "epoch": 4.354697102721686, "percentage": 13.64, "elapsed_time": "0:15:55", "remaining_time": "1:40:50"}
+{"current_steps": 630, "total_steps": 4544, "loss": 1.4231, "lr": 9.954873954152933e-06, "epoch": 4.424934152765584, "percentage": 13.86, "elapsed_time": "0:16:09", "remaining_time": "1:40:21"}
+{"current_steps": 640, "total_steps": 4544, "loss": 1.3745, "lr": 9.949578274091666e-06, "epoch": 4.495171202809482, "percentage": 14.08, "elapsed_time": "0:16:21", "remaining_time": "1:39:49"}
+{"current_steps": 650, "total_steps": 4544, "loss": 1.3663, "lr": 9.943990426745525e-06, "epoch": 4.565408252853381, "percentage": 14.3, "elapsed_time": "0:16:35", "remaining_time": "1:39:21"}
+{"current_steps": 660, "total_steps": 4544, "loss": 1.4177, "lr": 9.938110741958003e-06, "epoch": 4.635645302897278, "percentage": 14.52, "elapsed_time": "0:16:47", "remaining_time": "1:38:48"}
+{"current_steps": 670, "total_steps": 4544, "loss": 1.412, "lr": 9.931939566799385e-06, "epoch": 4.705882352941177, "percentage": 14.74, "elapsed_time": "0:16:59", "remaining_time": "1:38:16"}
+{"current_steps": 680, "total_steps": 4544, "loss": 1.4155, "lr": 9.925477265546258e-06, "epoch": 4.776119402985074, "percentage": 14.96, "elapsed_time": "0:17:11", "remaining_time": "1:37:43"}
+{"current_steps": 690, "total_steps": 4544, "loss": 1.4241, "lr": 9.918724219660013e-06, "epoch": 4.846356453028973, "percentage": 15.18, "elapsed_time": "0:17:23", "remaining_time": "1:37:10"}
+{"current_steps": 700, "total_steps": 4544, "loss": 1.3562, "lr": 9.911680827764329e-06, "epoch": 4.916593503072871, "percentage": 15.4, "elapsed_time": "0:17:36", "remaining_time": "1:36:42"}
+{"current_steps": 710, "total_steps": 4544, "loss": 1.3542, "lr": 9.90434750562163e-06, "epoch": 4.98683055311677, "percentage": 15.62, "elapsed_time": "0:17:48", "remaining_time": "1:36:09"}
+{"current_steps": 720, "total_steps": 4544, "loss": 1.305, "lr": 9.896724686108561e-06, "epoch": 5.057067603160667, "percentage": 15.85, "elapsed_time": "0:18:01", "remaining_time": "1:35:41"}
+{"current_steps": 730, "total_steps": 4544, "loss": 1.314, "lr": 9.888812819190419e-06, "epoch": 5.127304653204566, "percentage": 16.07, "elapsed_time": "0:18:13", "remaining_time": "1:35:11"}
+{"current_steps": 740, "total_steps": 4544, "loss": 1.2962, "lr": 9.88061237189461e-06, "epoch": 5.197541703248463, "percentage": 16.29, "elapsed_time": "0:18:25", "remaining_time": "1:34:44"}
+{"current_steps": 750, "total_steps": 4544, "loss": 1.2993, "lr": 9.872123828283063e-06, "epoch": 5.267778753292362, "percentage": 16.51, "elapsed_time": "0:18:37", "remaining_time": "1:34:15"}
+{"current_steps": 760, "total_steps": 4544, "loss": 1.2649, "lr": 9.863347689423666e-06, "epoch": 5.33801580333626, "percentage": 16.73, "elapsed_time": "0:18:50", "remaining_time": "1:33:46"}
+{"current_steps": 770, "total_steps": 4544, "loss": 1.2943, "lr": 9.854284473360694e-06, "epoch": 5.408252853380158, "percentage": 16.95, "elapsed_time": "0:19:01", "remaining_time": "1:33:14"}
+{"current_steps": 780, "total_steps": 4544, "loss": 1.2512, "lr": 9.84493471508421e-06, "epoch": 5.478489903424056, "percentage": 17.17, "elapsed_time": "0:19:12", "remaining_time": "1:32:43"}
+{"current_steps": 790, "total_steps": 4544, "loss": 1.3046, "lr": 9.835298966498511e-06, "epoch": 5.548726953467955, "percentage": 17.39, "elapsed_time": "0:19:24", "remaining_time": "1:32:13"}
+{"current_steps": 800, "total_steps": 4544, "loss": 1.2338, "lr": 9.82537779638953e-06, "epoch": 5.618964003511852, "percentage": 17.61, "elapsed_time": "0:19:36", "remaining_time": "1:31:44"}
+{"current_steps": 810, "total_steps": 4544, "loss": 1.2921, "lr": 9.815171790391269e-06, "epoch": 5.689201053555751, "percentage": 17.83, "elapsed_time": "0:19:47", "remaining_time": "1:31:15"}
+{"current_steps": 820, "total_steps": 4544, "loss": 1.2306, "lr": 9.804681550951228e-06, "epoch": 5.759438103599649, "percentage": 18.05, "elapsed_time": "0:19:59", "remaining_time": "1:30:45"}
+{"current_steps": 830, "total_steps": 4544, "loss": 1.2304, "lr": 9.793907697294844e-06, "epoch": 5.829675153643547, "percentage": 18.27, "elapsed_time": "0:20:10", "remaining_time": "1:30:14"}
+{"current_steps": 840, "total_steps": 4544, "loss": 1.2693, "lr": 9.782850865388941e-06, "epoch": 5.899912203687445, "percentage": 18.49, "elapsed_time": "0:20:21", "remaining_time": "1:29:46"}
+{"current_steps": 850, "total_steps": 4544, "loss": 1.2642, "lr": 9.77151170790419e-06, "epoch": 5.970149253731344, "percentage": 18.71, "elapsed_time": "0:20:33", "remaining_time": "1:29:19"}
+{"current_steps": 860, "total_steps": 4544, "loss": 1.2702, "lr": 9.759890894176574e-06, "epoch": 6.040386303775241, "percentage": 18.93, "elapsed_time": "0:20:44", "remaining_time": "1:28:52"}
+{"current_steps": 870, "total_steps": 4544, "loss": 1.2085, "lr": 9.747989110167887e-06, "epoch": 6.11062335381914, "percentage": 19.15, "elapsed_time": "0:20:55", "remaining_time": "1:28:23"}
+{"current_steps": 880, "total_steps": 4544, "loss": 1.1754, "lr": 9.735807058425241e-06, "epoch": 6.180860403863038, "percentage": 19.37, "elapsed_time": "0:21:06", "remaining_time": "1:27:54"}
+{"current_steps": 890, "total_steps": 4544, "loss": 1.1378, "lr": 9.723345458039595e-06, "epoch": 6.251097453906936, "percentage": 19.59, "elapsed_time": "0:21:18", "remaining_time": "1:27:28"}
+{"current_steps": 900, "total_steps": 4544, "loss": 1.1588, "lr": 9.710605044603305e-06, "epoch": 6.321334503950834, "percentage": 19.81, "elapsed_time": "0:21:29", "remaining_time": "1:27:00"}
+{"current_steps": 910, "total_steps": 4544, "loss": 1.1934, "lr": 9.697586570166707e-06, "epoch": 6.391571553994732, "percentage": 20.03, "elapsed_time": "0:21:40", "remaining_time": "1:26:32"}
+{"current_steps": 920, "total_steps": 4544, "loss": 1.1504, "lr": 9.684290803193721e-06, "epoch": 6.46180860403863, "percentage": 20.25, "elapsed_time": "0:21:50", "remaining_time": "1:26:03"}
+{"current_steps": 930, "total_steps": 4544, "loss": 1.1152, "lr": 9.670718528516495e-06, "epoch": 6.532045654082529, "percentage": 20.47, "elapsed_time": "0:22:01", "remaining_time": "1:25:35"}
+{"current_steps": 940, "total_steps": 4544, "loss": 1.1628, "lr": 9.65687054728907e-06, "epoch": 6.602282704126427, "percentage": 20.69, "elapsed_time": "0:22:12", "remaining_time": "1:25:07"}
+{"current_steps": 950, "total_steps": 4544, "loss": 1.1299, "lr": 9.642747676940094e-06, "epoch": 6.672519754170325, "percentage": 20.91, "elapsed_time": "0:22:23", "remaining_time": "1:24:41"}
+{"current_steps": 960, "total_steps": 4544, "loss": 1.1635, "lr": 9.62835075112457e-06, "epoch": 6.742756804214223, "percentage": 21.13, "elapsed_time": "0:22:33", "remaining_time": "1:24:12"}
+{"current_steps": 970, "total_steps": 4544, "loss": 1.1673, "lr": 9.61368061967464e-06, "epoch": 6.812993854258121, "percentage": 21.35, "elapsed_time": "0:22:43", "remaining_time": "1:23:43"}
+{"current_steps": 980, "total_steps": 4544, "loss": 1.1281, "lr": 9.598738148549434e-06, "epoch": 6.883230904302019, "percentage": 21.57, "elapsed_time": "0:22:53", "remaining_time": "1:23:16"}
+{"current_steps": 990, "total_steps": 4544, "loss": 1.1617, "lr": 9.583524219783938e-06, "epoch": 6.953467954345918, "percentage": 21.79, "elapsed_time": "0:23:04", "remaining_time": "1:22:49"}
+{"current_steps": 1000, "total_steps": 4544, "loss": 1.1436, "lr": 9.56803973143694e-06, "epoch": 7.023705004389815, "percentage": 22.01, "elapsed_time": "0:23:15", "remaining_time": "1:22:24"}
+{"current_steps": 1010, "total_steps": 4544, "loss": 1.0962, "lr": 9.552285597538014e-06, "epoch": 7.093942054433714, "percentage": 22.23, "elapsed_time": "0:23:25", "remaining_time": "1:21:56"}
+{"current_steps": 1020, "total_steps": 4544, "loss": 1.0893, "lr": 9.536262748033564e-06, "epoch": 7.164179104477612, "percentage": 22.45, "elapsed_time": "0:23:34", "remaining_time": "1:21:28"}
+{"current_steps": 1030, "total_steps": 4544, "loss": 1.0618, "lr": 9.519972128731937e-06, "epoch": 7.23441615452151, "percentage": 22.67, "elapsed_time": "0:23:44", "remaining_time": "1:21:01"}
+{"current_steps": 1040, "total_steps": 4544, "loss": 1.0986, "lr": 9.503414701247587e-06, "epoch": 7.304653204565408, "percentage": 22.89, "elapsed_time": "0:23:54", "remaining_time": "1:20:34"}
+{"current_steps": 1050, "total_steps": 4544, "loss": 1.1748, "lr": 9.486591442944313e-06, "epoch": 7.374890254609307, "percentage": 23.11, "elapsed_time": "0:24:05", "remaining_time": "1:20:09"}
+{"current_steps": 1060, "total_steps": 4544, "loss": 1.1092, "lr": 9.469503346877569e-06, "epoch": 7.445127304653204, "percentage": 23.33, "elapsed_time": "0:24:14", "remaining_time": "1:19:42"}
+{"current_steps": 1070, "total_steps": 4544, "loss": 1.1099, "lr": 9.452151421735846e-06, "epoch": 7.515364354697103, "percentage": 23.55, "elapsed_time": "0:24:24", "remaining_time": "1:19:16"}
+{"current_steps": 1080, "total_steps": 4544, "loss": 1.0662, "lr": 9.434536691781125e-06, "epoch": 7.585601404741001, "percentage": 23.77, "elapsed_time": "0:24:34", "remaining_time": "1:18:50"}
+{"current_steps": 1090, "total_steps": 4544, "loss": 1.0551, "lr": 9.416660196788423e-06, "epoch": 7.655838454784899, "percentage": 23.99, "elapsed_time": "0:24:44", "remaining_time": "1:18:24"}
+{"current_steps": 1100, "total_steps": 4544, "loss": 1.0643, "lr": 9.39852299198441e-06, "epoch": 7.726075504828797, "percentage": 24.21, "elapsed_time": "0:24:54", "remaining_time": "1:17:58"}
+{"current_steps": 1110, "total_steps": 4544, "loss": 1.0576, "lr": 9.380126147985122e-06, "epoch": 7.796312554872696, "percentage": 24.43, "elapsed_time": "0:25:03", "remaining_time": "1:17:31"}
+{"current_steps": 1120, "total_steps": 4544, "loss": 1.0703, "lr": 9.36147075073277e-06, "epoch": 7.866549604916593, "percentage": 24.65, "elapsed_time": "0:25:12", "remaining_time": "1:17:04"}
+{"current_steps": 1130, "total_steps": 4544, "loss": 1.0488, "lr": 9.34255790143163e-06, "epoch": 7.936786654960492, "percentage": 24.87, "elapsed_time": "0:25:21", "remaining_time": "1:16:37"}
+{"current_steps": 1140, "total_steps": 4544, "loss": 1.0843, "lr": 9.323388716483046e-06, "epoch": 8.00702370500439, "percentage": 25.09, "elapsed_time": "0:25:31", "remaining_time": "1:16:14"}
+{"current_steps": 1150, "total_steps": 4544, "loss": 1.0359, "lr": 9.303964327419524e-06, "epoch": 8.077260755048288, "percentage": 25.31, "elapsed_time": "0:25:41", "remaining_time": "1:15:49"}
+{"current_steps": 1160, "total_steps": 4544, "loss": 1.0227, "lr": 9.284285880837947e-06, "epoch": 8.147497805092186, "percentage": 25.53, "elapsed_time": "0:25:50", "remaining_time": "1:15:23"}
+{"current_steps": 1170, "total_steps": 4544, "loss": 0.997, "lr": 9.264354538331886e-06, "epoch": 8.217734855136085, "percentage": 25.75, "elapsed_time": "0:25:59", "remaining_time": "1:14:57"}
+{"current_steps": 1180, "total_steps": 4544, "loss": 1.0194, "lr": 9.244171476423037e-06, "epoch": 8.287971905179983, "percentage": 25.97, "elapsed_time": "0:26:08", "remaining_time": "1:14:31"}
+{"current_steps": 1190, "total_steps": 4544, "loss": 1.0264, "lr": 9.223737886491771e-06, "epoch": 8.35820895522388, "percentage": 26.19, "elapsed_time": "0:26:17", "remaining_time": "1:14:06"}
+{"current_steps": 1200, "total_steps": 4544, "loss": 1.037, "lr": 9.203054974706807e-06, "epoch": 8.428446005267778, "percentage": 26.41, "elapsed_time": "0:26:26", "remaining_time": "1:13:42"}
+{"current_steps": 1210, "total_steps": 4544, "loss": 1.0202, "lr": 9.182123961954016e-06, "epoch": 8.498683055311677, "percentage": 26.63, "elapsed_time": "0:26:35", "remaining_time": "1:13:17"}
+{"current_steps": 1220, "total_steps": 4544, "loss": 1.0577, "lr": 9.160946083764353e-06, "epoch": 8.568920105355575, "percentage": 26.85, "elapsed_time": "0:26:44", "remaining_time": "1:12:52"}
+{"current_steps": 1230, "total_steps": 4544, "loss": 1.009, "lr": 9.13952259024092e-06, "epoch": 8.639157155399474, "percentage": 27.07, "elapsed_time": "0:26:53", "remaining_time": "1:12:27"}
+{"current_steps": 1240, "total_steps": 4544, "loss": 0.9763, "lr": 9.117854745985183e-06, "epoch": 8.709394205443372, "percentage": 27.29, "elapsed_time": "0:27:02", "remaining_time": "1:12:01"}
+{"current_steps": 1250, "total_steps": 4544, "loss": 1.0406, "lr": 9.095943830022323e-06, "epoch": 8.779631255487269, "percentage": 27.51, "elapsed_time": "0:27:10", "remaining_time": "1:11:37"}
+{"current_steps": 1260, "total_steps": 4544, "loss": 1.0093, "lr": 9.073791135725722e-06, "epoch": 8.849868305531167, "percentage": 27.73, "elapsed_time": "0:27:19", "remaining_time": "1:11:12"}
+{"current_steps": 1270, "total_steps": 4544, "loss": 1.0146, "lr": 9.051397970740638e-06, "epoch": 8.920105355575066, "percentage": 27.95, "elapsed_time": "0:27:27", "remaining_time": "1:10:47"}
+{"current_steps": 1280, "total_steps": 4544, "loss": 1.0404, "lr": 9.028765656907005e-06, "epoch": 8.990342405618964, "percentage": 28.17, "elapsed_time": "0:27:35", "remaining_time": "1:10:22"}
+{"current_steps": 1290, "total_steps": 4544, "loss": 0.9933, "lr": 9.005895530181406e-06, "epoch": 9.060579455662863, "percentage": 28.39, "elapsed_time": "0:27:44", "remaining_time": "1:09:57"}
+{"current_steps": 1300, "total_steps": 4544, "loss": 0.9701, "lr": 8.982788940558216e-06, "epoch": 9.130816505706761, "percentage": 28.61, "elapsed_time": "0:27:52", "remaining_time": "1:09:33"}
+{"current_steps": 1310, "total_steps": 4544, "loss": 0.9554, "lr": 8.959447251989914e-06, "epoch": 9.201053555750658, "percentage": 28.83, "elapsed_time": "0:28:00", "remaining_time": "1:09:08"}
+{"current_steps": 1320, "total_steps": 4544, "loss": 1.0002, "lr": 8.935871842306569e-06, "epoch": 9.271290605794556, "percentage": 29.05, "elapsed_time": "0:28:08", "remaining_time": "1:08:44"}
+{"current_steps": 1330, "total_steps": 4544, "loss": 0.999, "lr": 8.912064103134505e-06, "epoch": 9.341527655838455, "percentage": 29.27, "elapsed_time": "0:28:17", "remaining_time": "1:08:21"}
+{"current_steps": 1340, "total_steps": 4544, "loss": 0.9726, "lr": 8.888025439814169e-06, "epoch": 9.411764705882353, "percentage": 29.49, "elapsed_time": "0:28:25", "remaining_time": "1:07:56"}
+{"current_steps": 1350, "total_steps": 4544, "loss": 1.0067, "lr": 8.863757271317154e-06, "epoch": 9.482001755926252, "percentage": 29.71, "elapsed_time": "0:28:33", "remaining_time": "1:07:33"}
+{"current_steps": 1360, "total_steps": 4544, "loss": 1.0181, "lr": 8.839261030162459e-06, "epoch": 9.552238805970148, "percentage": 29.93, "elapsed_time": "0:28:41", "remaining_time": "1:07:09"}
+{"current_steps": 1370, "total_steps": 4544, "loss": 0.9688, "lr": 8.814538162331913e-06, "epoch": 9.622475856014047, "percentage": 30.15, "elapsed_time": "0:28:48", "remaining_time": "1:06:45"}
+{"current_steps": 1380, "total_steps": 4544, "loss": 0.9309, "lr": 8.789590127184837e-06, "epoch": 9.692712906057945, "percentage": 30.37, "elapsed_time": "0:28:56", "remaining_time": "1:06:21"}
+{"current_steps": 1390, "total_steps": 4544, "loss": 0.987, "lr": 8.764418397371888e-06, "epoch": 9.762949956101844, "percentage": 30.59, "elapsed_time": "0:29:04", "remaining_time": "1:05:57"}
+{"current_steps": 1400, "total_steps": 4544, "loss": 0.9292, "lr": 8.739024458748128e-06, "epoch": 9.833187006145742, "percentage": 30.81, "elapsed_time": "0:29:12", "remaining_time": "1:05:34"}
+{"current_steps": 1410, "total_steps": 4544, "loss": 0.9909, "lr": 8.713409810285327e-06, "epoch": 9.90342405618964, "percentage": 31.03, "elapsed_time": "0:29:19", "remaining_time": "1:05:10"}
+{"current_steps": 1420, "total_steps": 4544, "loss": 0.9914, "lr": 8.687575963983477e-06, "epoch": 9.973661106233537, "percentage": 31.25, "elapsed_time": "0:29:26", "remaining_time": "1:04:47"}
+{"current_steps": 1430, "total_steps": 4544, "loss": 0.9369, "lr": 8.661524444781531e-06, "epoch": 10.043898156277436, "percentage": 31.47, "elapsed_time": "0:29:35", "remaining_time": "1:04:25"}
+{"current_steps": 1440, "total_steps": 4544, "loss": 0.9573, "lr": 8.635256790467402e-06, "epoch": 10.114135206321334, "percentage": 31.69, "elapsed_time": "0:29:42", "remaining_time": "1:04:01"}
+{"current_steps": 1450, "total_steps": 4544, "loss": 0.9491, "lr": 8.60877455158718e-06, "epoch": 10.184372256365233, "percentage": 31.91, "elapsed_time": "0:29:49", "remaining_time": "1:03:39"}
+{"current_steps": 1460, "total_steps": 4544, "loss": 0.8985, "lr": 8.582079291353607e-06, "epoch": 10.254609306409131, "percentage": 32.13, "elapsed_time": "0:29:57", "remaining_time": "1:03:16"}
+{"current_steps": 1470, "total_steps": 4544, "loss": 0.913, "lr": 8.555172585553804e-06, "epoch": 10.32484635645303, "percentage": 32.35, "elapsed_time": "0:30:04", "remaining_time": "1:02:52"}
+{"current_steps": 1480, "total_steps": 4544, "loss": 0.9727, "lr": 8.528056022456256e-06, "epoch": 10.395083406496926, "percentage": 32.57, "elapsed_time": "0:30:11", "remaining_time": "1:02:29"}
+{"current_steps": 1490, "total_steps": 4544, "loss": 0.9723, "lr": 8.500731202717056e-06, "epoch": 10.465320456540825, "percentage": 32.79, "elapsed_time": "0:30:18", "remaining_time": "1:02:07"}
+{"current_steps": 1500, "total_steps": 4544, "loss": 0.9626, "lr": 8.473199739285416e-06, "epoch": 10.535557506584723, "percentage": 33.01, "elapsed_time": "0:30:25", "remaining_time": "1:01:45"}
+{"current_steps": 1510, "total_steps": 4544, "loss": 0.9487, "lr": 8.445463257308463e-06, "epoch": 10.605794556628622, "percentage": 33.23, "elapsed_time": "0:30:32", "remaining_time": "1:01:22"}
+{"current_steps": 1520, "total_steps": 4544, "loss": 0.9832, "lr": 8.417523394035316e-06, "epoch": 10.67603160667252, "percentage": 33.45, "elapsed_time": "0:30:39", "remaining_time": "1:00:59"}
+{"current_steps": 1530, "total_steps": 4544, "loss": 0.9621, "lr": 8.389381798720417e-06, "epoch": 10.746268656716419, "percentage": 33.67, "elapsed_time": "0:30:46", "remaining_time": "1:00:36"}
+{"current_steps": 1540, "total_steps": 4544, "loss": 0.9222, "lr": 8.361040132526204e-06, "epoch": 10.816505706760315, "percentage": 33.89, "elapsed_time": "0:30:52", "remaining_time": "1:00:14"}
+{"current_steps": 1550, "total_steps": 4544, "loss": 0.9183, "lr": 8.332500068425038e-06, "epoch": 10.886742756804214, "percentage": 34.11, "elapsed_time": "0:30:59", "remaining_time": "0:59:52"}
+{"current_steps": 1560, "total_steps": 4544, "loss": 0.9173, "lr": 8.303763291100459e-06, "epoch": 10.956979806848112, "percentage": 34.33, "elapsed_time": "0:31:06", "remaining_time": "0:59:30"}
+{"current_steps": 1570, "total_steps": 4544, "loss": 0.9582, "lr": 8.274831496847735e-06, "epoch": 11.02721685689201, "percentage": 34.55, "elapsed_time": "0:31:13", "remaining_time": "0:59:09"}
+{"current_steps": 1580, "total_steps": 4544, "loss": 0.907, "lr": 8.245706393473734e-06, "epoch": 11.09745390693591, "percentage": 34.77, "elapsed_time": "0:31:20", "remaining_time": "0:58:46"}
+{"current_steps": 1590, "total_steps": 4544, "loss": 0.9339, "lr": 8.216389700196116e-06, "epoch": 11.167690956979808, "percentage": 34.99, "elapsed_time": "0:31:26", "remaining_time": "0:58:24"}
+{"current_steps": 1600, "total_steps": 4544, "loss": 0.9417, "lr": 8.186883147541846e-06, "epoch": 11.237928007023704, "percentage": 35.21, "elapsed_time": "0:31:34", "remaining_time": "0:58:05"}
+{"current_steps": 1610, "total_steps": 4544, "loss": 0.8978, "lr": 8.157188477245048e-06, "epoch": 11.308165057067603, "percentage": 35.43, "elapsed_time": "0:31:50", "remaining_time": "0:58:02"}
