Upload 13 files
Browse files- config.json +64 -0
- generation_config.json +16 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- source.spm +0 -0
- special_tokens_map.json +5 -0
- target.spm +0 -0
- tokenizer_config.json +12 -0
- trainer_state.json +1978 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,64 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "trans_model",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"activation_function": "swish",
|
| 5 |
+
"add_bias_logits": false,
|
| 6 |
+
"add_final_layer_norm": false,
|
| 7 |
+
"architectures": [
|
| 8 |
+
"MarianMTModel"
|
| 9 |
+
],
|
| 10 |
+
"attention_dropout": 0.0,
|
| 11 |
+
"bad_words_ids": [
|
| 12 |
+
[
|
| 13 |
+
65000
|
| 14 |
+
]
|
| 15 |
+
],
|
| 16 |
+
"bos_token_id": 0,
|
| 17 |
+
"classif_dropout": 0.0,
|
| 18 |
+
"classifier_dropout": 0.0,
|
| 19 |
+
"d_model": 512,
|
| 20 |
+
"decoder_attention_heads": 8,
|
| 21 |
+
"decoder_ffn_dim": 2048,
|
| 22 |
+
"decoder_layerdrop": 0.0,
|
| 23 |
+
"decoder_layers": 6,
|
| 24 |
+
"decoder_start_token_id": 65000,
|
| 25 |
+
"decoder_vocab_size": 65001,
|
| 26 |
+
"do_blenderbot_90_layernorm": false,
|
| 27 |
+
"dropout": 0.1,
|
| 28 |
+
"encoder_attention_heads": 8,
|
| 29 |
+
"encoder_ffn_dim": 2048,
|
| 30 |
+
"encoder_layerdrop": 0.0,
|
| 31 |
+
"encoder_layers": 6,
|
| 32 |
+
"eos_token_id": 0,
|
| 33 |
+
"extra_pos_embeddings": 0,
|
| 34 |
+
"force_bos_token_to_be_generated": false,
|
| 35 |
+
"forced_eos_token_id": 0,
|
| 36 |
+
"gradient_checkpointing": false,
|
| 37 |
+
"id2label": {
|
| 38 |
+
"0": "LABEL_0",
|
| 39 |
+
"1": "LABEL_1",
|
| 40 |
+
"2": "LABEL_2"
|
| 41 |
+
},
|
| 42 |
+
"init_std": 0.02,
|
| 43 |
+
"is_encoder_decoder": true,
|
| 44 |
+
"label2id": {
|
| 45 |
+
"LABEL_0": 0,
|
| 46 |
+
"LABEL_1": 1,
|
| 47 |
+
"LABEL_2": 2
|
| 48 |
+
},
|
| 49 |
+
"max_length": 512,
|
| 50 |
+
"max_position_embeddings": 512,
|
| 51 |
+
"model_type": "marian",
|
| 52 |
+
"normalize_before": false,
|
| 53 |
+
"normalize_embedding": false,
|
| 54 |
+
"num_beams": 4,
|
| 55 |
+
"num_hidden_layers": 6,
|
| 56 |
+
"pad_token_id": 65000,
|
| 57 |
+
"scale_embedding": true,
|
| 58 |
+
"share_encoder_decoder_embeddings": true,
|
| 59 |
+
"static_position_embeddings": true,
|
| 60 |
+
"torch_dtype": "float32",
|
| 61 |
+
"transformers_version": "4.30.2",
|
| 62 |
+
"use_cache": true,
|
| 63 |
+
"vocab_size": 65001
|
| 64 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bad_words_ids": [
|
| 4 |
+
[
|
| 5 |
+
65000
|
| 6 |
+
]
|
| 7 |
+
],
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"decoder_start_token_id": 65000,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"forced_eos_token_id": 0,
|
| 12 |
+
"max_length": 512,
|
| 13 |
+
"num_beams": 4,
|
| 14 |
+
"pad_token_id": 65000,
|
| 15 |
+
"transformers_version": "4.30.2"
|
| 16 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:68b975db1caf46413da27809a7786e61eb2f3c4e9c2647f16de1c82a4281fea9
|
| 3 |
+
size 619501061
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:072ccbd6ec6adf71e38319f0c08fa691c51e79def9360c13ee9ad6d4e195a30d
|
| 3 |
+
size 310022533
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ba189c7848e4096e993e8ab804f5bb42f8755bcda7a33135ed67461afe47ac38
|
| 3 |
+
size 14511
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:519e73e22f5a8054abe5b4756ca77bf6ddeb528bd80ca85e2b1e31b233f1adac
|
| 3 |
+
size 627
|
source.spm
ADDED
|
Binary file (806 kB). View file
|
|
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"unk_token": "<unk>"
|
| 5 |
+
}
|
target.spm
ADDED
|
Binary file (805 kB). View file
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"clean_up_tokenization_spaces": true,
|
| 3 |
+
"eos_token": "</s>",
|
| 4 |
+
"model_max_length": 512,
|
| 5 |
+
"pad_token": "<pad>",
|
| 6 |
+
"separate_vocabs": false,
|
| 7 |
+
"source_lang": "eng",
|
| 8 |
+
"sp_model_kwargs": {},
|
| 9 |
+
"target_lang": "zho",
|
| 10 |
+
"tokenizer_class": "MarianTokenizer",
|
| 11 |
+
"unk_token": "<unk>"
|
| 12 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1978 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 9.968985378821444,
|
| 5 |
+
"global_step": 157500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.03,