+{"current_steps": 1620, "total_steps": 4544, "loss": 0.9335, "lr": 8.12730744214419e-06, "epoch": 11.378402107111501, "percentage": 35.65, "elapsed_time": "0:32:07", "remaining_time": "0:57:59"}
+{"current_steps": 1630, "total_steps": 4544, "loss": 0.9623, "lr": 8.097241806078616e-06, "epoch": 11.4486391571554, "percentage": 35.87, "elapsed_time": "0:32:24", "remaining_time": "0:57:55"}
+{"current_steps": 1640, "total_steps": 4544, "loss": 0.9691, "lr": 8.066993343784427e-06, "epoch": 11.518876207199298, "percentage": 36.09, "elapsed_time": "0:32:41", "remaining_time": "0:57:53"}
+{"current_steps": 1650, "total_steps": 4544, "loss": 0.9208, "lr": 8.036563840789726e-06, "epoch": 11.589113257243195, "percentage": 36.31, "elapsed_time": "0:32:58", "remaining_time": "0:57:49"}
+{"current_steps": 1660, "total_steps": 4544, "loss": 0.9352, "lr": 8.005955093309217e-06, "epoch": 11.659350307287093, "percentage": 36.53, "elapsed_time": "0:33:14", "remaining_time": "0:57:44"}
+{"current_steps": 1670, "total_steps": 4544, "loss": 0.911, "lr": 7.975168908138174e-06, "epoch": 11.729587357330992, "percentage": 36.75, "elapsed_time": "0:33:30", "remaining_time": "0:57:40"}
+{"current_steps": 1680, "total_steps": 4544, "loss": 0.9109, "lr": 7.944207102545795e-06, "epoch": 11.79982440737489, "percentage": 36.97, "elapsed_time": "0:33:46", "remaining_time": "0:57:35"}
+{"current_steps": 1690, "total_steps": 4544, "loss": 0.946, "lr": 7.913071504167925e-06, "epoch": 11.870061457418789, "percentage": 37.19, "elapsed_time": "0:34:02", "remaining_time": "0:57:29"}
+{"current_steps": 1700, "total_steps": 4544, "loss": 0.8947, "lr": 7.881763950899175e-06, "epoch": 11.940298507462687, "percentage": 37.41, "elapsed_time": "0:34:19", "remaining_time": "0:57:24"}
+{"current_steps": 1710, "total_steps": 4544, "loss": 0.8951, "lr": 7.850286290784437e-06, "epoch": 12.010535557506584, "percentage": 37.63, "elapsed_time": "0:34:35", "remaining_time": "0:57:19"}
+{"current_steps": 1720, "total_steps": 4544, "loss": 0.893, "lr": 7.81864038190979e-06, "epoch": 12.080772607550482, "percentage": 37.85, "elapsed_time": "0:34:50", "remaining_time": "0:57:13"}
+{"current_steps": 1730, "total_steps": 4544, "loss": 0.8729, "lr": 7.786828092292821e-06, "epoch": 12.151009657594381, "percentage": 38.07, "elapsed_time": "0:35:06", "remaining_time": "0:57:06"}
+{"current_steps": 1740, "total_steps": 4544, "loss": 0.8704, "lr": 7.754851299772362e-06, "epoch": 12.22124670763828, "percentage": 38.29, "elapsed_time": "0:35:22", "remaining_time": "0:57:00"}
+{"current_steps": 1750, "total_steps": 4544, "loss": 0.8639, "lr": 7.722711891897641e-06, "epoch": 12.291483757682178, "percentage": 38.51, "elapsed_time": "0:35:38", "remaining_time": "0:56:54"}
+{"current_steps": 1760, "total_steps": 4544, "loss": 0.8624, "lr": 7.690411765816864e-06, "epoch": 12.361720807726076, "percentage": 38.73, "elapsed_time": "0:35:54", "remaining_time": "0:56:47"}
+{"current_steps": 1770, "total_steps": 4544, "loss": 0.878, "lr": 7.657952828165225e-06, "epoch": 12.431957857769973, "percentage": 38.95, "elapsed_time": "0:36:10", "remaining_time": "0:56:41"}
+{"current_steps": 1780, "total_steps": 4544, "loss": 0.8555, "lr": 7.625336994952364e-06, "epoch": 12.502194907813871, "percentage": 39.17, "elapsed_time": "0:36:25", "remaining_time": "0:56:34"}
+{"current_steps": 1790, "total_steps": 4544, "loss": 0.8889, "lr": 7.592566191449262e-06, "epoch": 12.57243195785777, "percentage": 39.39, "elapsed_time": "0:36:41", "remaining_time": "0:56:26"}
+{"current_steps": 1800, "total_steps": 4544, "loss": 0.8598, "lr": 7.559642352074606e-06, "epoch": 12.642669007901668, "percentage": 39.61, "elapsed_time": "0:36:56", "remaining_time": "0:56:19"}
+{"current_steps": 1810, "total_steps": 4544, "loss": 0.8856, "lr": 7.526567420280585e-06, "epoch": 12.712906057945567, "percentage": 39.83, "elapsed_time": "0:37:12", "remaining_time": "0:56:12"}
+{"current_steps": 1820, "total_steps": 4544, "loss": 0.8677, "lr": 7.4933433484381905e-06, "epoch": 12.783143107989464, "percentage": 40.05, "elapsed_time": "0:37:27", "remaining_time": "0:56:03"}
+{"current_steps": 1830, "total_steps": 4544, "loss": 0.9036, "lr": 7.459972097721954e-06, "epoch": 12.853380158033362, "percentage": 40.27, "elapsed_time": "0:37:42", "remaining_time": "0:55:55"}
+{"current_steps": 1840, "total_steps": 4544, "loss": 0.8607, "lr": 7.4264556379941895e-06, "epoch": 12.92361720807726, "percentage": 40.49, "elapsed_time": "0:37:58", "remaining_time": "0:55:47"}
+{"current_steps": 1850, "total_steps": 4544, "loss": 0.8699, "lr": 7.392795947688715e-06, "epoch": 12.993854258121159, "percentage": 40.71, "elapsed_time": "0:38:13", "remaining_time": "0:55:40"}
+{"current_steps": 1860, "total_steps": 4544, "loss": 0.8343, "lr": 7.3589950136940645e-06, "epoch": 13.064091308165057, "percentage": 40.93, "elapsed_time": "0:38:28", "remaining_time": "0:55:31"}
+{"current_steps": 1870, "total_steps": 4544, "loss": 0.8305, "lr": 7.325054831236211e-06, "epoch": 13.134328358208956, "percentage": 41.15, "elapsed_time": "0:38:43", "remaining_time": "0:55:22"}
+{"current_steps": 1880, "total_steps": 4544, "loss": 0.8039, "lr": 7.2909774037607775e-06, "epoch": 13.204565408252853, "percentage": 41.37, "elapsed_time": "0:38:58", "remaining_time": "0:55:14"}
+{"current_steps": 1890, "total_steps": 4544, "loss": 0.8274, "lr": 7.256764742814796e-06, "epoch": 13.274802458296751, "percentage": 41.59, "elapsed_time": "0:39:13", "remaining_time": "0:55:05"}
+{"current_steps": 1900, "total_steps": 4544, "loss": 0.8126, "lr": 7.222418867927948e-06, "epoch": 13.34503950834065, "percentage": 41.81, "elapsed_time": "0:39:28", "remaining_time": "0:54:56"}
+{"current_steps": 1910, "total_steps": 4544, "loss": 0.8155, "lr": 7.187941806493372e-06, "epoch": 13.415276558384548, "percentage": 42.03, "elapsed_time": "0:39:43", "remaining_time": "0:54:47"}
+{"current_steps": 1920, "total_steps": 4544, "loss": 0.8346, "lr": 7.153335593647974e-06, "epoch": 13.485513608428446, "percentage": 42.25, "elapsed_time": "0:39:58", "remaining_time": "0:54:37"}
+{"current_steps": 1930, "total_steps": 4544, "loss": 0.8275, "lr": 7.118602272152308e-06, "epoch": 13.555750658472345, "percentage": 42.47, "elapsed_time": "0:40:12", "remaining_time": "0:54:28"}
+{"current_steps": 1940, "total_steps": 4544, "loss": 0.8215, "lr": 7.083743892269987e-06, "epoch": 13.625987708516242, "percentage": 42.69, "elapsed_time": "0:40:27", "remaining_time": "0:54:17"}
+{"current_steps": 1950, "total_steps": 4544, "loss": 0.8181, "lr": 7.04876251164666e-06, "epoch": 13.69622475856014, "percentage": 42.91, "elapsed_time": "0:40:41", "remaining_time": "0:54:08"}
+{"current_steps": 1960, "total_steps": 4544, "loss": 0.8099, "lr": 7.013660195188553e-06, "epoch": 13.766461808604038, "percentage": 43.13, "elapsed_time": "0:40:56", "remaining_time": "0:53:58"}
+{"current_steps": 1970, "total_steps": 4544, "loss": 0.8278, "lr": 6.978439014940584e-06, "epoch": 13.836698858647937, "percentage": 43.35, "elapsed_time": "0:41:10", "remaining_time": "0:53:48"}
+{"current_steps": 1980, "total_steps": 4544, "loss": 0.8299, "lr": 6.943101049964042e-06, "epoch": 13.906935908691835, "percentage": 43.57, "elapsed_time": "0:41:24", "remaining_time": "0:53:37"}
+{"current_steps": 1990, "total_steps": 4544, "loss": 0.8003, "lr": 6.907648386213875e-06, "epoch": 13.977172958735734, "percentage": 43.79, "elapsed_time": "0:41:38", "remaining_time": "0:53:27"}
+{"current_steps": 2000, "total_steps": 4544, "loss": 0.7669, "lr": 6.872083116415547e-06, "epoch": 14.04741000877963, "percentage": 44.01, "elapsed_time": "0:41:54", "remaining_time": "0:53:18"}
+{"current_steps": 2000, "total_steps": 4544, "eval_loss": 1.4163318872451782, "epoch": 14.04741000877963, "percentage": 44.01, "elapsed_time": "0:42:23", "remaining_time": "0:53:54"}
+{"current_steps": 2010, "total_steps": 4544, "loss": 0.7701, "lr": 6.836407339941522e-06, "epoch": 14.117647058823529, "percentage": 44.23, "elapsed_time": "0:42:37", "remaining_time": "0:53:43"}
+{"current_steps": 2020, "total_steps": 4544, "loss": 0.7634, "lr": 6.800623162687325e-06, "epoch": 14.187884108867427, "percentage": 44.45, "elapsed_time": "0:42:50", "remaining_time": "0:53:32"}
+{"current_steps": 2030, "total_steps": 4544, "loss": 0.7579, "lr": 6.764732696947243e-06, "epoch": 14.258121158911326, "percentage": 44.67, "elapsed_time": "0:43:04", "remaining_time": "0:53:21"}
+{"current_steps": 2040, "total_steps": 4544, "loss": 0.7713, "lr": 6.728738061289634e-06, "epoch": 14.328358208955224, "percentage": 44.89, "elapsed_time": "0:43:18", "remaining_time": "0:53:09"}
+{"current_steps": 2050, "total_steps": 4544, "loss": 0.7512, "lr": 6.692641380431879e-06, "epoch": 14.398595258999123, "percentage": 45.11, "elapsed_time": "0:43:33", "remaining_time": "0:52:58"}
+{"current_steps": 2060, "total_steps": 4544, "loss": 0.7673, "lr": 6.6564447851149505e-06, "epoch": 14.46883230904302, "percentage": 45.33, "elapsed_time": "0:43:46", "remaining_time": "0:52:47"}
+{"current_steps": 2070, "total_steps": 4544, "loss": 0.747, "lr": 6.620150411977648e-06, "epoch": 14.539069359086918, "percentage": 45.55, "elapsed_time": "0:44:00", "remaining_time": "0:52:35"}
+{"current_steps": 2080, "total_steps": 4544, "loss": 0.7471, "lr": 6.5837604034304715e-06, "epoch": 14.609306409130816, "percentage": 45.77, "elapsed_time": "0:44:13", "remaining_time": "0:52:23"}
+{"current_steps": 2090, "total_steps": 4544, "loss": 0.7733, "lr": 6.547276907529152e-06, "epoch": 14.679543459174715, "percentage": 45.99, "elapsed_time": "0:44:27", "remaining_time": "0:52:11"}
+{"current_steps": 2100, "total_steps": 4544, "loss": 0.7356, "lr": 6.510702077847864e-06, "epoch": 14.749780509218613, "percentage": 46.21, "elapsed_time": "0:44:41", "remaining_time": "0:52:00"}
+{"current_steps": 2110, "total_steps": 4544, "loss": 0.7702, "lr": 6.474038073352098e-06, "epoch": 14.82001755926251, "percentage": 46.43, "elapsed_time": "0:44:54", "remaining_time": "0:51:48"}
+{"current_steps": 2120, "total_steps": 4544, "loss": 0.76, "lr": 6.4372870582712196e-06, "epoch": 14.890254609306409, "percentage": 46.65, "elapsed_time": "0:45:07", "remaining_time": "0:51:36"}
+{"current_steps": 2130, "total_steps": 4544, "loss": 0.7664, "lr": 6.4004512019707144e-06, "epoch": 14.960491659350307, "percentage": 46.88, "elapsed_time": "0:45:21", "remaining_time": "0:51:24"}
+{"current_steps": 2140, "total_steps": 4544, "loss": 0.7274, "lr": 6.363532678824145e-06, "epoch": 15.030728709394205, "percentage": 47.1, "elapsed_time": "0:45:34", "remaining_time": "0:51:12"}
+{"current_steps": 2150, "total_steps": 4544, "loss": 0.703, "lr": 6.326533668084783e-06, "epoch": 15.100965759438104, "percentage": 47.32, "elapsed_time": "0:45:48", "remaining_time": "0:51:00"}
+{"current_steps": 2160, "total_steps": 4544, "loss": 0.7176, "lr": 6.289456353756988e-06, "epoch": 15.171202809482002, "percentage": 47.54, "elapsed_time": "0:46:01", "remaining_time": "0:50:47"}
+{"current_steps": 2170, "total_steps": 4544, "loss": 0.6589, "lr": 6.252302924467276e-06, "epoch": 15.241439859525899, "percentage": 47.76, "elapsed_time": "0:46:14", "remaining_time": "0:50:35"}
+{"current_steps": 2180, "total_steps": 4544, "loss": 0.6822, "lr": 6.2150755733351305e-06, "epoch": 15.311676909569798, "percentage": 47.98, "elapsed_time": "0:46:27", "remaining_time": "0:50:22"}
+{"current_steps": 2190, "total_steps": 4544, "loss": 0.6947, "lr": 6.177776497843552e-06, "epoch": 15.381913959613696, "percentage": 48.2, "elapsed_time": "0:46:40", "remaining_time": "0:50:10"}
+{"current_steps": 2200, "total_steps": 4544, "loss": 0.7027, "lr": 6.140407899709333e-06, "epoch": 15.452151009657594, "percentage": 48.42, "elapsed_time": "0:46:53", "remaining_time": "0:49:58"}
+{"current_steps": 2210, "total_steps": 4544, "loss": 0.6625, "lr": 6.102971984753104e-06, "epoch": 15.522388059701493, "percentage": 48.64, "elapsed_time": "0:47:06", "remaining_time": "0:49:44"}
+{"current_steps": 2220, "total_steps": 4544, "loss": 0.7181, "lr": 6.065470962769119e-06, "epoch": 15.592625109745391, "percentage": 48.86, "elapsed_time": "0:47:19", "remaining_time": "0:49:32"}
+{"current_steps": 2230, "total_steps": 4544, "loss": 0.6745, "lr": 6.027907047394812e-06, "epoch": 15.662862159789288, "percentage": 49.08, "elapsed_time": "0:47:31", "remaining_time": "0:49:19"}
+{"current_steps": 2240, "total_steps": 4544, "loss": 0.6744, "lr": 5.990282455980145e-06, "epoch": 15.733099209833187, "percentage": 49.3, "elapsed_time": "0:47:44", "remaining_time": "0:49:06"}
+{"current_steps": 2250, "total_steps": 4544, "loss": 0.6877, "lr": 5.952599409456697e-06, "epoch": 15.803336259877085, "percentage": 49.52, "elapsed_time": "0:47:56", "remaining_time": "0:48:53"}
+{"current_steps": 2260, "total_steps": 4544, "loss": 0.7303, "lr": 5.914860132206584e-06, "epoch": 15.873573309920983, "percentage": 49.74, "elapsed_time": "0:48:09", "remaining_time": "0:48:40"}
+{"current_steps": 2270, "total_steps": 4544, "loss": 0.712, "lr": 5.877066851931151e-06, "epoch": 15.943810359964882, "percentage": 49.96, "elapsed_time": "0:48:21", "remaining_time": "0:48:26"}
+{"current_steps": 2280, "total_steps": 4544, "loss": 0.6674, "lr": 5.83922179951947e-06, "epoch": 16.01404741000878, "percentage": 50.18, "elapsed_time": "0:48:34", "remaining_time": "0:48:14"}
+{"current_steps": 2290, "total_steps": 4544, "loss": 0.6726, "lr": 5.8013272089166526e-06, "epoch": 16.08428446005268, "percentage": 50.4, "elapsed_time": "0:48:46", "remaining_time": "0:48:00"}
+{"current_steps": 2300, "total_steps": 4544, "loss": 0.6198, "lr": 5.763385316991995e-06, "epoch": 16.154521510096576, "percentage": 50.62, "elapsed_time": "0:48:59", "remaining_time": "0:47:47"}
+{"current_steps": 2310, "total_steps": 4544, "loss": 0.6223, "lr": 5.725398363406922e-06, "epoch": 16.224758560140476, "percentage": 50.84, "elapsed_time": "0:49:11", "remaining_time": "0:47:34"}
+{"current_steps": 2320, "total_steps": 4544, "loss": 0.6315, "lr": 5.687368590482797e-06, "epoch": 16.294995610184372, "percentage": 51.06, "elapsed_time": "0:49:23", "remaining_time": "0:47:20"}
+{"current_steps": 2330, "total_steps": 4544, "loss": 0.6418, "lr": 5.64929824306855e-06, "epoch": 16.36523266022827, "percentage": 51.28, "elapsed_time": "0:49:35", "remaining_time": "0:47:07"}
+{"current_steps": 2340, "total_steps": 4544, "loss": 0.6086, "lr": 5.611189568408173e-06, "epoch": 16.43546971027217, "percentage": 51.5, "elapsed_time": "0:49:47", "remaining_time": "0:46:53"}
+{"current_steps": 2350, "total_steps": 4544, "loss": 0.6375, "lr": 5.573044816008066e-06, "epoch": 16.505706760316066, "percentage": 51.72, "elapsed_time": "0:49:59", "remaining_time": "0:46:40"}
+{"current_steps": 2360, "total_steps": 4544, "loss": 0.6085, "lr": 5.534866237504252e-06, "epoch": 16.575943810359966, "percentage": 51.94, "elapsed_time": "0:50:11", "remaining_time": "0:46:26"}
+{"current_steps": 2370, "total_steps": 4544, "loss": 0.6355, "lr": 5.496656086529467e-06, "epoch": 16.646180860403863, "percentage": 52.16, "elapsed_time": "0:50:22", "remaining_time": "0:46:12"}
+{"current_steps": 2380, "total_steps": 4544, "loss": 0.6206, "lr": 5.458416618580126e-06, "epoch": 16.71641791044776, "percentage": 52.38, "elapsed_time": "0:50:34", "remaining_time": "0:45:58"}
+{"current_steps": 2390, "total_steps": 4544, "loss": 0.6445, "lr": 5.420150090883191e-06, "epoch": 16.78665496049166, "percentage": 52.6, "elapsed_time": "0:50:45", "remaining_time": "0:45:45"}
+{"current_steps": 2400, "total_steps": 4544, "loss": 0.6508, "lr": 5.381858762262927e-06, "epoch": 16.856892010535557, "percentage": 52.82, "elapsed_time": "0:50:58", "remaining_time": "0:45:31"}
+{"current_steps": 2410, "total_steps": 4544, "loss": 0.6198, "lr": 5.343544893007563e-06, "epoch": 16.927129060579457, "percentage": 53.04, "elapsed_time": "0:51:09", "remaining_time": "0:45:17"}
+{"current_steps": 2420, "total_steps": 4544, "loss": 0.6156, "lr": 5.305210744735874e-06, "epoch": 16.997366110623354, "percentage": 53.26, "elapsed_time": "0:51:20", "remaining_time": "0:45:03"}
+{"current_steps": 2430, "total_steps": 4544, "loss": 0.5986, "lr": 5.266858580263678e-06, "epoch": 17.06760316066725, "percentage": 53.48, "elapsed_time": "0:51:31", "remaining_time": "0:44:49"}
+{"current_steps": 2440, "total_steps": 4544, "loss": 0.5637, "lr": 5.228490663470271e-06, "epoch": 17.13784021071115, "percentage": 53.7, "elapsed_time": "0:51:43", "remaining_time": "0:44:35"}
+{"current_steps": 2450, "total_steps": 4544, "loss": 0.5738, "lr": 5.190109259164782e-06, "epoch": 17.208077260755047, "percentage": 53.92, "elapsed_time": "0:51:54", "remaining_time": "0:44:22"}
+{"current_steps": 2460, "total_steps": 4544, "loss": 0.5912, "lr": 5.151716632952495e-06, "epoch": 17.278314310798947, "percentage": 54.14, "elapsed_time": "0:52:05", "remaining_time": "0:44:07"}
+{"current_steps": 2470, "total_steps": 4544, "loss": 0.5782, "lr": 5.113315051101111e-06, "epoch": 17.348551360842844, "percentage": 54.36, "elapsed_time": "0:52:16", "remaining_time": "0:43:53"}
+{"current_steps": 2480, "total_steps": 4544, "loss": 0.5548, "lr": 5.074906780406962e-06, "epoch": 17.418788410886744, "percentage": 54.58, "elapsed_time": "0:52:27", "remaining_time": "0:43:39"}
+{"current_steps": 2490, "total_steps": 4544, "loss": 0.5678, "lr": 5.036494088061222e-06, "epoch": 17.48902546093064, "percentage": 54.8, "elapsed_time": "0:52:38", "remaining_time": "0:43:25"}
+{"current_steps": 2500, "total_steps": 4544, "loss": 0.5768, "lr": 4.998079241516068e-06, "epoch": 17.559262510974538, "percentage": 55.02, "elapsed_time": "0:52:49", "remaining_time": "0:43:11"}
+{"current_steps": 2510, "total_steps": 4544, "loss": 0.5794, "lr": 4.959664508350834e-06, "epoch": 17.629499561018438, "percentage": 55.24, "elapsed_time": "0:53:00", "remaining_time": "0:42:57"}
+{"current_steps": 2520, "total_steps": 4544, "loss": 0.5754, "lr": 4.921252156138163e-06, "epoch": 17.699736611062335, "percentage": 55.46, "elapsed_time": "0:53:10", "remaining_time": "0:42:42"}
+{"current_steps": 2530, "total_steps": 4544, "loss": 0.5781, "lr": 4.882844452310155e-06, "epoch": 17.769973661106235, "percentage": 55.68, "elapsed_time": "0:53:21", "remaining_time": "0:42:28"}
+{"current_steps": 2540, "total_steps": 4544, "loss": 0.5834, "lr": 4.844443664024517e-06, "epoch": 17.84021071115013, "percentage": 55.9, "elapsed_time": "0:53:32", "remaining_time": "0:42:14"}
+{"current_steps": 2550, "total_steps": 4544, "loss": 0.565, "lr": 4.8060520580307456e-06, "epoch": 17.91044776119403, "percentage": 56.12, "elapsed_time": "0:53:43", "remaining_time": "0:42:00"}
+{"current_steps": 2560, "total_steps": 4544, "loss": 0.6071, "lr": 4.767671900536315e-06, "epoch": 17.98068481123793, "percentage": 56.34, "elapsed_time": "0:53:53", "remaining_time": "0:41:46"}
+{"current_steps": 2570, "total_steps": 4544, "loss": 0.5734, "lr": 4.729305457072913e-06, "epoch": 18.050921861281825, "percentage": 56.56, "elapsed_time": "0:54:04", "remaining_time": "0:41:31"}
+{"current_steps": 2580, "total_steps": 4544, "loss": 0.5322, "lr": 4.690954992362699e-06, "epoch": 18.121158911325725, "percentage": 56.78, "elapsed_time": "0:54:14", "remaining_time": "0:41:17"}
+{"current_steps": 2590, "total_steps": 4544, "loss": 0.5304, "lr": 4.652622770184637e-06, "epoch": 18.191395961369622, "percentage": 57.0, "elapsed_time": "0:54:24", "remaining_time": "0:41:02"}
+{"current_steps": 2600, "total_steps": 4544, "loss": 0.5368, "lr": 4.6143110532408455e-06, "epoch": 18.261633011413522, "percentage": 57.22, "elapsed_time": "0:54:35", "remaining_time": "0:40:48"}
+{"current_steps": 2610, "total_steps": 4544, "loss": 0.5456, "lr": 4.576022103023053e-06, "epoch": 18.33187006145742, "percentage": 57.44, "elapsed_time": "0:54:45", "remaining_time": "0:40:34"}
+{"current_steps": 2620, "total_steps": 4544, "loss": 0.5699, "lr": 4.537758179679098e-06, "epoch": 18.402107111501316, "percentage": 57.66, "elapsed_time": "0:54:55", "remaining_time": "0:40:19"}
+{"current_steps": 2630, "total_steps": 4544, "loss": 0.5587, "lr": 4.499521541879508e-06, "epoch": 18.472344161545216, "percentage": 57.88, "elapsed_time": "0:55:05", "remaining_time": "0:40:05"}
+{"current_steps": 2640, "total_steps": 4544, "loss": 0.526, "lr": 4.461314446684189e-06, "epoch": 18.542581211589113, "percentage": 58.1, "elapsed_time": "0:55:15", "remaining_time": "0:39:50"}
+{"current_steps": 2650, "total_steps": 4544, "loss": 0.5593, "lr": 4.423139149409176e-06, "epoch": 18.612818261633013, "percentage": 58.32, "elapsed_time": "0:55:25", "remaining_time": "0:39:36"}
+{"current_steps": 2660, "total_steps": 4544, "loss": 0.5379, "lr": 4.384997903493519e-06, "epoch": 18.68305531167691, "percentage": 58.54, "elapsed_time": "0:55:35", "remaining_time": "0:39:22"}
+{"current_steps": 2670, "total_steps": 4544, "loss": 0.5503, "lr": 4.346892960366255e-06, "epoch": 18.753292361720806, "percentage": 58.76, "elapsed_time": "0:55:44", "remaining_time": "0:39:07"}
+{"current_steps": 2680, "total_steps": 4544, "loss": 0.5275, "lr": 4.30882656931352e-06, "epoch": 18.823529411764707, "percentage": 58.98, "elapsed_time": "0:55:54", "remaining_time": "0:38:53"}
+{"current_steps": 2690, "total_steps": 4544, "loss": 0.5515, "lr": 4.270800977345767e-06, "epoch": 18.893766461808603, "percentage": 59.2, "elapsed_time": "0:56:04", "remaining_time": "0:38:38"}
+{"current_steps": 2700, "total_steps": 4544, "loss": 0.5484, "lr": 4.232818429065128e-06, "epoch": 18.964003511852503, "percentage": 59.42, "elapsed_time": "0:56:13", "remaining_time": "0:38:24"}
+{"current_steps": 2710, "total_steps": 4544, "loss": 0.5451, "lr": 4.194881166532923e-06, "epoch": 19.0342405618964, "percentage": 59.64, "elapsed_time": "0:56:23", "remaining_time": "0:38:09"}
+{"current_steps": 2720, "total_steps": 4544, "loss": 0.5222, "lr": 4.156991429137317e-06, "epoch": 19.104477611940297, "percentage": 59.86, "elapsed_time": "0:56:32", "remaining_time": "0:37:55"}
+{"current_steps": 2730, "total_steps": 4544, "loss": 0.5167, "lr": 4.119151453461121e-06, "epoch": 19.174714661984197, "percentage": 60.08, "elapsed_time": "0:56:42", "remaining_time": "0:37:40"}
+{"current_steps": 2740, "total_steps": 4544, "loss": 0.5009, "lr": 4.081363473149778e-06, "epoch": 19.244951712028094, "percentage": 60.3, "elapsed_time": "0:56:51", "remaining_time": "0:37:26"}
+{"current_steps": 2750, "total_steps": 4544, "loss": 0.5206, "lr": 4.0436297187795085e-06, "epoch": 19.315188762071994, "percentage": 60.52, "elapsed_time": "0:57:01", "remaining_time": "0:37:11"}
+{"current_steps": 2760, "total_steps": 4544, "loss": 0.5241, "lr": 4.005952417725649e-06, "epoch": 19.38542581211589, "percentage": 60.74, "elapsed_time": "0:57:10", "remaining_time": "0:36:57"}
+{"current_steps": 2770, "total_steps": 4544, "loss": 0.5247, "lr": 3.968333794031165e-06, "epoch": 19.45566286215979, "percentage": 60.96, "elapsed_time": "0:57:19", "remaining_time": "0:36:42"}
+{"current_steps": 2780, "total_steps": 4544, "loss": 0.5138, "lr": 3.930776068275375e-06, "epoch": 19.525899912203688, "percentage": 61.18, "elapsed_time": "0:57:28", "remaining_time": "0:36:27"}
+{"current_steps": 2790, "total_steps": 4544, "loss": 0.5295, "lr": 3.89328145744287e-06, "epoch": 19.596136962247584, "percentage": 61.4, "elapsed_time": "0:57:37", "remaining_time": "0:36:13"}
+{"current_steps": 2800, "total_steps": 4544, "loss": 0.5235, "lr": 3.8558521747926434e-06, "epoch": 19.666374012291485, "percentage": 61.62, "elapsed_time": "0:57:46", "remaining_time": "0:35:59"}
+{"current_steps": 2810, "total_steps": 4544, "loss": 0.5198, "lr": 3.818490429727455e-06, "epoch": 19.73661106233538, "percentage": 61.84, "elapsed_time": "0:57:55", "remaining_time": "0:35:44"}
+{"current_steps": 2820, "total_steps": 4544, "loss": 0.5037, "lr": 3.7811984276634024e-06, "epoch": 19.80684811237928, "percentage": 62.06, "elapsed_time": "0:58:03", "remaining_time": "0:35:29"}
+{"current_steps": 2830, "total_steps": 4544, "loss": 0.5069, "lr": 3.743978369899748e-06, "epoch": 19.877085162423178, "percentage": 62.28, "elapsed_time": "0:58:12", "remaining_time": "0:35:15"}
+{"current_steps": 2840, "total_steps": 4544, "loss": 0.5333, "lr": 3.70683245348897e-06, "epoch": 19.947322212467075, "percentage": 62.5, "elapsed_time": "0:58:20", "remaining_time": "0:35:00"}
+{"current_steps": 2850, "total_steps": 4544, "loss": 0.4939, "lr": 3.6697628711070786e-06, "epoch": 20.017559262510975, "percentage": 62.72, "elapsed_time": "0:58:29", "remaining_time": "0:34:46"}
+{"current_steps": 2860, "total_steps": 4544, "loss": 0.4879, "lr": 3.632771810924184e-06, "epoch": 20.08779631255487, "percentage": 62.94, "elapsed_time": "0:58:38", "remaining_time": "0:34:31"}
+{"current_steps": 2870, "total_steps": 4544, "loss": 0.518, "lr": 3.5958614564753313e-06, "epoch": 20.158033362598772, "percentage": 63.16, "elapsed_time": "0:58:46", "remaining_time": "0:34:17"}
+{"current_steps": 2880, "total_steps": 4544, "loss": 0.4944, "lr": 3.559033986531608e-06, "epoch": 20.22827041264267, "percentage": 63.38, "elapsed_time": "0:58:55", "remaining_time": "0:34:02"}
+{"current_steps": 2890, "total_steps": 4544, "loss": 0.5026, "lr": 3.522291574971538e-06, "epoch": 20.298507462686565, "percentage": 63.6, "elapsed_time": "0:59:03", "remaining_time": "0:33:47"}
+{"current_steps": 2900, "total_steps": 4544, "loss": 0.5134, "lr": 3.4856363906527513e-06, "epoch": 20.368744512730466, "percentage": 63.82, "elapsed_time": "0:59:12", "remaining_time": "0:33:33"}
+{"current_steps": 2910, "total_steps": 4544, "loss": 0.5056, "lr": 3.449070597283972e-06, "epoch": 20.438981562774362, "percentage": 64.04, "elapsed_time": "0:59:20", "remaining_time": "0:33:19"}
+{"current_steps": 2920, "total_steps": 4544, "loss": 0.5035, "lr": 3.4125963532972878e-06, "epoch": 20.509218612818263, "percentage": 64.26, "elapsed_time": "0:59:28", "remaining_time": "0:33:04"}
+{"current_steps": 2930, "total_steps": 4544, "loss": 0.5134, "lr": 3.376215811720744e-06, "epoch": 20.57945566286216, "percentage": 64.48, "elapsed_time": "0:59:36", "remaining_time": "0:32:50"}
+{"current_steps": 2940, "total_steps": 4544, "loss": 0.4666, "lr": 3.3399311200512495e-06, "epoch": 20.64969271290606, "percentage": 64.7, "elapsed_time": "0:59:44", "remaining_time": "0:32:35"}
+{"current_steps": 2950, "total_steps": 4544, "loss": 0.4965, "lr": 3.3037444201278202e-06, "epoch": 20.719929762949956, "percentage": 64.92, "elapsed_time": "0:59:52", "remaining_time": "0:32:21"}
+{"current_steps": 2960, "total_steps": 4544, "loss": 0.4953, "lr": 3.267657848005139e-06, "epoch": 20.790166812993853, "percentage": 65.14, "elapsed_time": "1:00:00", "remaining_time": "0:32:06"}
+{"current_steps": 2970, "total_steps": 4544, "loss": 0.4914, "lr": 3.2316735338274795e-06, "epoch": 20.860403863037753, "percentage": 65.36, "elapsed_time": "1:00:07", "remaining_time": "0:31:51"}
+{"current_steps": 2980, "total_steps": 4544, "loss": 0.5002, "lr": 3.1957936017029513e-06, "epoch": 20.93064091308165, "percentage": 65.58, "elapsed_time": "1:00:15", "remaining_time": "0:31:37"}
+{"current_steps": 2990, "total_steps": 4544, "loss": 0.5163, "lr": 3.1600201695781335e-06, "epoch": 21.00087796312555, "percentage": 65.8, "elapsed_time": "1:00:23", "remaining_time": "0:31:23"}
+{"current_steps": 3000, "total_steps": 4544, "loss": 0.4863, "lr": 3.124355349113037e-06, "epoch": 21.071115013169447, "percentage": 66.02, "elapsed_time": "1:00:31", "remaining_time": "0:31:09"}
+{"current_steps": 3010, "total_steps": 4544, "loss": 0.4616, "lr": 3.0888012455564707e-06, "epoch": 21.141352063213343, "percentage": 66.24, "elapsed_time": "1:00:38", "remaining_time": "0:30:54"}
+{"current_steps": 3020, "total_steps": 4544, "loss": 0.4897, "lr": 3.0533599576217664e-06, "epoch": 21.211589113257244, "percentage": 66.46, "elapsed_time": "1:00:46", "remaining_time": "0:30:40"}
+{"current_steps": 3030, "total_steps": 4544, "loss": 0.4801, "lr": 3.0180335773628912e-06, "epoch": 21.28182616330114, "percentage": 66.68, "elapsed_time": "1:00:53", "remaining_time": "0:30:25"}
+{"current_steps": 3040, "total_steps": 4544, "loss": 0.5161, "lr": 2.982824190050958e-06, "epoch": 21.35206321334504, "percentage": 66.9, "elapsed_time": "1:01:01", "remaining_time": "0:30:11"}
+{"current_steps": 3050, "total_steps": 4544, "loss": 0.4854, "lr": 2.94773387405114e-06, "epoch": 21.422300263388937, "percentage": 67.12, "elapsed_time": "1:01:08", "remaining_time": "0:29:57"}
+{"current_steps": 3060, "total_steps": 4544, "loss": 0.5062, "lr": 2.912764700699978e-06, "epoch": 21.492537313432837, "percentage": 67.34, "elapsed_time": "1:01:16", "remaining_time": "0:29:42"}
+{"current_steps": 3070, "total_steps": 4544, "loss": 0.4976, "lr": 2.8779187341831205e-06, "epoch": 21.562774363476734, "percentage": 67.56, "elapsed_time": "1:01:23", "remaining_time": "0:29:28"}
+{"current_steps": 3080, "total_steps": 4544, "loss": 0.4782, "lr": 2.843198031413473e-06, "epoch": 21.63301141352063, "percentage": 67.78, "elapsed_time": "1:01:30", "remaining_time": "0:29:14"}
+{"current_steps": 3090, "total_steps": 4544, "loss": 0.4694, "lr": 2.808604641909781e-06, "epoch": 21.70324846356453, "percentage": 68.0, "elapsed_time": "1:01:37", "remaining_time": "0:28:59"}
+{"current_steps": 3100, "total_steps": 4544, "loss": 0.4864, "lr": 2.7741406076756484e-06, "epoch": 21.773485513608428, "percentage": 68.22, "elapsed_time": "1:01:44", "remaining_time": "0:28:45"}
+{"current_steps": 3110, "total_steps": 4544, "loss": 0.4775, "lr": 2.7398079630790064e-06, "epoch": 21.843722563652328, "percentage": 68.44, "elapsed_time": "1:01:51", "remaining_time": "0:28:31"}
+{"current_steps": 3120, "total_steps": 4544, "loss": 0.4923, "lr": 2.7056087347320238e-06, "epoch": 21.913959613696225, "percentage": 68.66, "elapsed_time": "1:01:58", "remaining_time": "0:28:17"}
+{"current_steps": 3130, "total_steps": 4544, "loss": 0.4862, "lr": 2.6715449413714778e-06, "epoch": 21.98419666374012, "percentage": 68.88, "elapsed_time": "1:02:05", "remaining_time": "0:28:02"}
+{"current_steps": 3140, "total_steps": 4544, "loss": 0.4765, "lr": 2.637618593739588e-06, "epoch": 22.05443371378402, "percentage": 69.1, "elapsed_time": "1:02:11", "remaining_time": "0:27:48"}
+{"current_steps": 3150, "total_steps": 4544, "loss": 0.4581, "lr": 2.603831694465333e-06, "epoch": 22.12467076382792, "percentage": 69.32, "elapsed_time": "1:02:18", "remaining_time": "0:27:34"}
+{"current_steps": 3160, "total_steps": 4544, "loss": 0.479, "lr": 2.57018623794623e-06, "epoch": 22.19490781387182, "percentage": 69.54, "elapsed_time": "1:02:25", "remaining_time": "0:27:20"}
+{"current_steps": 3170, "total_steps": 4544, "loss": 0.4777, "lr": 2.5366842102306144e-06, "epoch": 22.265144863915715, "percentage": 69.76, "elapsed_time": "1:02:31", "remaining_time": "0:27:06"}
+{"current_steps": 3180, "total_steps": 4544, "loss": 0.4865, "lr": 2.503327588900396e-06, "epoch": 22.335381913959615, "percentage": 69.98, "elapsed_time": "1:02:38", "remaining_time": "0:26:51"}
+{"current_steps": 3190, "total_steps": 4544, "loss": 0.4691, "lr": 2.4701183429543386e-06, "epoch": 22.405618964003512, "percentage": 70.2, "elapsed_time": "1:02:44", "remaining_time": "0:26:37"}
+{"current_steps": 3200, "total_steps": 4544, "loss": 0.4755, "lr": 2.437058432691819e-06, "epoch": 22.47585601404741, "percentage": 70.42, "elapsed_time": "1:02:52", "remaining_time": "0:26:24"}
+{"current_steps": 3210, "total_steps": 4544, "loss": 0.46, "lr": 2.4041498095971253e-06, "epoch": 22.54609306409131, "percentage": 70.64, "elapsed_time": "1:03:08", "remaining_time": "0:26:14"}
+{"current_steps": 3220, "total_steps": 4544, "loss": 0.4886, "lr": 2.3713944162242506e-06, "epoch": 22.616330114135206, "percentage": 70.86, "elapsed_time": "1:03:25", "remaining_time": "0:26:04"}
+{"current_steps": 3230, "total_steps": 4544, "loss": 0.488, "lr": 2.3387941860822395e-06, "epoch": 22.686567164179106, "percentage": 71.08, "elapsed_time": "1:03:42", "remaining_time": "0:25:55"}
+{"current_steps": 3240, "total_steps": 4544, "loss": 0.4788, "lr": 2.3063510435210456e-06, "epoch": 22.756804214223003, "percentage": 71.3, "elapsed_time": "1:03:59", "remaining_time": "0:25:45"}
+{"current_steps": 3250, "total_steps": 4544, "loss": 0.4856, "lr": 2.2740669036179464e-06, "epoch": 22.8270412642669, "percentage": 71.52, "elapsed_time": "1:04:16", "remaining_time": "0:25:35"}
+{"current_steps": 3260, "total_steps": 4544, "loss": 0.4672, "lr": 2.24194367206449e-06, "epoch": 22.8972783143108, "percentage": 71.74, "elapsed_time": "1:04:32", "remaining_time": "0:25:25"}
+{"current_steps": 3270, "total_steps": 4544, "loss": 0.4663, "lr": 2.209983245054014e-06, "epoch": 22.967515364354696, "percentage": 71.96, "elapsed_time": "1:04:48", "remaining_time": "0:25:14"}
+{"current_steps": 3280, "total_steps": 4544, "loss": 0.465, "lr": 2.178187509169713e-06, "epoch": 23.037752414398597, "percentage": 72.18, "elapsed_time": "1:05:04", "remaining_time": "0:25:04"}
+{"current_steps": 3290, "total_steps": 4544, "loss": 0.4592, "lr": 2.146558341273273e-06, "epoch": 23.107989464442493, "percentage": 72.4, "elapsed_time": "1:05:21", "remaining_time": "0:24:54"}
+{"current_steps": 3300, "total_steps": 4544, "loss": 0.4521, "lr": 2.115097608394084e-06, "epoch": 23.17822651448639, "percentage": 72.62, "elapsed_time": "1:05:37", "remaining_time": "0:24:44"}
+{"current_steps": 3310, "total_steps": 4544, "loss": 0.4396, "lr": 2.083807167619029e-06, "epoch": 23.24846356453029, "percentage": 72.84, "elapsed_time": "1:05:53", "remaining_time": "0:24:34"}
+{"current_steps": 3320, "total_steps": 4544, "loss": 0.4557, "lr": 2.0526888659828716e-06, "epoch": 23.318700614574187, "percentage": 73.06, "elapsed_time": "1:06:09", "remaining_time": "0:24:23"}
+{"current_steps": 3330, "total_steps": 4544, "loss": 0.4686, "lr": 2.0217445403592185e-06, "epoch": 23.388937664618087, "percentage": 73.28, "elapsed_time": "1:06:25", "remaining_time": "0:24:13"}
+{"current_steps": 3340, "total_steps": 4544, "loss": 0.4696, "lr": 1.990976017352097e-06, "epoch": 23.459174714661984, "percentage": 73.5, "elapsed_time": "1:06:41", "remaining_time": "0:24:02"}
+{"current_steps": 3350, "total_steps": 4544, "loss": 0.4566, "lr": 1.9603851131881256e-06, "epoch": 23.529411764705884, "percentage": 73.72, "elapsed_time": "1:06:58", "remaining_time": "0:23:52"}
+{"current_steps": 3360, "total_steps": 4544, "loss": 0.4642, "lr": 1.9299736336093137e-06, "epoch": 23.59964881474978, "percentage": 73.94, "elapsed_time": "1:07:13", "remaining_time": "0:23:41"}
+{"current_steps": 3370, "total_steps": 4544, "loss": 0.4371, "lr": 1.8997433737664673e-06, "epoch": 23.669885864793677, "percentage": 74.16, "elapsed_time": "1:07:28", "remaining_time": "0:23:30"}
+{"current_steps": 3380, "total_steps": 4544, "loss": 0.4616, "lr": 1.869696118113216e-06, "epoch": 23.740122914837578, "percentage": 74.38, "elapsed_time": "1:07:44", "remaining_time": "0:23:19"}
+{"current_steps": 3390, "total_steps": 4544, "loss": 0.4551, "lr": 1.8398336403006956e-06, "epoch": 23.810359964881474, "percentage": 74.6, "elapsed_time": "1:08:00", "remaining_time": "0:23:09"}
+{"current_steps": 3400, "total_steps": 4544, "loss": 0.4456, "lr": 1.8101577030728324e-06, "epoch": 23.880597014925375, "percentage": 74.82, "elapsed_time": "1:08:15", "remaining_time": "0:22:58"}
+{"current_steps": 3410, "total_steps": 4544, "loss": 0.4756, "lr": 1.7806700581623059e-06, "epoch": 23.95083406496927, "percentage": 75.04, "elapsed_time": "1:08:31", "remaining_time": "0:22:47"}
+{"current_steps": 3420, "total_steps": 4544, "loss": 0.4519, "lr": 1.7513724461871423e-06, "epoch": 24.021071115013168, "percentage": 75.26, "elapsed_time": "1:08:46", "remaining_time": "0:22:36"}
+{"current_steps": 3430, "total_steps": 4544, "loss": 0.4489, "lr": 1.7222665965479585e-06, "epoch": 24.091308165057068, "percentage": 75.48, "elapsed_time": "1:09:02", "remaining_time": "0:22:25"}
+{"current_steps": 3440, "total_steps": 4544, "loss": 0.4369, "lr": 1.6933542273258924e-06, "epoch": 24.161545215100965, "percentage": 75.7, "elapsed_time": "1:09:17", "remaining_time": "0:22:14"}
+{"current_steps": 3450, "total_steps": 4544, "loss": 0.4246, "lr": 1.6646370451811784e-06, "epoch": 24.231782265144865, "percentage": 75.92, "elapsed_time": "1:09:32", "remaining_time": "0:22:03"}
+{"current_steps": 3460, "total_steps": 4544, "loss": 0.4181, "lr": 1.6361167452524073e-06, "epoch": 24.302019315188762, "percentage": 76.14, "elapsed_time": "1:09:47", "remaining_time": "0:21:51"}
+{"current_steps": 3470, "total_steps": 4544, "loss": 0.4349, "lr": 1.6077950110564606e-06, "epoch": 24.37225636523266, "percentage": 76.36, "elapsed_time": "1:10:01", "remaining_time": "0:21:40"}
+{"current_steps": 3480, "total_steps": 4544, "loss": 0.4358, "lr": 1.5796735143891423e-06, "epoch": 24.44249341527656, "percentage": 76.58, "elapsed_time": "1:10:17", "remaining_time": "0:21:29"}
+{"current_steps": 3490, "total_steps": 4544, "loss": 0.4428, "lr": 1.551753915226491e-06, "epoch": 24.512730465320455, "percentage": 76.8, "elapsed_time": "1:10:31", "remaining_time": "0:21:18"}
+{"current_steps": 3500, "total_steps": 4544, "loss": 0.4538, "lr": 1.5240378616267887e-06, "epoch": 24.582967515364356, "percentage": 77.02, "elapsed_time": "1:10:47", "remaining_time": "0:21:06"}
+{"current_steps": 3510, "total_steps": 4544, "loss": 0.4311, "lr": 1.4965269896332884e-06, "epoch": 24.653204565408252, "percentage": 77.24, "elapsed_time": "1:11:01", "remaining_time": "0:20:55"}
+{"current_steps": 3520, "total_steps": 4544, "loss": 0.4664, "lr": 1.46922292317763e-06, "epoch": 24.723441615452153, "percentage": 77.46, "elapsed_time": "1:11:16", "remaining_time": "0:20:44"}
+{"current_steps": 3530, "total_steps": 4544, "loss": 0.4315, "lr": 1.4421272739839898e-06, "epoch": 24.79367866549605, "percentage": 77.68, "elapsed_time": "1:11:30", "remaining_time": "0:20:32"}
+{"current_steps": 3540, "total_steps": 4544, "loss": 0.4403, "lr": 1.4152416414739401e-06, "epoch": 24.863915715539946, "percentage": 77.9, "elapsed_time": "1:11:45", "remaining_time": "0:20:21"}
+{"current_steps": 3550, "total_steps": 4544, "loss": 0.4652, "lr": 1.3885676126720315e-06, "epoch": 24.934152765583846, "percentage": 78.12, "elapsed_time": "1:12:00", "remaining_time": "0:20:09"}
+{"current_steps": 3560, "total_steps": 4544, "loss": 0.4439, "lr": 1.362106762112123e-06, "epoch": 25.004389815627743, "percentage": 78.35, "elapsed_time": "1:12:14", "remaining_time": "0:19:58"}
+{"current_steps": 3570, "total_steps": 4544, "loss": 0.443, "lr": 1.3358606517444328e-06, "epoch": 25.074626865671643, "percentage": 78.57, "elapsed_time": "1:12:28", "remaining_time": "0:19:46"}
+{"current_steps": 3580, "total_steps": 4544, "loss": 0.416, "lr": 1.3098308308433411e-06, "epoch": 25.14486391571554, "percentage": 78.79, "elapsed_time": "1:12:42", "remaining_time": "0:19:34"}
+{"current_steps": 3590, "total_steps": 4544, "loss": 0.4035, "lr": 1.2840188359159329e-06, "epoch": 25.215100965759436, "percentage": 79.01, "elapsed_time": "1:12:56", "remaining_time": "0:19:23"}
+{"current_steps": 3600, "total_steps": 4544, "loss": 0.4279, "lr": 1.258426190611306e-06, "epoch": 25.285338015803337, "percentage": 79.23, "elapsed_time": "1:13:11", "remaining_time": "0:19:11"}
+{"current_steps": 3610, "total_steps": 4544, "loss": 0.4358, "lr": 1.2330544056306315e-06, "epoch": 25.355575065847233, "percentage": 79.45, "elapsed_time": "1:13:25", "remaining_time": "0:18:59"}
+{"current_steps": 3620, "total_steps": 4544, "loss": 0.4368, "lr": 1.2079049786379782e-06, "epoch": 25.425812115891134, "percentage": 79.67, "elapsed_time": "1:13:39", "remaining_time": "0:18:48"}
+{"current_steps": 3630, "total_steps": 4544, "loss": 0.4271, "lr": 1.1829793941719053e-06, "epoch": 25.49604916593503, "percentage": 79.89, "elapsed_time": "1:13:53", "remaining_time": "0:18:36"}
+{"current_steps": 3640, "total_steps": 4544, "loss": 0.4482, "lr": 1.1582791235578321e-06, "epoch": 25.56628621597893, "percentage": 80.11, "elapsed_time": "1:14:07", "remaining_time": "0:18:24"}
+{"current_steps": 3650, "total_steps": 4544, "loss": 0.4131, "lr": 1.1338056248211916e-06, "epoch": 25.636523266022827, "percentage": 80.33, "elapsed_time": "1:14:21", "remaining_time": "0:18:12"}
+{"current_steps": 3660, "total_steps": 4544, "loss": 0.4513, "lr": 1.1095603426013613e-06, "epoch": 25.706760316066724, "percentage": 80.55, "elapsed_time": "1:14:35", "remaining_time": "0:18:00"}
+{"current_steps": 3670, "total_steps": 4544, "loss": 0.4118, "lr": 1.0855447080663907e-06, "epoch": 25.776997366110624, "percentage": 80.77, "elapsed_time": "1:14:48", "remaining_time": "0:17:49"}
+{"current_steps": 3680, "total_steps": 4544, "loss": 0.4082, "lr": 1.0617601388285149e-06, "epoch": 25.84723441615452, "percentage": 80.99, "elapsed_time": "1:15:02", "remaining_time": "0:17:37"}
+{"current_steps": 3690, "total_steps": 4544, "loss": 0.428, "lr": 1.0382080388604866e-06, "epoch": 25.91747146619842, "percentage": 81.21, "elapsed_time": "1:15:15", "remaining_time": "0:17:25"}
+{"current_steps": 3700, "total_steps": 4544, "loss": 0.452, "lr": 1.0148897984126876e-06, "epoch": 25.987708516242318, "percentage": 81.43, "elapsed_time": "1:15:29", "remaining_time": "0:17:13"}
+{"current_steps": 3710, "total_steps": 4544, "loss": 0.4177, "lr": 9.918067939310766e-07, "epoch": 26.057945566286215, "percentage": 81.65, "elapsed_time": "1:15:43", "remaining_time": "0:17:01"}
+{"current_steps": 3720, "total_steps": 4544, "loss": 0.4091, "lr": 9.689603879759284e-07, "epoch": 26.128182616330115, "percentage": 81.87, "elapsed_time": "1:15:56", "remaining_time": "0:16:49"}
+{"current_steps": 3730, "total_steps": 4544, "loss": 0.4035, "lr": 9.463519291414131e-07, "epoch": 26.19841966637401, "percentage": 82.09, "elapsed_time": "1:16:09", "remaining_time": "0:16:37"}
+{"current_steps": 3740, "total_steps": 4544, "loss": 0.41, "lr": 9.239827519759842e-07, "epoch": 26.26865671641791, "percentage": 82.31, "elapsed_time": "1:16:23", "remaining_time": "0:16:25"}
+{"current_steps": 3750, "total_steps": 4544, "loss": 0.4306, "lr": 9.018541769036054e-07, "epoch": 26.33889376646181, "percentage": 82.53, "elapsed_time": "1:16:36", "remaining_time": "0:16:13"}
+{"current_steps": 3760, "total_steps": 4544, "loss": 0.4122, "lr": 8.799675101458033e-07, "epoch": 26.409130816505705, "percentage": 82.75, "elapsed_time": "1:16:49", "remaining_time": "0:16:01"}
+{"current_steps": 3770, "total_steps": 4544, "loss": 0.3986, "lr": 8.583240436445666e-07, "epoch": 26.479367866549605, "percentage": 82.97, "elapsed_time": "1:17:02", "remaining_time": "0:15:48"}
+{"current_steps": 3780, "total_steps": 4544, "loss": 0.437, "lr": 8.369250549860869e-07, "epoch": 26.549604916593502, "percentage": 83.19, "elapsed_time": "1:17:15", "remaining_time": "0:15:36"}
+{"current_steps": 3790, "total_steps": 4544, "loss": 0.4344, "lr": 8.157718073253351e-07, "epoch": 26.619841966637402, "percentage": 83.41, "elapsed_time": "1:17:28", "remaining_time": "0:15:24"}
+{"current_steps": 3800, "total_steps": 4544, "loss": 0.4269, "lr": 7.948655493115098e-07, "epoch": 26.6900790166813, "percentage": 83.63, "elapsed_time": "1:17:41", "remaining_time": "0:15:12"}
+{"current_steps": 3810, "total_steps": 4544, "loss": 0.3962, "lr": 7.742075150143225e-07, "epoch": 26.7603160667252, "percentage": 83.85, "elapsed_time": "1:17:54", "remaining_time": "0:15:00"}
+{"current_steps": 3820, "total_steps": 4544, "loss": 0.4376, "lr": 7.537989238511578e-07, "epoch": 26.830553116769096, "percentage": 84.07, "elapsed_time": "1:18:06", "remaining_time": "0:14:48"}
+{"current_steps": 3830, "total_steps": 4544, "loss": 0.4504, "lr": 7.336409805150901e-07, "epoch": 26.900790166812993, "percentage": 84.29, "elapsed_time": "1:18:19", "remaining_time": "0:14:36"}
+{"current_steps": 3840, "total_steps": 4544, "loss": 0.4166, "lr": 7.137348749037748e-07, "epoch": 26.971027216856893, "percentage": 84.51, "elapsed_time": "1:18:31", "remaining_time": "0:14:23"}
+{"current_steps": 3850, "total_steps": 4544, "loss": 0.4244, "lr": 6.940817820492024e-07, "epoch": 27.04126426690079, "percentage": 84.73, "elapsed_time": "1:18:44", "remaining_time": "0:14:11"}
+{"current_steps": 3860, "total_steps": 4544, "loss": 0.4096, "lr": 6.746828620483487e-07, "epoch": 27.11150131694469, "percentage": 84.95, "elapsed_time": "1:18:56", "remaining_time": "0:13:59"}
+{"current_steps": 3870, "total_steps": 4544, "loss": 0.3896, "lr": 6.555392599946903e-07, "epoch": 27.181738366988586, "percentage": 85.17, "elapsed_time": "1:19:09", "remaining_time": "0:13:47"}
+{"current_steps": 3880, "total_steps": 4544, "loss": 0.4363, "lr": 6.366521059106078e-07, "epoch": 27.251975417032483, "percentage": 85.39, "elapsed_time": "1:19:21", "remaining_time": "0:13:34"}
+{"current_steps": 3890, "total_steps": 4544, "loss": 0.4183, "lr": 6.180225146806878e-07, "epoch": 27.322212467076383, "percentage": 85.61, "elapsed_time": "1:19:33", "remaining_time": "0:13:22"}
+{"current_steps": 3900, "total_steps": 4544, "loss": 0.3911, "lr": 5.996515859859109e-07, "epoch": 27.39244951712028, "percentage": 85.83, "elapsed_time": "1:19:46", "remaining_time": "0:13:10"}
+{"current_steps": 3910, "total_steps": 4544, "loss": 0.4237, "lr": 5.815404042387379e-07, "epoch": 27.46268656716418, "percentage": 86.05, "elapsed_time": "1:19:58", "remaining_time": "0:12:58"}
+{"current_steps": 3920, "total_steps": 4544, "loss": 0.4077, "lr": 5.636900385191014e-07, "epoch": 27.532923617208077, "percentage": 86.27, "elapsed_time": "1:20:10", "remaining_time": "0:12:45"}
+{"current_steps": 3930, "total_steps": 4544, "loss": 0.407, "lr": 5.461015425112915e-07, "epoch": 27.603160667251977, "percentage": 86.49, "elapsed_time": "1:20:22", "remaining_time": "0:12:33"}
+{"current_steps": 3940, "total_steps": 4544, "loss": 0.436, "lr": 5.287759544417687e-07, "epoch": 27.673397717295874, "percentage": 86.71, "elapsed_time": "1:20:34", "remaining_time": "0:12:21"}
+{"current_steps": 3950, "total_steps": 4544, "loss": 0.4014, "lr": 5.117142970178712e-07, "epoch": 27.74363476733977, "percentage": 86.93, "elapsed_time": "1:20:47", "remaining_time": "0:12:08"}
+{"current_steps": 3960, "total_steps": 4544, "loss": 0.4272, "lr": 4.949175773674502e-07, "epoch": 27.81387181738367, "percentage": 87.15, "elapsed_time": "1:20:58", "remaining_time": "0:11:56"}
+{"current_steps": 3970, "total_steps": 4544, "loss": 0.4179, "lr": 4.783867869794157e-07, "epoch": 27.884108867427567, "percentage": 87.37, "elapsed_time": "1:21:10", "remaining_time": "0:11:44"}
+{"current_steps": 3980, "total_steps": 4544, "loss": 0.4351, "lr": 4.6212290164521554e-07, "epoch": 27.954345917471468, "percentage": 87.59, "elapsed_time": "1:21:21", "remaining_time": "0:11:31"}
+{"current_steps": 3990, "total_steps": 4544, "loss": 0.3982, "lr": 4.461268814012304e-07, "epoch": 28.024582967515364, "percentage": 87.81, "elapsed_time": "1:21:33", "remaining_time": "0:11:19"}
+{"current_steps": 4000, "total_steps": 4544, "loss": 0.4249, "lr": 4.3039967047210865e-07, "epoch": 28.09482001755926, "percentage": 88.03, "elapsed_time": "1:21:45", "remaining_time": "0:11:07"}
+{"current_steps": 4000, "total_steps": 4544, "eval_loss": 1.6928666830062866, "epoch": 28.09482001755926, "percentage": 88.03, "elapsed_time": "1:21:56", "remaining_time": "0:11:08"}
+{"current_steps": 4010, "total_steps": 4544, "loss": 0.406, "lr": 4.1494219721502917e-07, "epoch": 28.16505706760316, "percentage": 88.25, "elapsed_time": "1:22:07", "remaining_time": "0:10:56"}
+{"current_steps": 4020, "total_steps": 4544, "loss": 0.4208, "lr": 3.997553740648974e-07, "epoch": 28.235294117647058, "percentage": 88.47, "elapsed_time": "1:22:18", "remaining_time": "0:10:43"}
+{"current_steps": 4030, "total_steps": 4544, "loss": 0.4113, "lr": 3.8484009748049053e-07, "epoch": 28.30553116769096, "percentage": 88.69, "elapsed_time": "1:22:30", "remaining_time": "0:10:31"}
+{"current_steps": 4040, "total_steps": 4544, "loss": 0.4131, "lr": 3.7019724789154e-07, "epoch": 28.375768217734855, "percentage": 88.91, "elapsed_time": "1:22:42", "remaining_time": "0:10:19"}
+{"current_steps": 4050, "total_steps": 4544, "loss": 0.4192, "lr": 3.558276896467555e-07, "epoch": 28.44600526777875, "percentage": 89.13, "elapsed_time": "1:22:53", "remaining_time": "0:10:06"}
+{"current_steps": 4060, "total_steps": 4544, "loss": 0.3869, "lr": 3.4173227096281124e-07, "epoch": 28.516242317822652, "percentage": 89.35, "elapsed_time": "1:23:04", "remaining_time": "0:09:54"}
+{"current_steps": 4070, "total_steps": 4544, "loss": 0.4353, "lr": 3.279118238742729e-07, "epoch": 28.58647936786655, "percentage": 89.57, "elapsed_time": "1:23:15", "remaining_time": "0:09:41"}
+{"current_steps": 4080, "total_steps": 4544, "loss": 0.3981, "lr": 3.143671641844831e-07, "epoch": 28.65671641791045, "percentage": 89.79, "elapsed_time": "1:23:25", "remaining_time": "0:09:29"}
+{"current_steps": 4090, "total_steps": 4544, "loss": 0.3976, "lr": 3.0109909141740614e-07, "epoch": 28.726953467954345, "percentage": 90.01, "elapsed_time": "1:23:36", "remaining_time": "0:09:16"}
+{"current_steps": 4100, "total_steps": 4544, "loss": 0.4075, "lr": 2.881083887704339e-07, "epoch": 28.797190517998246, "percentage": 90.23, "elapsed_time": "1:23:47", "remaining_time": "0:09:04"}
+{"current_steps": 4110, "total_steps": 4544, "loss": 0.4224, "lr": 2.753958230681547e-07, "epoch": 28.867427568042142, "percentage": 90.45, "elapsed_time": "1:23:58", "remaining_time": "0:08:52"}
+{"current_steps": 4120, "total_steps": 4544, "loss": 0.4291, "lr": 2.6296214471708826e-07, "epoch": 28.93766461808604, "percentage": 90.67, "elapsed_time": "1:24:09", "remaining_time": "0:08:39"}
+{"current_steps": 4130, "total_steps": 4544, "loss": 0.4209, "lr": 2.5080808766138996e-07, "epoch": 29.00790166812994, "percentage": 90.89, "elapsed_time": "1:24:19", "remaining_time": "0:08:27"}
+{"current_steps": 4140, "total_steps": 4544, "loss": 0.426, "lr": 2.3893436933952575e-07, "epoch": 29.078138718173836, "percentage": 91.11, "elapsed_time": "1:24:30", "remaining_time": "0:08:14"}
+{"current_steps": 4150, "total_steps": 4544, "loss": 0.4265, "lr": 2.2734169064192623e-07, "epoch": 29.148375768217736, "percentage": 91.33, "elapsed_time": "1:24:41", "remaining_time": "0:08:02"}
+{"current_steps": 4160, "total_steps": 4544, "loss": 0.3948, "lr": 2.1603073586961067e-07, "epoch": 29.218612818261633, "percentage": 91.55, "elapsed_time": "1:24:51", "remaining_time": "0:07:50"}
+{"current_steps": 4170, "total_steps": 4544, "loss": 0.4023, "lr": 2.0500217269379618e-07, "epoch": 29.28884986830553, "percentage": 91.77, "elapsed_time": "1:25:02", "remaining_time": "0:07:37"}
+{"current_steps": 4180, "total_steps": 4544, "loss": 0.4033, "lr": 1.9425665211648238e-07, "epoch": 29.35908691834943, "percentage": 91.99, "elapsed_time": "1:25:12", "remaining_time": "0:07:25"}
+{"current_steps": 4190, "total_steps": 4544, "loss": 0.4199, "lr": 1.837948084320268e-07, "epoch": 29.429323968393327, "percentage": 92.21, "elapsed_time": "1:25:22", "remaining_time": "0:07:12"}
+{"current_steps": 4200, "total_steps": 4544, "loss": 0.4347, "lr": 1.736172591897023e-07, "epoch": 29.499561018437227, "percentage": 92.43, "elapsed_time": "1:25:33", "remaining_time": "0:07:00"}
+{"current_steps": 4210, "total_steps": 4544, "loss": 0.3949, "lr": 1.6372460515724498e-07, "epoch": 29.569798068481123, "percentage": 92.65, "elapsed_time": "1:25:43", "remaining_time": "0:06:48"}
+{"current_steps": 4220, "total_steps": 4544, "loss": 0.4254, "lr": 1.541174302853876e-07, "epoch": 29.64003511852502, "percentage": 92.87, "elapsed_time": "1:25:53", "remaining_time": "0:06:35"}
+{"current_steps": 4230, "total_steps": 4544, "loss": 0.3997, "lr": 1.4479630167339554e-07, "epoch": 29.71027216856892, "percentage": 93.09, "elapsed_time": "1:26:02", "remaining_time": "0:06:23"}
+{"current_steps": 4240, "total_steps": 4544, "loss": 0.3875, "lr": 1.3576176953558783e-07, "epoch": 29.780509218612817, "percentage": 93.31, "elapsed_time": "1:26:12", "remaining_time": "0:06:10"}
+{"current_steps": 4250, "total_steps": 4544, "loss": 0.4052, "lr": 1.2701436716885897e-07, "epoch": 29.850746268656717, "percentage": 93.53, "elapsed_time": "1:26:22", "remaining_time": "0:05:58"}
+{"current_steps": 4260, "total_steps": 4544, "loss": 0.4227, "lr": 1.1855461092119991e-07, "epoch": 29.920983318700614, "percentage": 93.75, "elapsed_time": "1:26:32", "remaining_time": "0:05:46"}
+{"current_steps": 4270, "total_steps": 4544, "loss": 0.4192, "lr": 1.1038300016121883e-07, "epoch": 29.991220368744514, "percentage": 93.97, "elapsed_time": "1:26:42", "remaining_time": "0:05:33"}
+{"current_steps": 4280, "total_steps": 4544, "loss": 0.4067, "lr": 1.025000172486651e-07, "epoch": 30.06145741878841, "percentage": 94.19, "elapsed_time": "1:26:51", "remaining_time": "0:05:21"}
+{"current_steps": 4290, "total_steps": 4544, "loss": 0.4177, "lr": 9.490612750595096e-08, "epoch": 30.131694468832308, "percentage": 94.41, "elapsed_time": "1:27:01", "remaining_time": "0:05:09"}
+{"current_steps": 4300, "total_steps": 4544, "loss": 0.4126, "lr": 8.760177919069302e-08, "epoch": 30.201931518876208, "percentage": 94.63, "elapsed_time": "1:27:11", "remaining_time": "0:04:56"}
+{"current_steps": 4310, "total_steps": 4544, "loss": 0.4098, "lr": 8.058740346924221e-08, "epoch": 30.272168568920105, "percentage": 94.85, "elapsed_time": "1:27:20", "remaining_time": "0:04:44"}
+{"current_steps": 4320, "total_steps": 4544, "loss": 0.4183, "lr": 7.386341439124145e-08, "epoch": 30.342405618964005, "percentage": 95.07, "elapsed_time": "1:27:29", "remaining_time": "0:04:32"}
+{"current_steps": 4330, "total_steps": 4544, "loss": 0.4249, "lr": 6.74302088651796e-08, "epoch": 30.4126426690079, "percentage": 95.29, "elapsed_time": "1:27:39", "remaining_time": "0:04:19"}
+{"current_steps": 4340, "total_steps": 4544, "loss": 0.4003, "lr": 6.128816663496296e-08, "epoch": 30.482879719051798, "percentage": 95.51, "elapsed_time": "1:27:48", "remaining_time": "0:04:07"}
+{"current_steps": 4350, "total_steps": 4544, "loss": 0.4085, "lr": 5.543765025750103e-08, "epoch": 30.5531167690957, "percentage": 95.73, "elapsed_time": "1:27:57", "remaining_time": "0:03:55"}
+{"current_steps": 4360, "total_steps": 4544, "loss": 0.437, "lr": 4.987900508130417e-08, "epoch": 30.623353819139595, "percentage": 95.95, "elapsed_time": "1:28:07", "remaining_time": "0:03:43"}
+{"current_steps": 4370, "total_steps": 4544, "loss": 0.4117, "lr": 4.461255922609986e-08, "epoch": 30.693590869183495, "percentage": 96.17, "elapsed_time": "1:28:16", "remaining_time": "0:03:30"}
+{"current_steps": 4380, "total_steps": 4544, "loss": 0.4054, "lr": 3.963862356346049e-08, "epoch": 30.763827919227392, "percentage": 96.39, "elapsed_time": "1:28:25", "remaining_time": "0:03:18"}
+{"current_steps": 4390, "total_steps": 4544, "loss": 0.3948, "lr": 3.49574916984563e-08, "epoch": 30.834064969271292, "percentage": 96.61, "elapsed_time": "1:28:34", "remaining_time": "0:03:06"}
+{"current_steps": 4400, "total_steps": 4544, "loss": 0.3932, "lr": 3.056943995232431e-08, "epoch": 30.90430201931519, "percentage": 96.83, "elapsed_time": "1:28:43", "remaining_time": "0:02:54"}
+{"current_steps": 4410, "total_steps": 4544, "loss": 0.426, "lr": 2.6474727346155194e-08, "epoch": 30.974539069359086, "percentage": 97.05, "elapsed_time": "1:28:52", "remaining_time": "0:02:42"}
+{"current_steps": 4420, "total_steps": 4544, "loss": 0.4219, "lr": 2.2673595585605557e-08, "epoch": 31.044776119402986, "percentage": 97.27, "elapsed_time": "1:29:00", "remaining_time": "0:02:29"}
+{"current_steps": 4430, "total_steps": 4544, "loss": 0.426, "lr": 1.9166269046628215e-08, "epoch": 31.115013169446883, "percentage": 97.49, "elapsed_time": "1:29:09", "remaining_time": "0:02:17"}
+{"current_steps": 4440, "total_steps": 4544, "loss": 0.4222, "lr": 1.5952954762230575e-08, "epoch": 31.185250219490783, "percentage": 97.71, "elapsed_time": "1:29:18", "remaining_time": "0:02:05"}
+{"current_steps": 4450, "total_steps": 4544, "loss": 0.4178, "lr": 1.3033842410251074e-08, "epoch": 31.25548726953468, "percentage": 97.93, "elapsed_time": "1:29:27", "remaining_time": "0:01:53"}
+{"current_steps": 4460, "total_steps": 4544, "loss": 0.4068, "lr": 1.0409104302164241e-08, "epoch": 31.325724319578576, "percentage": 98.15, "elapsed_time": "1:29:35", "remaining_time": "0:01:41"}
+{"current_steps": 4470, "total_steps": 4544, "loss": 0.4126, "lr": 8.078895372908846e-09, "epoch": 31.395961369622476, "percentage": 98.37, "elapsed_time": "1:29:43", "remaining_time": "0:01:29"}
+{"current_steps": 4480, "total_steps": 4544, "loss": 0.4136, "lr": 6.0433531717424275e-09, "epoch": 31.466198419666373, "percentage": 98.59, "elapsed_time": "1:29:52", "remaining_time": "0:01:17"}
+{"current_steps": 4490, "total_steps": 4544, "loss": 0.4096, "lr": 4.302597854121127e-09, "epoch": 31.536435469710273, "percentage": 98.81, "elapsed_time": "1:30:00", "remaining_time": "0:01:04"}
+{"current_steps": 4500, "total_steps": 4544, "loss": 0.4092, "lr": 2.8567321746092446e-09, "epoch": 31.60667251975417, "percentage": 99.03, "elapsed_time": "1:30:08", "remaining_time": "0:00:52"}
+{"current_steps": 4510, "total_steps": 4544, "loss": 0.4072, "lr": 1.705841480810766e-09, "epoch": 31.67690956979807, "percentage": 99.25, "elapsed_time": "1:30:16", "remaining_time": "0:00:40"}
+{"current_steps": 4520, "total_steps": 4544, "loss": 0.4261, "lr": 8.499937083339404e-10, "epoch": 31.747146619841967, "percentage": 99.47, "elapsed_time": "1:30:24", "remaining_time": "0:00:28"}
+{"current_steps": 4530, "total_steps": 4544, "loss": 0.4138, "lr": 2.892393767800483e-10, "epoch": 31.817383669885864, "percentage": 99.69, "elapsed_time": "1:30:32", "remaining_time": "0:00:16"}
+{"current_steps": 4540, "total_steps": 4544, "loss": 0.3897, "lr": 2.3611586760785566e-11, "epoch": 31.887620719929764, "percentage": 99.91, "elapsed_time": "1:30:40", "remaining_time": "0:00:04"}
+{"current_steps": 4544, "total_steps": 4544, "epoch": 31.91571553994732, "percentage": 100.0, "elapsed_time": "1:31:46", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,3236 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 31.91571553994732,
+  "eval_steps": 2000,
+  "global_step": 4544,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.07023705004389816,
+      "grad_norm": 6.03125,
+      "learning_rate": 2.197802197802198e-07,
+      "loss": 2.4857,
+      "step": 10
+    },
+    {
+      "epoch": 0.14047410008779632,
+      "grad_norm": 5.03125,
+      "learning_rate": 4.395604395604396e-07,
+      "loss": 2.5113,
+      "step": 20
+    },
+    {
+      "epoch": 0.21071115013169447,
+      "grad_norm": 4.71875,
+      "learning_rate": 6.593406593406594e-07,
+      "loss": 2.4922,
+      "step": 30
+    },
+    {
+      "epoch": 0.28094820017559263,
+      "grad_norm": 5.09375,
+      "learning_rate": 8.791208791208792e-07,
+      "loss": 2.4394,
+      "step": 40
+    },
+    {
+      "epoch": 0.35118525021949076,
+      "grad_norm": 16.75,
+      "learning_rate": 1.098901098901099e-06,
+      "loss": 2.4773,
+      "step": 50
+    },
+    {
+      "epoch": 0.42142230026338895,
+      "grad_norm": 14.5625,
+      "learning_rate": 1.3186813186813187e-06,
+      "loss": 2.3524,
+      "step": 60
+    },
+    {
+      "epoch": 0.4916593503072871,
+      "grad_norm": 15.125,
+      "learning_rate": 1.5384615384615387e-06,
+      "loss": 2.42,
+      "step": 70
+    },
+    {
+      "epoch": 0.5618964003511853,
+      "grad_norm": 14.0625,
+      "learning_rate": 1.7582417582417585e-06,
+      "loss": 2.3288,
+      "step": 80
+    },
+    {
+      "epoch": 0.6321334503950834,
+      "grad_norm": 12.6875,
+      "learning_rate": 1.9780219780219782e-06,
+      "loss": 2.2519,
+      "step": 90
+    },
+    {
+      "epoch": 0.7023705004389815,
+      "grad_norm": 2.53125,
+      "learning_rate": 2.197802197802198e-06,
+      "loss": 2.2021,
+      "step": 100
+    },
+    {
+      "epoch": 0.7726075504828798,
+      "grad_norm": 2.296875,
+      "learning_rate": 2.4175824175824177e-06,
+      "loss": 2.2151,
+      "step": 110
+    },
+    {
+      "epoch": 0.8428446005267779,
+      "grad_norm": 2.015625,
+      "learning_rate": 2.6373626373626375e-06,
+      "loss": 2.1102,
+      "step": 120
+    },
+    {
+      "epoch": 0.913081650570676,
+      "grad_norm": 2.109375,
+      "learning_rate": 2.8571428571428573e-06,
+      "loss": 2.1034,
+      "step": 130
+    },
+    {
+      "epoch": 0.9833187006145742,
+      "grad_norm": 2.078125,
+      "learning_rate": 3.0769230769230774e-06,
+      "loss": 2.0737,
+      "step": 140
+    },
+    {
+      "epoch": 1.0535557506584723,
+      "grad_norm": 1.7578125,
+      "learning_rate": 3.2967032967032968e-06,
+      "loss": 2.0345,
+      "step": 150
+    },
+    {
+      "epoch": 1.1237928007023705,
+      "grad_norm": 2.109375,
+      "learning_rate": 3.516483516483517e-06,
+      "loss": 1.9568,
+      "step": 160
+    },
+    {
+      "epoch": 1.1940298507462686,
+      "grad_norm": 2.1875,
+      "learning_rate": 3.7362637362637367e-06,
+      "loss": 1.9494,
+      "step": 170
+    },
+    {
+      "epoch": 1.2642669007901668,
+      "grad_norm": 1.875,
+      "learning_rate": 3.9560439560439565e-06,
+      "loss": 1.8932,
+      "step": 180
+    },
+    {
+      "epoch": 1.334503950834065,
+      "grad_norm": 1.6875,
+      "learning_rate": 4.175824175824177e-06,
+      "loss": 1.9219,
+      "step": 190
+    },
+    {
+      "epoch": 1.404741000877963,
+      "grad_norm": 2.1875,
+      "learning_rate": 4.395604395604396e-06,
+      "loss": 1.9751,
+      "step": 200
+    },
+    {
+      "epoch": 1.4749780509218613,
+      "grad_norm": 1.7265625,
+      "learning_rate": 4.615384615384616e-06,
+      "loss": 1.8729,
+      "step": 210
+    },
+    {
+      "epoch": 1.5452151009657595,
+      "grad_norm": 1.765625,
+      "learning_rate": 4.8351648351648355e-06,
+      "loss": 1.885,
+      "step": 220
+    },
+    {
+      "epoch": 1.6154521510096576,
+      "grad_norm": 1.546875,
+      "learning_rate": 5.054945054945055e-06,
+      "loss": 1.8549,
+      "step": 230
+    },
+    {
+      "epoch": 1.6856892010535558,
+      "grad_norm": 2.046875,
+      "learning_rate": 5.274725274725275e-06,
+      "loss": 1.8736,
+      "step": 240
+    },
+    {
+      "epoch": 1.755926251097454,
+      "grad_norm": 1.6875,
+      "learning_rate": 5.494505494505495e-06,
+      "loss": 1.8388,
+      "step": 250
+    },
+    {
+      "epoch": 1.826163301141352,
+      "grad_norm": 2.09375,
+      "learning_rate": 5.7142857142857145e-06,
+      "loss": 1.8676,
+      "step": 260
+    },
+    {
+      "epoch": 1.89640035118525,
+      "grad_norm": 2.0,
+      "learning_rate": 5.934065934065935e-06,
+      "loss": 1.8181,
+      "step": 270
+    },
+    {
+      "epoch": 1.9666374012291485,
+      "grad_norm": 1.890625,
+      "learning_rate": 6.153846153846155e-06,
+      "loss": 1.8007,
+      "step": 280
+    },
+    {
+      "epoch": 2.0368744512730466,
+      "grad_norm": 1.8046875,
+      "learning_rate": 6.373626373626373e-06,
+      "loss": 1.7727,
+      "step": 290
+    },
+    {
+      "epoch": 2.1071115013169446,
+      "grad_norm": 1.8203125,
+      "learning_rate": 6.5934065934065935e-06,
+      "loss": 1.7038,
+      "step": 300
+    },
+    {
+      "epoch": 2.177348551360843,
+      "grad_norm": 1.7578125,
+      "learning_rate": 6.813186813186814e-06,
+      "loss": 1.7398,
+      "step": 310
+    },
+    {
+      "epoch": 2.247585601404741,
+      "grad_norm": 2.15625,
+      "learning_rate": 7.032967032967034e-06,
+      "loss": 1.7468,
+      "step": 320
+    },
+    {
+      "epoch": 2.317822651448639,
+      "grad_norm": 1.9296875,
+      "learning_rate": 7.252747252747253e-06,
+      "loss": 1.7326,
+      "step": 330
+    },
+    {
+      "epoch": 2.388059701492537,
+      "grad_norm": 1.8671875,
+      "learning_rate": 7.472527472527473e-06,
+      "loss": 1.7779,
+      "step": 340
+    },
+    {
+      "epoch": 2.4582967515364356,
+      "grad_norm": 1.96875,
+      "learning_rate": 7.692307692307694e-06,
+      "loss": 1.6381,
+      "step": 350
+    },
+    {
+      "epoch": 2.5285338015803336,
+      "grad_norm": 1.71875,
+      "learning_rate": 7.912087912087913e-06,
+      "loss": 1.6605,
+      "step": 360
+    },
+    {
+      "epoch": 2.5987708516242316,
+      "grad_norm": 1.703125,
+      "learning_rate": 8.131868131868132e-06,
+      "loss": 1.6683,
+      "step": 370
+    },
+    {
+      "epoch": 2.66900790166813,
+      "grad_norm": 1.6640625,
+      "learning_rate": 8.351648351648353e-06,
+      "loss": 1.6819,
+      "step": 380
+    },
+    {
+      "epoch": 2.739244951712028,
+      "grad_norm": 1.984375,
+      "learning_rate": 8.571428571428571e-06,
+      "loss": 1.7072,
+      "step": 390
+    },
+    {
+      "epoch": 2.809482001755926,
+      "grad_norm": 1.8671875,
+      "learning_rate": 8.791208791208792e-06,
+      "loss": 1.6747,
+      "step": 400
+    },
+    {
+      "epoch": 2.8797190517998246,
+      "grad_norm": 1.8671875,
+      "learning_rate": 9.010989010989011e-06,
+      "loss": 1.6596,
+      "step": 410
+    },
+    {
+      "epoch": 2.9499561018437226,
+      "grad_norm": 1.7265625,
+      "learning_rate": 9.230769230769232e-06,
+      "loss": 1.65,
+      "step": 420
+    },
+    {
+      "epoch": 3.0201931518876206,
+      "grad_norm": 1.53125,
+      "learning_rate": 9.450549450549452e-06,
+      "loss": 1.6037,
+      "step": 430
+    },
+    {
+      "epoch": 3.090430201931519,
+      "grad_norm": 1.78125,
+      "learning_rate": 9.670329670329671e-06,
+      "loss": 1.6015,
+      "step": 440
+    },
+    {
+      "epoch": 3.160667251975417,
+      "grad_norm": 1.859375,
+      "learning_rate": 9.890109890109892e-06,
+      "loss": 1.5708,
+      "step": 450
+    },
+    {
+      "epoch": 3.230904302019315,
+      "grad_norm": 1.671875,
+      "learning_rate": 9.99996310691202e-06,
+      "loss": 1.5778,
+      "step": 460
+    },
+    {
+      "epoch": 3.3011413520632136,
+      "grad_norm": 1.609375,
+      "learning_rate": 9.999667965474806e-06,
+      "loss": 1.5528,
+      "step": 470
+    },
+    {
+      "epoch": 3.3713784021071116,
+      "grad_norm": 1.984375,
+      "learning_rate": 9.9990777000222e-06,
+      "loss": 1.5539,
+      "step": 480
+    },
+    {
+      "epoch": 3.4416154521510096,
+      "grad_norm": 2.5,
+      "learning_rate": 9.998192345396817e-06,
+      "loss": 1.5289,
+      "step": 490
+    },
+    {
+      "epoch": 3.511852502194908,
+      "grad_norm": 1.484375,
+      "learning_rate": 9.997011953860014e-06,
+      "loss": 1.5335,
+      "step": 500
+    },
+    {
+      "epoch": 3.582089552238806,
+      "grad_norm": 1.4765625,
+      "learning_rate": 9.99553659508879e-06,
+      "loss": 1.5565,
+      "step": 510
+    },
+    {
+      "epoch": 3.652326602282704,
+      "grad_norm": 1.46875,
+      "learning_rate": 9.993766356171694e-06,
+      "loss": 1.5414,
+      "step": 520
+    },
+    {
+      "epoch": 3.722563652326602,
+      "grad_norm": 1.40625,
+      "learning_rate": 9.991701341603667e-06,
+      "loss": 1.5271,
+      "step": 530
+    },
+    {
+      "epoch": 3.7928007023705006,
+      "grad_norm": 1.4921875,
+      "learning_rate": 9.989341673279881e-06,
+      "loss": 1.4782,
+      "step": 540
+    },
+    {
+      "epoch": 3.8630377524143986,
+      "grad_norm": 1.6015625,
+      "learning_rate": 9.986687490488545e-06,
+      "loss": 1.5165,
+      "step": 550
+    },
+    {
+      "epoch": 3.9332748024582966,
+      "grad_norm": 1.6953125,
+      "learning_rate": 9.98373894990268e-06,
+      "loss": 1.507,
+      "step": 560
+    },
+    {
+      "epoch": 4.003511852502195,
+      "grad_norm": 1.65625,
+      "learning_rate": 9.980496225570869e-06,
+      "loss": 1.5207,
+      "step": 570
+    },
+    {
+      "epoch": 4.073748902546093,
+      "grad_norm": 1.59375,
+      "learning_rate": 9.976959508906992e-06,
+      "loss": 1.4324,
+      "step": 580
+    },
+    {
+      "epoch": 4.143985952589992,
+      "grad_norm": 1.5859375,
+      "learning_rate": 9.973129008678915e-06,
+      "loss": 1.4171,
+      "step": 590
+    },
+    {
+      "epoch": 4.214223002633889,
+      "grad_norm": 1.4375,
+      "learning_rate": 9.969004950996175e-06,
+      "loss": 1.4378,
+      "step": 600
+    },
+    {
+      "epoch": 4.284460052677788,
+      "grad_norm": 1.3828125,
+      "learning_rate": 9.964587579296631e-06,
+      "loss": 1.3886,
+      "step": 610
+    },
+    {
+      "epoch": 4.354697102721686,
+      "grad_norm": 1.6796875,
+      "learning_rate": 9.959877154332095e-06,
+      "loss": 1.4154,
+      "step": 620
+    },
+    {
+      "epoch": 4.424934152765584,
+      "grad_norm": 1.625,
+      "learning_rate": 9.954873954152933e-06,
+      "loss": 1.4231,
+      "step": 630
+    },
+    {
+      "epoch": 4.495171202809482,
+      "grad_norm": 1.4609375,
+      "learning_rate": 9.949578274091666e-06,
+      "loss": 1.3745,
+      "step": 640
+    },
+    {
+      "epoch": 4.565408252853381,
+      "grad_norm": 1.6796875,
+      "learning_rate": 9.943990426745525e-06,
+      "loss": 1.3663,
+      "step": 650
+    },
+    {
+      "epoch": 4.635645302897278,
+      "grad_norm": 1.5,
+      "learning_rate": 9.938110741958003e-06,
+      "loss": 1.4177,
+      "step": 660
+    },
+    {
+      "epoch": 4.705882352941177,
+      "grad_norm": 1.6171875,
+      "learning_rate": 9.931939566799385e-06,
+      "loss": 1.412,
+      "step": 670
+    },
+    {
+      "epoch": 4.776119402985074,
+      "grad_norm": 1.6953125,
+      "learning_rate": 9.925477265546258e-06,
+      "loss": 1.4155,
+      "step": 680
+    },
+    {
+      "epoch": 4.846356453028973,
+      "grad_norm": 1.5859375,
+      "learning_rate": 9.918724219660013e-06,
+      "loss": 1.4241,
+      "step": 690
+    },
+    {
+      "epoch": 4.916593503072871,
+      "grad_norm": 1.5234375,
+      "learning_rate": 9.911680827764329e-06,
+      "loss": 1.3562,
+      "step": 700
+    },
+    {
+      "epoch": 4.98683055311677,
+      "grad_norm": 1.4609375,
+      "learning_rate": 9.90434750562163e-06,
+      "loss": 1.3542,
+      "step": 710
+    },
+    {
+      "epoch": 5.057067603160667,
+      "grad_norm": 1.484375,
+      "learning_rate": 9.896724686108561e-06,
+      "loss": 1.305,
+      "step": 720
+    },
+    {
+      "epoch": 5.127304653204566,
+      "grad_norm": 1.4609375,
+      "learning_rate": 9.888812819190419e-06,
+      "loss": 1.314,
+      "step": 730
+    },
+    {
+      "epoch": 5.197541703248463,
+      "grad_norm": 1.5390625,
+      "learning_rate": 9.88061237189461e-06,
+      "loss": 1.2962,
+      "step": 740
+    },
+    {
+      "epoch": 5.267778753292362,
+      "grad_norm": 1.4375,
+      "learning_rate": 9.872123828283063e-06,
+      "loss": 1.2993,
+      "step": 750
+    },
+    {
+      "epoch": 5.33801580333626,
+      "grad_norm": 1.4140625,
+      "learning_rate": 9.863347689423666e-06,
+      "loss": 1.2649,
+      "step": 760
+    },
+    {
+      "epoch": 5.408252853380158,
+      "grad_norm": 1.6796875,
+      "learning_rate": 9.854284473360694e-06,
+      "loss": 1.2943,
+      "step": 770
+    },
+    {
+      "epoch": 5.478489903424056,
+      "grad_norm": 1.3203125,
+      "learning_rate": 9.84493471508421e-06,
+      "loss": 1.2512,
+      "step": 780
+    },
+    {
+      "epoch": 5.548726953467955,
+      "grad_norm": 1.375,
+      "learning_rate": 9.835298966498511e-06,
+      "loss": 1.3046,
+      "step": 790
+    },
+    {
+      "epoch": 5.618964003511852,
+      "grad_norm": 1.390625,
+      "learning_rate": 9.82537779638953e-06,
+      "loss": 1.2338,
+      "step": 800
+    },
+    {
+      "epoch": 5.689201053555751,
+      "grad_norm": 1.3359375,
+      "learning_rate": 9.815171790391269e-06,
+      "loss": 1.2921,
+      "step": 810
+    },
+    {
+      "epoch": 5.759438103599649,
+      "grad_norm": 1.3046875,
+      "learning_rate": 9.804681550951228e-06,
+      "loss": 1.2306,
+      "step": 820
+    },
+    {
+      "epoch": 5.829675153643547,
+      "grad_norm": 1.3515625,
+      "learning_rate": 9.793907697294844e-06,
+      "loss": 1.2304,
+      "step": 830
+    },
+    {
+      "epoch": 5.899912203687445,
+      "grad_norm": 1.1796875,
+      "learning_rate": 9.782850865388941e-06,
+      "loss": 1.2693,
+      "step": 840
+    },
+    {
+      "epoch": 5.970149253731344,
+      "grad_norm": 1.2578125,
+      "learning_rate": 9.77151170790419e-06,
+      "loss": 1.2642,
+      "step": 850
+    },
+    {
+      "epoch": 6.040386303775241,
+      "grad_norm": 1.3203125,
+      "learning_rate": 9.759890894176574e-06,
+      "loss": 1.2702,
+      "step": 860
+    },
+    {
+      "epoch": 6.11062335381914,
+      "grad_norm": 1.328125,
+      "learning_rate": 9.747989110167887e-06,
+      "loss": 1.2085,
+      "step": 870
+    },
+    {
+      "epoch": 6.180860403863038,
+      "grad_norm": 1.265625,
+      "learning_rate": 9.735807058425241e-06,
+      "loss": 1.1754,
+      "step": 880
+    },
+    {
+      "epoch": 6.251097453906936,
+      "grad_norm": 1.1484375,
+      "learning_rate": 9.723345458039595e-06,
+      "loss": 1.1378,
+      "step": 890
+    },
+    {
+      "epoch": 6.321334503950834,
+      "grad_norm": 1.1640625,
+      "learning_rate": 9.710605044603305e-06,
+      "loss": 1.1588,
+      "step": 900
+    },
+    {
+      "epoch": 6.391571553994732,
+      "grad_norm": 1.21875,
+      "learning_rate": 9.697586570166707e-06,
+      "loss": 1.1934,
+      "step": 910
+    },
+    {
+      "epoch": 6.46180860403863,
+      "grad_norm": 1.0859375,
+      "learning_rate": 9.684290803193721e-06,
+      "loss": 1.1504,
+      "step": 920
+    },
+    {
+      "epoch": 6.532045654082529,
+      "grad_norm": 1.0859375,
+      "learning_rate": 9.670718528516495e-06,
+      "loss": 1.1152,
+      "step": 930
+    },
+    {
+      "epoch": 6.602282704126427,
+      "grad_norm": 1.171875,
+      "learning_rate": 9.65687054728907e-06,
+      "loss": 1.1628,
+      "step": 940
+    },
+    {
+      "epoch": 6.672519754170325,
+      "grad_norm": 1.109375,
+      "learning_rate": 9.642747676940094e-06,
+      "loss": 1.1299,
+      "step": 950
+    },
+    {
+      "epoch": 6.742756804214223,
+      "grad_norm": 1.125,
+      "learning_rate": 9.62835075112457e-06,
+      "loss": 1.1635,
+      "step": 960
+    },
+    {
+      "epoch": 6.812993854258121,
+      "grad_norm": 0.953125,
+      "learning_rate": 9.61368061967464e-06,
+      "loss": 1.1673,
+      "step": 970
+    },
+    {
+      "epoch": 6.883230904302019,
+      "grad_norm": 0.953125,
+      "learning_rate": 9.598738148549434e-06,
+      "loss": 1.1281,
+      "step": 980
+    },
+    {
+      "epoch": 6.953467954345918,
+      "grad_norm": 1.078125,
+      "learning_rate": 9.583524219783938e-06,
+      "loss": 1.1617,
+      "step": 990
+    },
+    {
+      "epoch": 7.023705004389815,
+      "grad_norm": 1.0,
+      "learning_rate": 9.56803973143694e-06,
+      "loss": 1.1436,
+      "step": 1000
+    },
+    {
+      "epoch": 7.093942054433714,
+      "grad_norm": 1.0078125,
+      "learning_rate": 9.552285597538014e-06,
+      "loss": 1.0962,
+      "step": 1010
+    },
+    {
+      "epoch": 7.164179104477612,
+      "grad_norm": 0.8828125,
+      "learning_rate": 9.536262748033564e-06,
+      "loss": 1.0893,
+      "step": 1020
+    },
+    {
+      "epoch": 7.23441615452151,
+      "grad_norm": 1.0234375,
+      "learning_rate": 9.519972128731937e-06,
+      "loss": 1.0618,
+      "step": 1030
+    },
+    {
+      "epoch": 7.304653204565408,
+      "grad_norm": 1.0078125,
+      "learning_rate": 9.503414701247587e-06,
+      "loss": 1.0986,
+      "step": 1040
+    },
+    {
+      "epoch": 7.374890254609307,
+      "grad_norm": 1.0546875,
+      "learning_rate": 9.486591442944313e-06,
+      "loss": 1.1748,
+      "step": 1050
+    },
+    {
+      "epoch": 7.445127304653204,
+      "grad_norm": 1.1484375,
+      "learning_rate": 9.469503346877569e-06,
+      "loss": 1.1092,
+      "step": 1060
+    },
+    {
+      "epoch": 7.515364354697103,
+      "grad_norm": 0.9296875,
+      "learning_rate": 9.452151421735846e-06,
+      "loss": 1.1099,
+      "step": 1070
+    },
+    {
+      "epoch": 7.585601404741001,
+      "grad_norm": 0.99609375,
+      "learning_rate": 9.434536691781125e-06,
+      "loss": 1.0662,
+      "step": 1080
+    },
+    {
+      "epoch": 7.655838454784899,
+      "grad_norm": 1.046875,
+      "learning_rate": 9.416660196788423e-06,
+      "loss": 1.0551,
+      "step": 1090
+    },
+    {
+      "epoch": 7.726075504828797,
+      "grad_norm": 0.94921875,
+      "learning_rate": 9.39852299198441e-06,
+      "loss": 1.0643,
+      "step": 1100
+    },
+    {
+      "epoch": 7.796312554872696,
+      "grad_norm": 0.98828125,
+      "learning_rate": 9.380126147985122e-06,
+      "loss": 1.0576,
+      "step": 1110
+    },
+    {
+      "epoch": 7.866549604916593,
+      "grad_norm": 0.95703125,
+      "learning_rate": 9.36147075073277e-06,
+      "loss": 1.0703,
+      "step": 1120
+    },
+    {
+      "epoch": 7.936786654960492,
+      "grad_norm": 0.95703125,
+      "learning_rate": 9.34255790143163e-06,
+      "loss": 1.0488,
+      "step": 1130
+    },
+    {
+      "epoch": 8.00702370500439,
+      "grad_norm": 0.8125,
+      "learning_rate": 9.323388716483046e-06,
+      "loss": 1.0843,
+      "step": 1140
+    },
+    {
+      "epoch": 8.077260755048288,
+      "grad_norm": 0.81640625,
+      "learning_rate": 9.303964327419524e-06,
+      "loss": 1.0359,
+      "step": 1150
+    },
+    {
+      "epoch": 8.147497805092186,
+      "grad_norm": 0.84375,
+      "learning_rate": 9.284285880837947e-06,
+      "loss": 1.0227,
+      "step": 1160
+    },
+    {
+      "epoch": 8.217734855136085,
+      "grad_norm": 0.87109375,
+      "learning_rate": 9.264354538331886e-06,
+      "loss": 0.997,
+      "step": 1170
+    },
+    {
+      "epoch": 8.287971905179983,
+      "grad_norm": 0.875,
+      "learning_rate": 9.244171476423037e-06,
+      "loss": 1.0194,
+      "step": 1180
+    },
+    {
+      "epoch": 8.35820895522388,
+      "grad_norm": 0.86328125,
+      "learning_rate": 9.223737886491771e-06,
+      "loss": 1.0264,
+      "step": 1190
+    },
+    {
+      "epoch": 8.428446005267778,
+      "grad_norm": 0.84375,
+      "learning_rate": 9.203054974706807e-06,
+      "loss": 1.037,
+      "step": 1200
+    },
+    {
+      "epoch": 8.498683055311677,
+      "grad_norm": 0.76171875,
+      "learning_rate": 9.182123961954016e-06,
+      "loss": 1.0202,
+      "step": 1210
+    },
+    {
+      "epoch": 8.568920105355575,
+      "grad_norm": 0.7265625,
+      "learning_rate": 9.160946083764353e-06,
+      "loss": 1.0577,
+      "step": 1220
+    },
+    {
+      "epoch": 8.639157155399474,
+      "grad_norm": 0.75,
+      "learning_rate": 9.13952259024092e-06,
+      "loss": 1.009,
+      "step": 1230
+    },
+    {
+      "epoch": 8.709394205443372,
+      "grad_norm": 0.74609375,
+      "learning_rate": 9.117854745985183e-06,
+      "loss": 0.9763,
+      "step": 1240
+    },
+    {
+      "epoch": 8.779631255487269,
+      "grad_norm": 0.80859375,
+      "learning_rate": 9.095943830022323e-06,
+      "loss": 1.0406,
+      "step": 1250
+    },
+    {
+      "epoch": 8.849868305531167,
+      "grad_norm": 0.8359375,
+      "learning_rate": 9.073791135725722e-06,
+      "loss": 1.0093,
+      "step": 1260
+    },
+    {
+      "epoch": 8.920105355575066,
+      "grad_norm": 0.7890625,
+      "learning_rate": 9.051397970740638e-06,
+      "loss": 1.0146,
+      "step": 1270
+    },
+    {
+      "epoch": 8.990342405618964,
+      "grad_norm": 0.75390625,
+      "learning_rate": 9.028765656907005e-06,
+      "loss": 1.0404,
+      "step": 1280
+    },
+    {
+      "epoch": 9.060579455662863,
+      "grad_norm": 0.75,
+      "learning_rate": 9.005895530181406e-06,
+      "loss": 0.9933,
+      "step": 1290
+    },
+    {
+      "epoch": 9.130816505706761,
+      "grad_norm": 0.78515625,
+      "learning_rate": 8.982788940558216e-06,
+      "loss": 0.9701,
+      "step": 1300
+    },
+    {
+      "epoch": 9.201053555750658,
+      "grad_norm": 0.81640625,
+      "learning_rate": 8.959447251989914e-06,
+      "loss": 0.9554,
+      "step": 1310
+    },
+    {
+      "epoch": 9.271290605794556,
+      "grad_norm": 0.81640625,
+      "learning_rate": 8.935871842306569e-06,
+      "loss": 1.0002,
+      "step": 1320
+    },
+    {
+      "epoch": 9.341527655838455,
+      "grad_norm": 0.83203125,
+      "learning_rate": 8.912064103134505e-06,
+      "loss": 0.999,
+      "step": 1330
+    },
+    {
+      "epoch": 9.411764705882353,
+      "grad_norm": 0.81640625,
+      "learning_rate": 8.888025439814169e-06,
+      "loss": 0.9726,
+      "step": 1340
+    },
+    {
+      "epoch": 9.482001755926252,
+      "grad_norm": 0.80078125,
+      "learning_rate": 8.863757271317154e-06,
+      "loss": 1.0067,
+      "step": 1350
+    },
+    {
+      "epoch": 9.552238805970148,
+      "grad_norm": 0.81640625,
+      "learning_rate": 8.839261030162459e-06,
+      "loss": 1.0181,
+      "step": 1360
+    },
+    {
+      "epoch": 9.622475856014047,
+      "grad_norm": 0.875,
+      "learning_rate": 8.814538162331913e-06,
+      "loss": 0.9688,
+      "step": 1370
+    },
+    {
+      "epoch": 9.692712906057945,
+      "grad_norm": 0.76171875,
+      "learning_rate": 8.789590127184837e-06,
+      "loss": 0.9309,
+      "step": 1380
+    },
+    {
+      "epoch": 9.762949956101844,
+      "grad_norm": 0.7734375,
+      "learning_rate": 8.764418397371888e-06,
+      "loss": 0.987,
+      "step": 1390
+    },
+    {
+      "epoch": 9.833187006145742,
+      "grad_norm": 0.86328125,
+      "learning_rate": 8.739024458748128e-06,
+      "loss": 0.9292,
+      "step": 1400
+    },
+    {
+      "epoch": 9.90342405618964,
+      "grad_norm": 0.87890625,
+      "learning_rate": 8.713409810285327e-06,
+      "loss": 0.9909,
+      "step": 1410
+    },
+    {
+      "epoch": 9.973661106233537,
+      "grad_norm": 0.9296875,
+      "learning_rate": 8.687575963983477e-06,
+      "loss": 0.9914,
+      "step": 1420
+    },
+    {
+      "epoch": 10.043898156277436,
+      "grad_norm": 0.8359375,
+      "learning_rate": 8.661524444781531e-06,
+      "loss": 0.9369,
+      "step": 1430
+    },
+    {
+      "epoch": 10.114135206321334,
+      "grad_norm": 0.87109375,
+      "learning_rate": 8.635256790467402e-06,
+      "loss": 0.9573,
+      "step": 1440
+    },
+    {
+      "epoch": 10.184372256365233,
+      "grad_norm": 0.92578125,
+      "learning_rate": 8.60877455158718e-06,
+      "loss": 0.9491,
+      "step": 1450
+    },
+    {
+      "epoch": 10.254609306409131,
+      "grad_norm": 0.96484375,
+      "learning_rate": 8.582079291353607e-06,
+      "loss": 0.8985,
+      "step": 1460
+    },
+    {
+      "epoch": 10.32484635645303,
+      "grad_norm": 0.95703125,
+      "learning_rate": 8.555172585553804e-06,
+      "loss": 0.913,
+      "step": 1470
+    },
+    {
+      "epoch": 10.395083406496926,
+      "grad_norm": 0.94921875,
+      "learning_rate": 8.528056022456256e-06,
+      "loss": 0.9727,
+      "step": 1480
+    },
+    {
+      "epoch": 10.465320456540825,
+      "grad_norm": 0.9453125,
+      "learning_rate": 8.500731202717056e-06,
+      "loss": 0.9723,
+      "step": 1490
+    },
+    {
+      "epoch": 10.535557506584723,
+      "grad_norm": 1.0859375,
+      "learning_rate": 8.473199739285416e-06,
+      "loss": 0.9626,
+      "step": 1500
+    },
+    {
+      "epoch": 10.605794556628622,
+      "grad_norm": 1.2109375,
+      "learning_rate": 8.445463257308463e-06,
+      "loss": 0.9487,
+      "step": 1510
+    },
+    {
+      "epoch": 10.67603160667252,
+      "grad_norm": 1.1875,
+      "learning_rate": 8.417523394035316e-06,
+      "loss": 0.9832,
+      "step": 1520
+    },
+    {
+      "epoch": 10.746268656716419,
+      "grad_norm": 1.203125,
+      "learning_rate": 8.389381798720417e-06,
+      "loss": 0.9621,
+      "step": 1530
+    },
+    {
+      "epoch": 10.816505706760315,
+      "grad_norm": 1.1875,
+      "learning_rate": 8.361040132526204e-06,
+      "loss": 0.9222,
+      "step": 1540
+    },
+    {
+      "epoch": 10.886742756804214,
+      "grad_norm": 2.109375,
+      "learning_rate": 8.332500068425038e-06,
+      "loss": 0.9183,
+      "step": 1550
+    },
+    {
+      "epoch": 10.956979806848112,
+      "grad_norm": 2.34375,
+      "learning_rate": 8.303763291100459e-06,
+      "loss": 0.9173,
+      "step": 1560
+    },
+    {
+      "epoch": 11.02721685689201,
+      "grad_norm": 2.09375,
+      "learning_rate": 8.274831496847735e-06,
+      "loss": 0.9582,
+      "step": 1570
+    },
+    {
+      "epoch": 11.09745390693591,
+      "grad_norm": 1.828125,
+      "learning_rate": 8.245706393473734e-06,
+      "loss": 0.907,
+      "step": 1580
+    },
+    {
+      "epoch": 11.167690956979808,
+      "grad_norm": 1.8828125,
+      "learning_rate": 8.216389700196116e-06,
+      "loss": 0.9339,
+      "step": 1590
+    },
+    {
+      "epoch": 11.237928007023704,
+      "grad_norm": 4.375,
+      "learning_rate": 8.186883147541846e-06,
+      "loss": 0.9417,
+      "step": 1600
+    },
+    {
+      "epoch": 11.308165057067603,
+      "grad_norm": 6.53125,
+      "learning_rate": 8.157188477245048e-06,
+      "loss": 0.8978,
+      "step": 1610
+    },
+    {
+      "epoch": 11.378402107111501,
+      "grad_norm": 4.625,
+      "learning_rate": 8.12730744214419e-06,
+      "loss": 0.9335,
+      "step": 1620
+    },
+    {
+      "epoch": 11.4486391571554,
+      "grad_norm": 5.6875,
+      "learning_rate": 8.097241806078616e-06,
+      "loss": 0.9623,
+      "step": 1630
+    },
+    {
+      "epoch": 11.518876207199298,
+      "grad_norm": 4.71875,
+      "learning_rate": 8.066993343784427e-06,
+      "loss": 0.9691,
+      "step": 1640
+    },
+    {
+      "epoch": 11.589113257243195,
+      "grad_norm": 8.625,
+      "learning_rate": 8.036563840789726e-06,
+      "loss": 0.9208,
+      "step": 1650
+    },
+    {
+      "epoch": 11.659350307287093,
+      "grad_norm": 9.5625,
+      "learning_rate": 8.005955093309217e-06,
+      "loss": 0.9352,
+      "step": 1660
+    },
+    {
+      "epoch": 11.729587357330992,
+      "grad_norm": 8.625,
+      "learning_rate": 7.975168908138174e-06,
+      "loss": 0.911,
+      "step": 1670
+    },
+    {
+      "epoch": 11.79982440737489,
+      "grad_norm": 10.3125,
+      "learning_rate": 7.944207102545795e-06,
+      "loss": 0.9109,
+      "step": 1680
+    },
+    {
+      "epoch": 11.870061457418789,
+      "grad_norm": 8.375,
+      "learning_rate": 7.913071504167925e-06,
+      "loss": 0.946,
+      "step": 1690
+    },
+    {
+      "epoch": 11.940298507462687,
+      "grad_norm": 2.78125,
+      "learning_rate": 7.881763950899175e-06,
+      "loss": 0.8947,
+      "step": 1700
+    },
+    {
+      "epoch": 12.010535557506584,
+      "grad_norm": 2.28125,
+      "learning_rate": 7.850286290784437e-06,
+      "loss": 0.8951,
+      "step": 1710
+    },
+    {
+      "epoch": 12.080772607550482,
+      "grad_norm": 2.34375,
+      "learning_rate": 7.81864038190979e-06,
+      "loss": 0.893,
+      "step": 1720
+    },
+    {
+      "epoch": 12.151009657594381,
+      "grad_norm": 2.84375,
+      "learning_rate": 7.786828092292821e-06,
+      "loss": 0.8729,
+      "step": 1730
+    },
+    {
+      "epoch": 12.22124670763828,
+      "grad_norm": 2.890625,
+      "learning_rate": 7.754851299772362e-06,
+      "loss": 0.8704,
+      "step": 1740
+    },
+    {
+      "epoch": 12.291483757682178,
+      "grad_norm": 2.171875,
+      "learning_rate": 7.722711891897641e-06,
+      "loss": 0.8639,
+      "step": 1750
+    },
+    {
+      "epoch": 12.361720807726076,
+      "grad_norm": 2.921875,
+      "learning_rate": 7.690411765816864e-06,
+      "loss": 0.8624,
+      "step": 1760
+    },
+    {
+      "epoch": 12.431957857769973,
+      "grad_norm": 2.96875,
+      "learning_rate": 7.657952828165225e-06,
+      "loss": 0.878,
+      "step": 1770
+    },
+    {
+      "epoch": 12.502194907813871,
+      "grad_norm": 2.546875,
+      "learning_rate": 7.625336994952364e-06,
+      "loss": 0.8555,
+      "step": 1780
+    },
+    {
+      "epoch": 12.57243195785777,
+      "grad_norm": 2.265625,
+      "learning_rate": 7.592566191449262e-06,
+      "loss": 0.8889,
+      "step": 1790
+    },
+    {
+      "epoch": 12.642669007901668,
+      "grad_norm": 2.453125,
+      "learning_rate": 7.559642352074606e-06,
+      "loss": 0.8598,
+      "step": 1800
+    },
+    {
+      "epoch": 12.712906057945567,
+      "grad_norm": 2.234375,
+      "learning_rate": 7.526567420280585e-06,
+      "loss": 0.8856,
+      "step": 1810
+    },
+    {
+      "epoch": 12.783143107989464,
+      "grad_norm": 2.25,
+      "learning_rate": 7.4933433484381905e-06,
+      "loss": 0.8677,
+      "step": 1820
+    },
+    {
+      "epoch": 12.853380158033362,
+      "grad_norm": 2.84375,
+      "learning_rate": 7.459972097721954e-06,
+      "loss": 0.9036,
+      "step": 1830
+    },
+    {
+      "epoch": 12.92361720807726,
+      "grad_norm": 2.171875,
+      "learning_rate": 7.4264556379941895e-06,
+      "loss": 0.8607,
+      "step": 1840
+    },
+    {
+      "epoch": 12.993854258121159,
+      "grad_norm": 2.28125,
+      "learning_rate": 7.392795947688715e-06,
+      "loss": 0.8699,
+      "step": 1850
+    },
+    {
+      "epoch": 13.064091308165057,
+      "grad_norm": 2.296875,
+      "learning_rate": 7.3589950136940645e-06,
+      "loss": 0.8343,
+      "step": 1860
+    },
+    {
+      "epoch": 13.134328358208956,
+      "grad_norm": 2.34375,
+      "learning_rate": 7.325054831236211e-06,
+      "loss": 0.8305,
+      "step": 1870
+    },
+    {
+      "epoch": 13.204565408252853,
+      "grad_norm": 2.3125,
+      "learning_rate": 7.2909774037607775e-06,
+      "loss": 0.8039,
+      "step": 1880
+    },
+    {
+      "epoch": 13.274802458296751,
+      "grad_norm": 3.0,
+      "learning_rate": 7.256764742814796e-06,
+      "loss": 0.8274,
+      "step": 1890
+    },
+    {
+      "epoch": 13.34503950834065,
+      "grad_norm": 2.171875,
+      "learning_rate": 7.222418867927948e-06,
+      "loss": 0.8126,
+      "step": 1900
+    },
+    {
+      "epoch": 13.415276558384548,
+      "grad_norm": 2.4375,
+      "learning_rate": 7.187941806493372e-06,
+      "loss": 0.8155,
+      "step": 1910
+    },
+    {
+      "epoch": 13.485513608428446,
+      "grad_norm": 2.125,
+      "learning_rate": 7.153335593647974e-06,
+      "loss": 0.8346,
+      "step": 1920
+    },
+    {
+      "epoch": 13.555750658472345,
+      "grad_norm": 2.234375,
+      "learning_rate": 7.118602272152308e-06,
+      "loss": 0.8275,
+      "step": 1930
+    },
+    {
+      "epoch": 13.625987708516242,
+      "grad_norm": 2.25,
+      "learning_rate": 7.083743892269987e-06,
+      "loss": 0.8215,
+      "step": 1940
+    },
+    {
+      "epoch": 13.69622475856014,
+      "grad_norm": 1.890625,
+      "learning_rate": 7.04876251164666e-06,
+      "loss": 0.8181,
+      "step": 1950
+    },
+    {
+      "epoch": 13.766461808604038,
+      "grad_norm": 2.171875,
+      "learning_rate": 7.013660195188553e-06,
+      "loss": 0.8099,
+      "step": 1960
+    },
+    {
+      "epoch": 13.836698858647937,
+      "grad_norm": 2.125,
+      "learning_rate": 6.978439014940584e-06,
+      "loss": 0.8278,
+      "step": 1970
+    },
+    {
+      "epoch": 13.906935908691835,
+      "grad_norm": 2.21875,
+      "learning_rate": 6.943101049964042e-06,
+      "loss": 0.8299,
+      "step": 1980
+    },
+    {
+      "epoch": 13.977172958735734,
+      "grad_norm": 2.328125,
+      "learning_rate": 6.907648386213875e-06,
+      "loss": 0.8003,
+      "step": 1990
+    },
+    {
+      "epoch": 14.04741000877963,
+      "grad_norm": 2.0625,
+      "learning_rate": 6.872083116415547e-06,
+      "loss": 0.7669,
+      "step": 2000
+    },
+    {
+      "epoch": 14.04741000877963,
+      "eval_loss": 1.4163318872451782,
+      "eval_runtime": 28.9493,
+      "eval_samples_per_second": 8.774,
+      "eval_steps_per_second": 8.774,
+      "step": 2000
+    },
+    {
+      "epoch": 14.117647058823529,
+      "grad_norm": 1.8984375,
+      "learning_rate": 6.836407339941522e-06,
+      "loss": 0.7701,
+      "step": 2010
+    },
+    {
+      "epoch": 14.187884108867427,
+      "grad_norm": 1.921875,
+      "learning_rate": 6.800623162687325e-06,
+      "loss": 0.7634,
+      "step": 2020
+    },
+    {
+      "epoch": 14.258121158911326,
+      "grad_norm": 2.34375,
+      "learning_rate": 6.764732696947243e-06,
+      "loss": 0.7579,
+      "step": 2030
+    },
+    {
+      "epoch": 14.328358208955224,
+      "grad_norm": 2.5,
+      "learning_rate": 6.728738061289634e-06,
+      "loss": 0.7713,
+      "step": 2040
+    },
+    {
+      "epoch": 14.398595258999123,
+      "grad_norm": 1.9453125,
+      "learning_rate": 6.692641380431879e-06,
+      "loss": 0.7512,
+      "step": 2050
+    },
+    {
+      "epoch": 14.46883230904302,
+      "grad_norm": 2.125,
+      "learning_rate": 6.6564447851149505e-06,
+      "loss": 0.7673,
+      "step": 2060
+    },
+    {
+      "epoch": 14.539069359086918,
+      "grad_norm": 2.0,
+      "learning_rate": 6.620150411977648e-06,
+      "loss": 0.747,
+      "step": 2070
+    },
+    {
+      "epoch": 14.609306409130816,
+      "grad_norm": 1.96875,
+      "learning_rate": 6.5837604034304715e-06,
+      "loss": 0.7471,
+      "step": 2080
+    },
+    {
+      "epoch": 14.679543459174715,
+      "grad_norm": 1.9453125,
+      "learning_rate": 6.547276907529152e-06,
+      "loss": 0.7733,
+      "step": 2090
+    },
+    {
+      "epoch": 14.749780509218613,
+      "grad_norm": 1.59375,
+      "learning_rate": 6.510702077847864e-06,
+      "loss": 0.7356,
+      "step": 2100
+    },
+    {
+      "epoch": 14.82001755926251,
+      "grad_norm": 1.9921875,
+      "learning_rate": 6.474038073352098e-06,
+      "loss": 0.7702,
+      "step": 2110
+    },
+    {
+      "epoch": 14.890254609306409,
+      "grad_norm": 1.96875,
+      "learning_rate": 6.4372870582712196e-06,
+      "loss": 0.76,
+      "step": 2120
+    },
+    {
+      "epoch": 14.960491659350307,
+      "grad_norm": 1.8359375,
+      "learning_rate": 6.4004512019707144e-06,
+      "loss": 0.7664,
+      "step": 2130
+    },
+    {
+      "epoch": 15.030728709394205,
+      "grad_norm": 1.5859375,
+      "learning_rate": 6.363532678824145e-06,
+      "loss": 0.7274,
+      "step": 2140
+    },
+    {
+      "epoch": 15.100965759438104,
+      "grad_norm": 1.8828125,
+      "learning_rate": 6.326533668084783e-06,
+      "loss": 0.703,
+      "step": 2150
+    },
+    {
+      "epoch": 15.171202809482002,
+      "grad_norm": 1.9921875,
+      "learning_rate": 6.289456353756988e-06,
+      "loss": 0.7176,
+      "step": 2160
+    },
+    {
+      "epoch": 15.241439859525899,
+      "grad_norm": 1.9453125,
+      "learning_rate": 6.252302924467276e-06,
+      "loss": 0.6589,
+      "step": 2170
+    },
+    {
+      "epoch": 15.311676909569798,
+      "grad_norm": 1.7578125,
+      "learning_rate": 6.2150755733351305e-06,
+      "loss": 0.6822,
+      "step": 2180
+    },
+    {
+      "epoch": 15.381913959613696,
+      "grad_norm": 1.734375,
+      "learning_rate": 6.177776497843552e-06,
+      "loss": 0.6947,
+      "step": 2190
+    },
+    {
+      "epoch": 15.452151009657594,
+      "grad_norm": 1.7578125,
+      "learning_rate": 6.140407899709333e-06,
+      "loss": 0.7027,
+      "step": 2200
+    },
+    {
+      "epoch": 15.522388059701493,
+      "grad_norm": 1.765625,
+      "learning_rate": 6.102971984753104e-06,
+      "loss": 0.6625,
+      "step": 2210
+    },
+    {
+      "epoch": 15.592625109745391,
+      "grad_norm": 1.8359375,
+      "learning_rate": 6.065470962769119e-06,
+      "loss": 0.7181,
+      "step": 2220
+    },
+    {
+      "epoch": 15.662862159789288,
+      "grad_norm": 1.7421875,
+      "learning_rate": 6.027907047394812e-06,
+      "loss": 0.6745,
+      "step": 2230
+    },
+    {
+      "epoch": 15.733099209833187,
+      "grad_norm": 1.6953125,
+      "learning_rate": 5.990282455980145e-06,
+      "loss": 0.6744,
+      "step": 2240
+    },
+    {
+      "epoch": 15.803336259877085,
+      "grad_norm": 1.9609375,
+      "learning_rate": 5.952599409456697e-06,
+      "loss": 0.6877,
+      "step": 2250
+    },
+    {
+      "epoch": 15.873573309920983,
+      "grad_norm": 1.765625,
+      "learning_rate": 5.914860132206584e-06,
+      "loss": 0.7303,
+      "step": 2260
+    },
+    {
+      "epoch": 15.943810359964882,
+      "grad_norm": 1.71875,
+      "learning_rate": 5.877066851931151e-06,
+      "loss": 0.712,
+      "step": 2270
+    },
+    {
+      "epoch": 16.01404741000878,
+      "grad_norm": 1.859375,
+      "learning_rate": 5.83922179951947e-06,
+      "loss": 0.6674,
+      "step": 2280
+    },
+    {
+      "epoch": 16.08428446005268,
+      "grad_norm": 2.03125,
+      "learning_rate": 5.8013272089166526e-06,
+      "loss": 0.6726,
+      "step": 2290
+    },
+    {
+      "epoch": 16.154521510096576,
+      "grad_norm": 1.7109375,
+      "learning_rate": 5.763385316991995e-06,
+      "loss": 0.6198,
+      "step": 2300
+    },
+    {
+      "epoch": 16.224758560140476,
+      "grad_norm": 1.8046875,
+      "learning_rate": 5.725398363406922e-06,
+      "loss": 0.6223,
+      "step": 2310
+    },
+    {
+      "epoch": 16.294995610184372,
+      "grad_norm": 1.8671875,
+      "learning_rate": 5.687368590482797e-06,
+      "loss": 0.6315,
+      "step": 2320
+    },
+    {
+      "epoch": 16.36523266022827,
+      "grad_norm": 1.78125,
+      "learning_rate": 5.64929824306855e-06,
+      "loss": 0.6418,
+      "step": 2330
+    },
+    {
+      "epoch": 16.43546971027217,
+      "grad_norm": 1.6875,
+      "learning_rate": 5.611189568408173e-06,
+      "loss": 0.6086,
+      "step": 2340
+    },
+    {
+      "epoch": 16.505706760316066,
+      "grad_norm": 1.8515625,
+      "learning_rate": 5.573044816008066e-06,
+      "loss": 0.6375,
+      "step": 2350
+    },
+    {
+      "epoch": 16.575943810359966,
+      "grad_norm": 1.5390625,
+      "learning_rate": 5.534866237504252e-06,
+      "loss": 0.6085,
+      "step": 2360
+    },
+    {
+      "epoch": 16.646180860403863,
+      "grad_norm": 1.7109375,
+      "learning_rate": 5.496656086529467e-06,
+      "loss": 0.6355,
+      "step": 2370
+    },
+    {
+      "epoch": 16.71641791044776,
+      "grad_norm": 1.7109375,
+      "learning_rate": 5.458416618580126e-06,
+      "loss": 0.6206,
+      "step": 2380
+    },
+    {
+      "epoch": 16.78665496049166,
+      "grad_norm": 1.4921875,
+      "learning_rate": 5.420150090883191e-06,
+      "loss": 0.6445,
+      "step": 2390
+    },
+    {
+      "epoch": 16.856892010535557,
+      "grad_norm": 1.546875,
+      "learning_rate": 5.381858762262927e-06,
+      "loss": 0.6508,
+      "step": 2400
+    },
+    {
+      "epoch": 16.927129060579457,
+      "grad_norm": 1.578125,
+      "learning_rate": 5.343544893007563e-06,
+      "loss": 0.6198,
+      "step": 2410
+    },
+    {
+      "epoch": 16.997366110623354,
+      "grad_norm": 1.4921875,
+      "learning_rate": 5.305210744735874e-06,
+      "loss": 0.6156,
+      "step": 2420
+    },
+    {
+      "epoch": 17.06760316066725,
+      "grad_norm": 1.515625,
+      "learning_rate": 5.266858580263678e-06,
+      "loss": 0.5986,
+      "step": 2430
+    },
+    {
+      "epoch": 17.13784021071115,
+      "grad_norm": 1.578125,
+      "learning_rate": 5.228490663470271e-06,
+      "loss": 0.5637,
+      "step": 2440
+    },
+    {
+      "epoch": 17.208077260755047,
+      "grad_norm": 1.390625,
+      "learning_rate": 5.190109259164782e-06,
+      "loss": 0.5738,
+      "step": 2450
+    },
+    {
+      "epoch": 17.278314310798947,
+      "grad_norm": 1.5,
+      "learning_rate": 5.151716632952495e-06,
+      "loss": 0.5912,
+      "step": 2460
+    },
+    {
+      "epoch": 17.348551360842844,
+      "grad_norm": 1.28125,
+      "learning_rate": 5.113315051101111e-06,
+      "loss": 0.5782,
+      "step": 2470
+    },
+    {
+      "epoch": 17.418788410886744,
+      "grad_norm": 1.5546875,
+      "learning_rate": 5.074906780406962e-06,
+      "loss": 0.5548,
+      "step": 2480
+    },
+    {
+      "epoch": 17.48902546093064,
+      "grad_norm": 1.3359375,
+      "learning_rate": 5.036494088061222e-06,
+      "loss": 0.5678,
+      "step": 2490
+    },
+    {
+      "epoch": 17.559262510974538,
+      "grad_norm": 1.1953125,
+      "learning_rate": 4.998079241516068e-06,
+      "loss": 0.5768,
+      "step": 2500
+    },
+    {
+      "epoch": 17.629499561018438,
+      "grad_norm": 1.21875,
+      "learning_rate": 4.959664508350834e-06,
+      "loss": 0.5794,
+      "step": 2510
+    },
+    {
+      "epoch": 17.699736611062335,
+      "grad_norm": 1.234375,
+      "learning_rate": 4.921252156138163e-06,
+      "loss": 0.5754,
+      "step": 2520
+    },
+    {
+      "epoch": 17.769973661106235,
+      "grad_norm": 1.28125,
+      "learning_rate": 4.882844452310155e-06,
+      "loss": 0.5781,
+      "step": 2530
+    },
+    {
+      "epoch": 17.84021071115013,
+      "grad_norm": 1.265625,
+      "learning_rate": 4.844443664024517e-06,
+      "loss": 0.5834,
+      "step": 2540
+    },
+    {
+      "epoch": 17.91044776119403,
+      "grad_norm": 1.15625,
+      "learning_rate": 4.8060520580307456e-06,
+      "loss": 0.565,
+      "step": 2550
+    },
+    {
+      "epoch": 17.98068481123793,
+      "grad_norm": 1.15625,
+      "learning_rate": 4.767671900536315e-06,
+      "loss": 0.6071,
+      "step": 2560
+    },
+    {
+      "epoch": 18.050921861281825,
+      "grad_norm": 1.078125,
+      "learning_rate": 4.729305457072913e-06,
+      "loss": 0.5734,
+      "step": 2570
+    },
+    {
+      "epoch": 18.121158911325725,
+      "grad_norm": 1.1796875,
+      "learning_rate": 4.690954992362699e-06,
+      "loss": 0.5322,
+      "step": 2580
+    },
+    {
+      "epoch": 18.191395961369622,
+      "grad_norm": 1.171875,
+      "learning_rate": 4.652622770184637e-06,
+      "loss": 0.5304,
+      "step": 2590
+    },
+    {
+      "epoch": 18.261633011413522,
+      "grad_norm": 1.0859375,
+      "learning_rate": 4.6143110532408455e-06,
+      "loss": 0.5368,
+      "step": 2600
+    },
+    {
+      "epoch": 18.33187006145742,
+      "grad_norm": 1.1484375,
+      "learning_rate": 4.576022103023053e-06,
+      "loss": 0.5456,
+      "step": 2610
+    },
+    {
+      "epoch": 18.402107111501316,
+      "grad_norm": 1.0859375,
+      "learning_rate": 4.537758179679098e-06,
+      "loss": 0.5699,
+      "step": 2620
+    },
+    {
+      "epoch": 18.472344161545216,
+      "grad_norm": 1.0859375,
+      "learning_rate": 4.499521541879508e-06,
+      "loss": 0.5587,
+      "step": 2630
+    },
+    {
+      "epoch": 18.542581211589113,
+      "grad_norm": 1.1328125,
+      "learning_rate": 4.461314446684189e-06,
+      "loss": 0.526,
+      "step": 2640
+    },
+    {
+      "epoch": 18.612818261633013,
+      "grad_norm": 1.0078125,
+      "learning_rate": 4.423139149409176e-06,
+      "loss": 0.5593,
+      "step": 2650
+    },
+    {
+      "epoch": 18.68305531167691,
+      "grad_norm": 1.0703125,
+      "learning_rate": 4.384997903493519e-06,
+      "loss": 0.5379,
+      "step": 2660
+    },
+    {
+      "epoch": 18.753292361720806,
+      "grad_norm": 1.078125,
+      "learning_rate": 4.346892960366255e-06,
+      "loss": 0.5503,
+      "step": 2670
+    },
+    {
+      "epoch": 18.823529411764707,
+      "grad_norm": 1.03125,
+      "learning_rate": 4.30882656931352e-06,
+      "loss": 0.5275,
+      "step": 2680
+    },
+    {
+      "epoch": 18.893766461808603,
+      "grad_norm": 1.1484375,
+      "learning_rate": 4.270800977345767e-06,
+      "loss": 0.5515,
+      "step": 2690
+    },
+    {
+      "epoch": 18.964003511852503,
+      "grad_norm": 1.0234375,
+      "learning_rate": 4.232818429065128e-06,
+      "loss": 0.5484,
+      "step": 2700
+    },
+    {
+      "epoch": 19.0342405618964,
+      "grad_norm": 0.9140625,
+      "learning_rate": 4.194881166532923e-06,
+      "loss": 0.5451,
+      "step": 2710
+    },
+    {
+      "epoch": 19.104477611940297,
+      "grad_norm": 1.015625,
+      "learning_rate": 4.156991429137317e-06,
+      "loss": 0.5222,
+      "step": 2720
+    },
+    {
+      "epoch": 19.174714661984197,
+      "grad_norm": 0.9921875,
+      "learning_rate": 4.119151453461121e-06,
+      "loss": 0.5167,
+      "step": 2730
+    },
+    {
+      "epoch": 19.244951712028094,
+      "grad_norm": 1.0078125,
+      "learning_rate": 4.081363473149778e-06,
+      "loss": 0.5009,
+      "step": 2740
+    },
+    {
+      "epoch": 19.315188762071994,
+      "grad_norm": 0.94140625,
+      "learning_rate": 4.0436297187795085e-06,
+      "loss": 0.5206,
+      "step": 2750
+    },
+    {
+      "epoch": 19.38542581211589,
+      "grad_norm": 0.96484375,
+      "learning_rate": 4.005952417725649e-06,
+      "loss": 0.5241,
+      "step": 2760
+    },
+    {
+      "epoch": 19.45566286215979,
+      "grad_norm": 0.91796875,
+      "learning_rate": 3.968333794031165e-06,
+      "loss": 0.5247,
+      "step": 2770
+    },
+    {
+      "epoch": 19.525899912203688,
+      "grad_norm": 0.8671875,
+      "learning_rate": 3.930776068275375e-06,
+      "loss": 0.5138,
+      "step": 2780
+    },
+    {
+      "epoch": 19.596136962247584,
+      "grad_norm": 0.9453125,
+      "learning_rate": 3.89328145744287e-06,
+      "loss": 0.5295,
+      "step": 2790
+    },
+    {
+      "epoch": 19.666374012291485,
+      "grad_norm": 0.88671875,
+      "learning_rate": 3.8558521747926434e-06,
+      "loss": 0.5235,
+      "step": 2800
+    },
+    {
+      "epoch": 19.73661106233538,
+      "grad_norm": 0.87890625,
+      "learning_rate": 3.818490429727455e-06,
+      "loss": 0.5198,
+      "step": 2810
+    },
+    {
+      "epoch": 19.80684811237928,
+      "grad_norm": 0.90234375,
+      "learning_rate": 3.7811984276634024e-06,
+      "loss": 0.5037,
+      "step": 2820
+    },
+    {
+      "epoch": 19.877085162423178,
+      "grad_norm": 0.82421875,
+      "learning_rate": 3.743978369899748e-06,
+      "loss": 0.5069,
+      "step": 2830
+    },
+    {
+      "epoch": 19.947322212467075,
+      "grad_norm": 0.90625,
+      "learning_rate": 3.70683245348897e-06,
+      "loss": 0.5333,
+      "step": 2840
+    },
+    {
+      "epoch": 20.017559262510975,
+      "grad_norm": 0.7890625,
+      "learning_rate": 3.6697628711070786e-06,
+      "loss": 0.4939,
+      "step": 2850
+    },
+    {
+      "epoch": 20.08779631255487,
+      "grad_norm": 0.78125,
+      "learning_rate": 3.632771810924184e-06,
+      "loss": 0.4879,
+      "step": 2860
+    },
+    {
+      "epoch": 20.158033362598772,
+      "grad_norm": 0.82421875,
+      "learning_rate": 3.5958614564753313e-06,
+      "loss": 0.518,
+      "step": 2870
+    },
+    {
+      "epoch": 20.22827041264267,
+      "grad_norm": 0.79296875,
+      "learning_rate": 3.559033986531608e-06,
+      "loss": 0.4944,
+      "step": 2880
+    },
+    {
+      "epoch": 20.298507462686565,
+      "grad_norm": 0.8125,
+      "learning_rate": 3.522291574971538e-06,
+      "loss": 0.5026,
+      "step": 2890
+    },
+    {
+      "epoch": 20.368744512730466,
+      "grad_norm": 0.8359375,
+      "learning_rate": 3.4856363906527513e-06,
+      "loss": 0.5134,
+      "step": 2900
+    },
+    {
+      "epoch": 20.438981562774362,
+      "grad_norm": 0.69140625,
+      "learning_rate": 3.449070597283972e-06,
+      "loss": 0.5056,
+      "step": 2910
+    },
+    {
+      "epoch": 20.509218612818263,
+      "grad_norm": 0.8828125,
+      "learning_rate": 3.4125963532972878e-06,
+      "loss": 0.5035,
+      "step": 2920
+    },
+    {
+      "epoch": 20.57945566286216,
+      "grad_norm": 0.82421875,
+      "learning_rate": 3.376215811720744e-06,
+      "loss": 0.5134,
+      "step": 2930
+    },
+    {
+      "epoch": 20.64969271290606,
+      "grad_norm": 0.81640625,
+      "learning_rate": 3.3399311200512495e-06,
+      "loss": 0.4666,
+      "step": 2940
+    },
+    {
+      "epoch": 20.719929762949956,
+      "grad_norm": 0.78125,
+      "learning_rate": 3.3037444201278202e-06,
+      "loss": 0.4965,
+      "step": 2950
+    },
+    {
+      "epoch": 20.790166812993853,
+      "grad_norm": 0.80078125,
+      "learning_rate": 3.267657848005139e-06,
+      "loss": 0.4953,
+      "step": 2960
+    },
+    {
+      "epoch": 20.860403863037753,
+      "grad_norm": 0.8125,
+      "learning_rate": 3.2316735338274795e-06,
+      "loss": 0.4914,
+      "step": 2970
+    },
+    {
+      "epoch": 20.93064091308165,
+      "grad_norm": 0.83203125,
+      "learning_rate": 3.1957936017029513e-06,
+      "loss": 0.5002,
+      "step": 2980
+    },
+    {
+      "epoch": 21.00087796312555,
+      "grad_norm": 0.79296875,
+      "learning_rate": 3.1600201695781335e-06,
+      "loss": 0.5163,
+      "step": 2990
+    },
+    {
+      "epoch": 21.071115013169447,
+      "grad_norm": 0.87109375,
+      "learning_rate": 3.124355349113037e-06,
+      "loss": 0.4863,
+      "step": 3000
+    },
+    {
+      "epoch": 21.141352063213343,
+      "grad_norm": 0.84375,
+      "learning_rate": 3.0888012455564707e-06,
+      "loss": 0.4616,
+      "step": 3010
+    },
+    {
+      "epoch": 21.211589113257244,
+      "grad_norm": 0.91796875,
+      "learning_rate": 3.0533599576217664e-06,
+      "loss": 0.4897,
+      "step": 3020
+    },
+    {
+      "epoch": 21.28182616330114,
+      "grad_norm": 0.87109375,
+      "learning_rate": 3.0180335773628912e-06,
+      "loss": 0.4801,
+      "step": 3030
+    },
+    {
+      "epoch": 21.35206321334504,
+      "grad_norm": 0.8984375,
+      "learning_rate": 2.982824190050958e-06,
+      "loss": 0.5161,
+      "step": 3040
+    },
+    {
+      "epoch": 21.422300263388937,
+      "grad_norm": 0.84375,
+      "learning_rate": 2.94773387405114e-06,
+      "loss": 0.4854,
+      "step": 3050
+    },
+    {
+      "epoch": 21.492537313432837,
+      "grad_norm": 0.953125,
+      "learning_rate": 2.912764700699978e-06,
+      "loss": 0.5062,
+      "step": 3060
+    },
+    {
+      "epoch": 21.562774363476734,
+      "grad_norm": 1.0234375,
+      "learning_rate": 2.8779187341831205e-06,
+      "loss": 0.4976,
+      "step": 3070
+    },
+    {
+      "epoch": 21.63301141352063,
+      "grad_norm": 1.046875,
+      "learning_rate": 2.843198031413473e-06,
+      "loss": 0.4782,
+      "step": 3080
+    },
+    {
+      "epoch": 21.70324846356453,
+      "grad_norm": 1.09375,
+      "learning_rate": 2.808604641909781e-06,
+      "loss": 0.4694,
+      "step": 3090
+    },
+    {
+      "epoch": 21.773485513608428,
+      "grad_norm": 1.0078125,
+      "learning_rate": 2.7741406076756484e-06,
+      "loss": 0.4864,
+      "step": 3100
+    },
+    {
+      "epoch": 21.843722563652328,
+      "grad_norm": 1.03125,
+      "learning_rate": 2.7398079630790064e-06,
+      "loss": 0.4775,
+      "step": 3110
+    },
+    {
+      "epoch": 21.913959613696225,
+      "grad_norm": 1.0703125,
+      "learning_rate": 2.7056087347320238e-06,
+      "loss": 0.4923,
+      "step": 3120
+    },
+    {
+      "epoch": 21.98419666374012,
+      "grad_norm": 1.015625,
+      "learning_rate": 2.6715449413714778e-06,
+      "loss": 0.4862,
+      "step": 3130
+    },
+    {
+      "epoch": 22.05443371378402,
+      "grad_norm": 0.9921875,
+      "learning_rate": 2.637618593739588e-06,
+      "loss": 0.4765,
+      "step": 3140
+    },
+    {
+      "epoch": 22.12467076382792,
+      "grad_norm": 1.453125,
+      "learning_rate": 2.603831694465333e-06,
+      "loss": 0.4581,
+      "step": 3150
+    },
+    {
+      "epoch": 22.19490781387182,
+      "grad_norm": 1.5390625,
+      "learning_rate": 2.57018623794623e-06,
+      "loss": 0.479,
+      "step": 3160
+    },
+    {
+      "epoch": 22.265144863915715,
+      "grad_norm": 1.4375,
+      "learning_rate": 2.5366842102306144e-06,
+      "loss": 0.4777,
+      "step": 3170
+    },
+    {
+      "epoch": 22.335381913959615,
+      "grad_norm": 1.4921875,
+      "learning_rate": 2.503327588900396e-06,
+      "loss": 0.4865,
+      "step": 3180
+    },
+    {
+      "epoch": 22.405618964003512,
+      "grad_norm": 1.28125,
+      "learning_rate": 2.4701183429543386e-06,
+      "loss": 0.4691,
+      "step": 3190
+    },
+    {
+      "epoch": 22.47585601404741,
+      "grad_norm": 4.15625,
+      "learning_rate": 2.437058432691819e-06,
+      "loss": 0.4755,
+      "step": 3200
+    },
+    {
+      "epoch": 22.54609306409131,
+      "grad_norm": 4.53125,
+      "learning_rate": 2.4041498095971253e-06,
+      "loss": 0.46,
+      "step": 3210
+    },
+    {
+      "epoch": 22.616330114135206,
+      "grad_norm": 4.09375,
+      "learning_rate": 2.3713944162242506e-06,
+      "loss": 0.4886,
+      "step": 3220
+    },
+    {
+      "epoch": 22.686567164179106,
+      "grad_norm": 4.65625,
+      "learning_rate": 2.3387941860822395e-06,
+      "loss": 0.488,
+      "step": 3230
+    },
+    {
+      "epoch": 22.756804214223003,
+      "grad_norm": 4.4375,
+      "learning_rate": 2.3063510435210456e-06,
+      "loss": 0.4788,
+      "step": 3240
+    },
+    {
+      "epoch": 22.8270412642669,
+      "grad_norm": 8.6875,
+      "learning_rate": 2.2740669036179464e-06,
+      "loss": 0.4856,
+      "step": 3250
+    },
+    {
+      "epoch": 22.8972783143108,
+      "grad_norm": 8.5,
+      "learning_rate": 2.24194367206449e-06,
+      "loss": 0.4672,
+      "step": 3260
+    },
+    {
+      "epoch": 22.967515364354696,
+      "grad_norm": 9.875,
+      "learning_rate": 2.209983245054014e-06,
+      "loss": 0.4663,
+      "step": 3270
+    },
+    {
+      "epoch": 23.037752414398597,
+      "grad_norm": 8.6875,
+      "learning_rate": 2.178187509169713e-06,
+      "loss": 0.465,
+      "step": 3280
+    },
+    {
+      "epoch": 23.107989464442493,
+      "grad_norm": 10.375,
+      "learning_rate": 2.146558341273273e-06,
+      "loss": 0.4592,
+      "step": 3290
+    },
+    {
+      "epoch": 23.17822651448639,
+      "grad_norm": 2.65625,
+      "learning_rate": 2.115097608394084e-06,
+      "loss": 0.4521,
+      "step": 3300
+    },
+    {
+      "epoch": 23.24846356453029,
+      "grad_norm": 3.0,
+      "learning_rate": 2.083807167619029e-06,
+      "loss": 0.4396,
+      "step": 3310
+    },
+    {
+      "epoch": 23.318700614574187,
+      "grad_norm": 2.859375,
+      "learning_rate": 2.0526888659828716e-06,
+      "loss": 0.4557,
+      "step": 3320
+    },
+    {
+      "epoch": 23.388937664618087,
+      "grad_norm": 2.828125,
+      "learning_rate": 2.0217445403592185e-06,
+      "loss": 0.4686,
+      "step": 3330
+    },
+    {
+      "epoch": 23.459174714661984,
+      "grad_norm": 2.4375,
+      "learning_rate": 1.990976017352097e-06,
+      "loss": 0.4696,
+      "step": 3340
+    },
+    {
+      "epoch": 23.529411764705884,
+      "grad_norm": 2.625,
+      "learning_rate": 1.9603851131881256e-06,
+      "loss": 0.4566,
+      "step": 3350
+    },
+    {
+      "epoch": 23.59964881474978,
+      "grad_norm": 2.9375,
+      "learning_rate": 1.9299736336093137e-06,
+      "loss": 0.4642,
+      "step": 3360
+    },
+    {
+      "epoch": 23.669885864793677,
+      "grad_norm": 2.8125,
+      "learning_rate": 1.8997433737664673e-06,
+      "loss": 0.4371,
+      "step": 3370
+    },
+    {
+      "epoch": 23.740122914837578,
+      "grad_norm": 2.703125,
+      "learning_rate": 1.869696118113216e-06,
+      "loss": 0.4616,
+      "step": 3380
+    },
+    {
+      "epoch": 23.810359964881474,
+      "grad_norm": 2.859375,
+      "learning_rate": 1.8398336403006956e-06,
+      "loss": 0.4551,
+      "step": 3390
+    },
+    {
+      "epoch": 23.880597014925375,
+      "grad_norm": 2.390625,
+      "learning_rate": 1.8101577030728324e-06,
+      "loss": 0.4456,
+      "step": 3400
+    },
+    {
+      "epoch": 23.95083406496927,
+      "grad_norm": 2.625,
+      "learning_rate": 1.7806700581623059e-06,
+      "loss": 0.4756,
+      "step": 3410
+    },
+    {
+      "epoch": 24.021071115013168,
+      "grad_norm": 2.453125,
+      "learning_rate": 1.7513724461871423e-06,
+      "loss": 0.4519,
+      "step": 3420
+    },
+    {
+      "epoch": 24.091308165057068,
+      "grad_norm": 2.484375,
+      "learning_rate": 1.7222665965479585e-06,
+      "loss": 0.4489,
+      "step": 3430
+    },
+    {
+      "epoch": 24.161545215100965,
+      "grad_norm": 2.171875,
+      "learning_rate": 1.6933542273258924e-06,
+      "loss": 0.4369,
+      "step": 3440
+    },
+    {
+      "epoch": 24.231782265144865,
+      "grad_norm": 2.359375,
+      "learning_rate": 1.6646370451811784e-06,
+      "loss": 0.4246,
+      "step": 3450
+    },
+    {
+      "epoch": 24.302019315188762,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.6361167452524073e-06,
+      "loss": 0.4181,
+      "step": 3460
+    },
+    {
+      "epoch": 24.37225636523266,
+      "grad_norm": 2.328125,
+      "learning_rate": 1.6077950110564606e-06,
+      "loss": 0.4349,
+      "step": 3470
+    },
+    {
+      "epoch": 24.44249341527656,
+      "grad_norm": 2.515625,
+      "learning_rate": 1.5796735143891423e-06,
+      "loss": 0.4358,
+      "step": 3480
+    },
+    {
+      "epoch": 24.512730465320455,
+      "grad_norm": 2.71875,
+      "learning_rate": 1.551753915226491e-06,
+      "loss": 0.4428,
+      "step": 3490
+    },
+    {
+      "epoch": 24.582967515364356,
+      "grad_norm": 2.5,
+      "learning_rate": 1.5240378616267887e-06,
+      "loss": 0.4538,
+      "step": 3500
+    },
+    {
+      "epoch": 24.653204565408252,
+      "grad_norm": 2.234375,
+      "learning_rate": 1.4965269896332884e-06,
+      "loss": 0.4311,
+      "step": 3510
+    },
+    {
+      "epoch": 24.723441615452153,
+      "grad_norm": 2.140625,
+      "learning_rate": 1.46922292317763e-06,
+      "loss": 0.4664,
+      "step": 3520
+    },
+    {
+      "epoch": 24.79367866549605,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.4421272739839898e-06,
+      "loss": 0.4315,
+      "step": 3530
+    },
+    {
+      "epoch": 24.863915715539946,
+      "grad_norm": 2.234375,
+      "learning_rate": 1.4152416414739401e-06,
+      "loss": 0.4403,
+      "step": 3540
+    },
+    {
+      "epoch": 24.934152765583846,
+      "grad_norm": 2.34375,
+      "learning_rate": 1.3885676126720315e-06,
+      "loss": 0.4652,
+      "step": 3550
+    },
+    {
+      "epoch": 25.004389815627743,
+      "grad_norm": 2.140625,
+      "learning_rate": 1.362106762112123e-06,
+      "loss": 0.4439,
+      "step": 3560
+    },
+    {
+      "epoch": 25.074626865671643,
+      "grad_norm": 2.3125,
+      "learning_rate": 1.3358606517444328e-06,
+      "loss": 0.443,
+      "step": 3570
+    },
+    {
+      "epoch": 25.14486391571554,
+      "grad_norm": 2.0,
+      "learning_rate": 1.3098308308433411e-06,
+      "loss": 0.416,
+      "step": 3580
+    },
+    {
+      "epoch": 25.215100965759436,
+      "grad_norm": 2.171875,
+      "learning_rate": 1.2840188359159329e-06,
+      "loss": 0.4035,
+      "step": 3590
+    },
+    {
+      "epoch": 25.285338015803337,
+      "grad_norm": 1.9765625,
+      "learning_rate": 1.258426190611306e-06,
+      "loss": 0.4279,
+      "step": 3600
+    },
+    {
+      "epoch": 25.355575065847233,
+      "grad_norm": 2.171875,
+      "learning_rate": 1.2330544056306315e-06,
+      "loss": 0.4358,
+      "step": 3610
+    },
+    {
+      "epoch": 25.425812115891134,
+      "grad_norm": 1.9765625,
+      "learning_rate": 1.2079049786379782e-06,
+      "loss": 0.4368,
+      "step": 3620
+    },
+    {
+      "epoch": 25.49604916593503,
+      "grad_norm": 2.109375,
+      "learning_rate": 1.1829793941719053e-06,
+      "loss": 0.4271,
+      "step": 3630
+    },
+    {
+      "epoch": 25.56628621597893,
+      "grad_norm": 2.078125,
+      "learning_rate": 1.1582791235578321e-06,
+      "loss": 0.4482,
+      "step": 3640
+    },
+    {
+      "epoch": 25.636523266022827,
+      "grad_norm": 2.140625,
+      "learning_rate": 1.1338056248211916e-06,
+      "loss": 0.4131,
+      "step": 3650
+    },
+    {
+      "epoch": 25.706760316066724,
+      "grad_norm": 2.125,
+      "learning_rate": 1.1095603426013613e-06,
+      "loss": 0.4513,
+      "step": 3660
+    },
+    {
+      "epoch": 25.776997366110624,
+      "grad_norm": 2.15625,
+      "learning_rate": 1.0855447080663907e-06,
+      "loss": 0.4118,
+      "step": 3670
+    },
+    {
+      "epoch": 25.84723441615452,
+      "grad_norm": 2.1875,
+      "learning_rate": 1.0617601388285149e-06,
+      "loss": 0.4082,
+      "step": 3680
+    },
+    {
+      "epoch": 25.91747146619842,
+      "grad_norm": 2.421875,
+      "learning_rate": 1.0382080388604866e-06,
+      "loss": 0.428,
+      "step": 3690
+    },
+    {
+      "epoch": 25.987708516242318,
+      "grad_norm": 2.28125,
+      "learning_rate": 1.0148897984126876e-06,
+      "loss": 0.452,
+      "step": 3700
+    },
+    {
+      "epoch": 26.057945566286215,
+      "grad_norm": 2.0625,
+      "learning_rate": 9.918067939310766e-07,
+      "loss": 0.4177,
+      "step": 3710
+    },
+    {
+      "epoch": 26.128182616330115,
+      "grad_norm": 2.046875,
+      "learning_rate": 9.689603879759284e-07,
+      "loss": 0.4091,
+      "step": 3720
+    },
+    {
+      "epoch": 26.19841966637401,
+      "grad_norm": 1.90625,
+      "learning_rate": 9.463519291414131e-07,
+      "loss": 0.4035,
+      "step": 3730
+    },
+    {
+      "epoch": 26.26865671641791,
+      "grad_norm": 1.7578125,
+      "learning_rate": 9.239827519759842e-07,
+      "loss": 0.41,
+      "step": 3740
+    },
+    {
+      "epoch": 26.33889376646181,
+      "grad_norm": 2.25,
+      "learning_rate": 9.018541769036054e-07,
+      "loss": 0.4306,
+      "step": 3750
+    },
+    {
+      "epoch": 26.409130816505705,
+      "grad_norm": 2.015625,
+      "learning_rate": 8.799675101458033e-07,
+      "loss": 0.4122,
+      "step": 3760
+    },
+    {
+      "epoch": 26.479367866549605,
+      "grad_norm": 1.8984375,
+      "learning_rate": 8.583240436445666e-07,
+      "loss": 0.3986,
+      "step": 3770
+    },
+    {
+      "epoch": 26.549604916593502,
+      "grad_norm": 2.046875,
+      "learning_rate": 8.369250549860869e-07,
+      "loss": 0.437,
+      "step": 3780
+    },
+    {
+      "epoch": 26.619841966637402,
+      "grad_norm": 2.328125,
+      "learning_rate": 8.157718073253351e-07,
+      "loss": 0.4344,
+      "step": 3790
+    },
+    {
+      "epoch": 26.6900790166813,
+      "grad_norm": 1.8828125,
+      "learning_rate": 7.948655493115098e-07,
+      "loss": 0.4269,
+      "step": 3800
+    },
+    {
+      "epoch": 26.7603160667252,
+      "grad_norm": 1.9765625,
+      "learning_rate": 7.742075150143225e-07,
+      "loss": 0.3962,
+      "step": 3810
+    },
+    {
+      "epoch": 26.830553116769096,
+      "grad_norm": 1.9765625,
+      "learning_rate": 7.537989238511578e-07,
+      "loss": 0.4376,
+      "step": 3820
+    },
+    {
+      "epoch": 26.900790166812993,
+      "grad_norm": 1.734375,
+      "learning_rate": 7.336409805150901e-07,
+      "loss": 0.4504,
+      "step": 3830
+    },
+    {
+      "epoch": 26.971027216856893,
+      "grad_norm": 1.984375,
+      "learning_rate": 7.137348749037748e-07,
+      "loss": 0.4166,
+      "step": 3840
+    },
+    {
+      "epoch": 27.04126426690079,
+      "grad_norm": 1.9453125,
+      "learning_rate": 6.940817820492024e-07,
+      "loss": 0.4244,
+      "step": 3850
+    },
+    {
+      "epoch": 27.11150131694469,
+      "grad_norm": 1.9609375,
+      "learning_rate": 6.746828620483487e-07,
+      "loss": 0.4096,
+      "step": 3860
+    },
+    {
+      "epoch": 27.181738366988586,
+      "grad_norm": 1.78125,
+      "learning_rate": 6.555392599946903e-07,
+      "loss": 0.3896,
+      "step": 3870
+    },
+    {
+      "epoch": 27.251975417032483,
+      "grad_norm": 1.9375,
+      "learning_rate": 6.366521059106078e-07,
+      "loss": 0.4363,
+      "step": 3880
+    },
+    {
+      "epoch": 27.322212467076383,
+      "grad_norm": 1.9296875,
+      "learning_rate": 6.180225146806878e-07,
+      "loss": 0.4183,
+      "step": 3890
+    },
+    {
+      "epoch": 27.39244951712028,
+      "grad_norm": 1.8125,
+      "learning_rate": 5.996515859859109e-07,
+      "loss": 0.3911,
+      "step": 3900
+    },
+    {
+      "epoch": 27.46268656716418,
+      "grad_norm": 1.8671875,
+      "learning_rate": 5.815404042387379e-07,
+      "loss": 0.4237,
+      "step": 3910
+    },
+    {
+      "epoch": 27.532923617208077,
+      "grad_norm": 1.78125,
+      "learning_rate": 5.636900385191014e-07,
+      "loss": 0.4077,
+      "step": 3920
+    },
+    {
+      "epoch": 27.603160667251977,
+      "grad_norm": 1.859375,
+      "learning_rate": 5.461015425112915e-07,
+      "loss": 0.407,
+      "step": 3930
+    },
+    {
+      "epoch": 27.673397717295874,
+      "grad_norm": 2.015625,
+      "learning_rate": 5.287759544417687e-07,
+      "loss": 0.436,
+      "step": 3940
+    },
+    {
+      "epoch": 27.74363476733977,
+      "grad_norm": 1.6953125,
+      "learning_rate": 5.117142970178712e-07,
+      "loss": 0.4014,
+      "step": 3950
+    },
+    {
+      "epoch": 27.81387181738367,
+      "grad_norm": 1.6640625,
+      "learning_rate": 4.949175773674502e-07,
+      "loss": 0.4272,
+      "step": 3960
+    },
+    {
+      "epoch": 27.884108867427567,
+      "grad_norm": 1.640625,
+      "learning_rate": 4.783867869794157e-07,
+      "loss": 0.4179,
+      "step": 3970
+    },
+    {
+      "epoch": 27.954345917471468,
+      "grad_norm": 1.75,
+      "learning_rate": 4.6212290164521554e-07,
+      "loss": 0.4351,
+      "step": 3980
+    },
+    {
+      "epoch": 28.024582967515364,
+      "grad_norm": 1.609375,
+      "learning_rate": 4.461268814012304e-07,
+      "loss": 0.3982,
+      "step": 3990
+    },
+    {
+      "epoch": 28.09482001755926,
+      "grad_norm": 1.4921875,
+      "learning_rate": 4.3039967047210865e-07,
+      "loss": 0.4249,
+      "step": 4000
+    },
+    {
+      "epoch": 28.09482001755926,
+      "eval_loss": 1.6928666830062866,
+      "eval_runtime": 10.5811,
+      "eval_samples_per_second": 24.005,
+      "eval_steps_per_second": 24.005,
+      "step": 4000
+    },
+    {
+      "epoch": 28.16505706760316,
+      "grad_norm": 1.515625,
+      "learning_rate": 4.1494219721502917e-07,
+      "loss": 0.406,
+      "step": 4010
+    },
+    {
+      "epoch": 28.235294117647058,
+      "grad_norm": 1.5703125,
+      "learning_rate": 3.997553740648974e-07,
+      "loss": 0.4208,
+      "step": 4020
+    },
+    {
+      "epoch": 28.30553116769096,
+      "grad_norm": 1.625,
+      "learning_rate": 3.8484009748049053e-07,
+      "loss": 0.4113,
+      "step": 4030
+    },
+    {
+      "epoch": 28.375768217734855,
+      "grad_norm": 1.546875,
+      "learning_rate": 3.7019724789154e-07,
+      "loss": 0.4131,
+      "step": 4040
+    },
+    {
+      "epoch": 28.44600526777875,
+      "grad_norm": 1.4453125,
+      "learning_rate": 3.558276896467555e-07,
+      "loss": 0.4192,
+      "step": 4050
+    },
+    {
+      "epoch": 28.516242317822652,
+      "grad_norm": 1.46875,
+      "learning_rate": 3.4173227096281124e-07,
+      "loss": 0.3869,
+      "step": 4060
+    },
+    {
+      "epoch": 28.58647936786655,
+      "grad_norm": 1.4609375,
+      "learning_rate": 3.279118238742729e-07,
+      "loss": 0.4353,
+      "step": 4070
+    },
+    {
+      "epoch": 28.65671641791045,
+      "grad_norm": 1.4453125,
+      "learning_rate": 3.143671641844831e-07,
+      "loss": 0.3981,
+      "step": 4080
+    },
+    {
+      "epoch": 28.726953467954345,
+      "grad_norm": 1.4765625,
+      "learning_rate": 3.0109909141740614e-07,
+      "loss": 0.3976,
+      "step": 4090
+    },
+    {
+      "epoch": 28.797190517998246,
+      "grad_norm": 1.28125,
+      "learning_rate": 2.881083887704339e-07,
+      "loss": 0.4075,
+      "step": 4100
+    },
+    {
+      "epoch": 28.867427568042142,
+      "grad_norm": 1.2421875,
+      "learning_rate": 2.753958230681547e-07,
+      "loss": 0.4224,
+      "step": 4110
+    },
+    {
+      "epoch": 28.93766461808604,
+      "grad_norm": 1.3125,
+      "learning_rate": 2.6296214471708826e-07,
+      "loss": 0.4291,
+      "step": 4120
+    },
+    {
+      "epoch": 29.00790166812994,
+      "grad_norm": 1.375,
+      "learning_rate": 2.5080808766138996e-07,
+      "loss": 0.4209,
+      "step": 4130
+    },
+    {
+      "epoch": 29.078138718173836,
+      "grad_norm": 1.2578125,
+      "learning_rate": 2.3893436933952575e-07,
+      "loss": 0.426,
+      "step": 4140
+    },
+    {
+      "epoch": 29.148375768217736,
+      "grad_norm": 1.109375,
+      "learning_rate": 2.2734169064192623e-07,
+      "loss": 0.4265,
+      "step": 4150
+    },
+    {
+      "epoch": 29.218612818261633,
+      "grad_norm": 1.234375,
+      "learning_rate": 2.1603073586961067e-07,
+      "loss": 0.3948,
+      "step": 4160
+    },
+    {
+      "epoch": 29.28884986830553,
+      "grad_norm": 1.2734375,
+      "learning_rate": 2.0500217269379618e-07,
+      "loss": 0.4023,
+      "step": 4170
+    },
+    {
+      "epoch": 29.35908691834943,
+      "grad_norm": 1.171875,
+      "learning_rate": 1.9425665211648238e-07,
+      "loss": 0.4033,
+      "step": 4180
+    },
+    {
+      "epoch": 29.429323968393327,
+      "grad_norm": 1.15625,
+      "learning_rate": 1.837948084320268e-07,
+      "loss": 0.4199,
+      "step": 4190
+    },
+    {
+      "epoch": 29.499561018437227,
+      "grad_norm": 1.1953125,
+      "learning_rate": 1.736172591897023e-07,
+      "loss": 0.4347,
+      "step": 4200
+    },
+    {
+      "epoch": 29.569798068481123,
+      "grad_norm": 1.1484375,
+      "learning_rate": 1.6372460515724498e-07,
+      "loss": 0.3949,
+      "step": 4210
+    },
+    {
+      "epoch": 29.64003511852502,
+      "grad_norm": 1.140625,
+      "learning_rate": 1.541174302853876e-07,
+      "loss": 0.4254,
+      "step": 4220
+    },
+    {
+      "epoch": 29.71027216856892,
+      "grad_norm": 1.1171875,
+      "learning_rate": 1.4479630167339554e-07,
+      "loss": 0.3997,
+      "step": 4230
+    },
+    {
+      "epoch": 29.780509218612817,
+      "grad_norm": 1.1328125,
+      "learning_rate": 1.3576176953558783e-07,
+      "loss": 0.3875,
+      "step": 4240
+    },
+    {
+      "epoch": 29.850746268656717,
+      "grad_norm": 1.234375,
+      "learning_rate": 1.2701436716885897e-07,
+      "loss": 0.4052,
+      "step": 4250
+    },
+    {
+      "epoch": 29.920983318700614,
+      "grad_norm": 1.0859375,
+      "learning_rate": 1.1855461092119991e-07,
+      "loss": 0.4227,
+      "step": 4260
+    },
+    {
+      "epoch": 29.991220368744514,
+      "grad_norm": 1.1875,
+      "learning_rate": 1.1038300016121883e-07,
+      "loss": 0.4192,
+      "step": 4270
+    },
+    {
+      "epoch": 30.06145741878841,
+      "grad_norm": 1.171875,
+      "learning_rate": 1.025000172486651e-07,
+      "loss": 0.4067,
+      "step": 4280
+    },
+    {
+      "epoch": 30.131694468832308,
+      "grad_norm": 1.1015625,
+      "learning_rate": 9.490612750595096e-08,
+      "loss": 0.4177,
+      "step": 4290
+    },
+    {
+      "epoch": 30.201931518876208,
+      "grad_norm": 0.96875,
+      "learning_rate": 8.760177919069302e-08,
+      "loss": 0.4126,
+      "step": 4300
+    },
+    {
+      "epoch": 30.272168568920105,
+      "grad_norm": 0.984375,
+      "learning_rate": 8.058740346924221e-08,
+      "loss": 0.4098,
+      "step": 4310
+    },
+    {
+      "epoch": 30.342405618964005,
+      "grad_norm": 1.03125,
+      "learning_rate": 7.386341439124145e-08,
+      "loss": 0.4183,
+      "step": 4320
+    },
+    {
+      "epoch": 30.4126426690079,
+      "grad_norm": 0.98828125,
+      "learning_rate": 6.74302088651796e-08,
+      "loss": 0.4249,
+      "step": 4330
+    },
+    {
+      "epoch": 30.482879719051798,
+      "grad_norm": 0.96484375,
+      "learning_rate": 6.128816663496296e-08,
+      "loss": 0.4003,
+      "step": 4340
+    },
+    {
+      "epoch": 30.5531167690957,
+      "grad_norm": 0.984375,
+      "learning_rate": 5.543765025750103e-08,
+      "loss": 0.4085,
+      "step": 4350
+    },
+    {
+      "epoch": 30.623353819139595,
+      "grad_norm": 0.95703125,
+      "learning_rate": 4.987900508130417e-08,
+      "loss": 0.437,
+      "step": 4360
+    },
+    {
+      "epoch": 30.693590869183495,
+      "grad_norm": 0.89453125,
+      "learning_rate": 4.461255922609986e-08,
+      "loss": 0.4117,
+      "step": 4370
+    },
+    {
+      "epoch": 30.763827919227392,
+      "grad_norm": 0.87109375,
+      "learning_rate": 3.963862356346049e-08,
+      "loss": 0.4054,
+      "step": 4380
+    },
+    {
+      "epoch": 30.834064969271292,
+      "grad_norm": 0.93359375,
+      "learning_rate": 3.49574916984563e-08,
+      "loss": 0.3948,
+      "step": 4390
+    },
+    {
+      "epoch": 30.90430201931519,
+      "grad_norm": 0.9453125,
+      "learning_rate": 3.056943995232431e-08,
+      "loss": 0.3932,
+      "step": 4400
+    },
+    {
+      "epoch": 30.974539069359086,
+      "grad_norm": 0.8671875,
+      "learning_rate": 2.6474727346155194e-08,
+      "loss": 0.426,
+      "step": 4410
+    },
+    {
+      "epoch": 31.044776119402986,
+      "grad_norm": 0.89453125,
+      "learning_rate": 2.2673595585605557e-08,
+      "loss": 0.4219,
+      "step": 4420
+    },
+    {
+      "epoch": 31.115013169446883,
+      "grad_norm": 0.85546875,
+      "learning_rate": 1.9166269046628215e-08,
+      "loss": 0.426,
+      "step": 4430
+    },
+    {
+      "epoch": 31.185250219490783,
+      "grad_norm": 0.87890625,
+      "learning_rate": 1.5952954762230575e-08,
+      "loss": 0.4222,
+      "step": 4440
+    },
+    {
+      "epoch": 31.25548726953468,
+      "grad_norm": 0.8359375,
+      "learning_rate": 1.3033842410251074e-08,
+      "loss": 0.4178,
+      "step": 4450
+    },
+    {
+      "epoch": 31.325724319578576,
+      "grad_norm": 0.84375,
+      "learning_rate": 1.0409104302164241e-08,
+      "loss": 0.4068,
+      "step": 4460
+    },
+    {
+      "epoch": 31.395961369622476,
+      "grad_norm": 0.82421875,
+      "learning_rate": 8.078895372908846e-09,
+      "loss": 0.4126,
+      "step": 4470
+    },
+    {
+      "epoch": 31.466198419666373,
+      "grad_norm": 0.83984375,
+      "learning_rate": 6.0433531717424275e-09,
+      "loss": 0.4136,
+      "step": 4480
+    },
+    {
+      "epoch": 31.536435469710273,
+      "grad_norm": 0.83984375,
+      "learning_rate": 4.302597854121127e-09,
+      "loss": 0.4096,
+      "step": 4490
+    },
+    {
+      "epoch": 31.60667251975417,
+      "grad_norm": 0.85546875,
+      "learning_rate": 2.8567321746092446e-09,
+      "loss": 0.4092,
+      "step": 4500
+    },
+    {
+      "epoch": 31.67690956979807,
+      "grad_norm": 0.8515625,
+      "learning_rate": 1.705841480810766e-09,
+      "loss": 0.4072,
+      "step": 4510
+    },
+    {
+      "epoch": 31.747146619841967,
+      "grad_norm": 0.8671875,
+      "learning_rate": 8.499937083339404e-10,
+      "loss": 0.4261,
+      "step": 4520
+    },
+    {
+      "epoch": 31.817383669885864,
+      "grad_norm": 0.84765625,
+      "learning_rate": 2.892393767800483e-10,
+      "loss": 0.4138,
+      "step": 4530
+    },
+    {
+      "epoch": 31.887620719929764,
+      "grad_norm": 0.87109375,
+      "learning_rate": 2.3611586760785566e-11,
+      "loss": 0.3897,
+      "step": 4540
+    },
+    {
+      "epoch": 31.91571553994732,
+      "step": 4544,
+      "total_flos": 9.547395325030564e+17,
+      "train_loss": 0.8426483250682203,
+      "train_runtime": 5512.317,
+      "train_samples_per_second": 13.218,
+      "train_steps_per_second": 0.824
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 4544,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 32,
+  "save_steps": 0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 9.547395325030564e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c8decbe7e60165a3368e714030855da00ff4c16274ac50fdc33a5238a89a182e
+size 5368