|
| 12 |
+
"learning_rate": 1.9936704854737643e-05,
|
| 13 |
+
"loss": 2.1885,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.06,
|
| 18 |
+
"learning_rate": 1.9873409709475284e-05,
|
| 19 |
+
"loss": 2.1503,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.09,
|
| 24 |
+
"learning_rate": 1.9810114564212926e-05,
|
| 25 |
+
"loss": 2.1207,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.13,
|
| 30 |
+
"learning_rate": 1.974681941895057e-05,
|
| 31 |
+
"loss": 2.093,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.16,
|
| 36 |
+
"learning_rate": 1.968352427368821e-05,
|
| 37 |
+
"loss": 2.0938,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.19,
|
| 42 |
+
"learning_rate": 1.9620229128425853e-05,
|
| 43 |
+
"loss": 2.0842,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.22,
|
| 48 |
+
"learning_rate": 1.9556933983163494e-05,
|
| 49 |
+
"loss": 2.0991,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.25,
|
| 54 |
+
"learning_rate": 1.9493638837901136e-05,
|
| 55 |
+
"loss": 2.0715,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.28,
|
| 60 |
+
"learning_rate": 1.9430343692638777e-05,
|
| 61 |
+
"loss": 2.073,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.32,
|
| 66 |
+
"learning_rate": 1.936704854737642e-05,
|
| 67 |
+
"loss": 2.08,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.35,
|
| 72 |
+
"learning_rate": 1.930375340211406e-05,
|
| 73 |
+
"loss": 2.0494,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.38,
|
| 78 |
+
"learning_rate": 1.92404582568517e-05,
|
| 79 |
+
"loss": 2.0616,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.41,
|
| 84 |
+
"learning_rate": 1.9177163111589342e-05,
|
| 85 |
+
"loss": 2.0571,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.44,
|
| 90 |
+
"learning_rate": 1.9113867966326984e-05,
|
| 91 |
+
"loss": 2.0561,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.47,
|
| 96 |
+
"learning_rate": 1.9050572821064625e-05,
|
| 97 |
+
"loss": 2.0658,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.51,
|
| 102 |
+
"learning_rate": 1.8987277675802266e-05,
|
| 103 |
+
"loss": 2.0275,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.54,
|
| 108 |
+
"learning_rate": 1.8923982530539908e-05,
|
| 109 |
+
"loss": 2.0335,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.57,
|
| 114 |
+
"learning_rate": 1.8860687385277552e-05,
|
| 115 |
+
"loss": 2.0312,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.6,
|
| 120 |
+
"learning_rate": 1.8797392240015194e-05,
|
| 121 |
+
"loss": 2.0138,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.63,
|
| 126 |
+
"learning_rate": 1.8734097094752835e-05,
|
| 127 |
+
"loss": 2.0005,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.66,
|
| 132 |
+
"learning_rate": 1.8670801949490476e-05,
|
| 133 |
+
"loss": 2.0248,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.7,
|
| 138 |
+
"learning_rate": 1.8607506804228118e-05,
|
| 139 |
+
"loss": 2.0152,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.73,
|
| 144 |
+
"learning_rate": 1.854421165896576e-05,
|
| 145 |
+
"loss": 2.0168,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.76,
|
| 150 |
+
"learning_rate": 1.84809165137034e-05,
|
| 151 |
+
"loss": 2.0196,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.79,
|
| 156 |
+
"learning_rate": 1.841762136844104e-05,
|
| 157 |
+
"loss": 2.0003,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.82,
|
| 162 |
+
"learning_rate": 1.8354326223178683e-05,
|
| 163 |
+
"loss": 2.0267,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.85,
|
| 168 |
+
"learning_rate": 1.8291031077916324e-05,
|
| 169 |
+
"loss": 2.0022,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.89,
|
| 174 |
+
"learning_rate": 1.8227735932653966e-05,
|
| 175 |
+
"loss": 2.0102,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.92,
|
| 180 |
+
"learning_rate": 1.8164440787391607e-05,
|
| 181 |
+
"loss": 2.0,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.95,
|
| 186 |
+
"learning_rate": 1.810114564212925e-05,
|
| 187 |
+
"loss": 1.9751,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.98,
|
| 192 |
+
"learning_rate": 1.803785049686689e-05,
|
| 193 |
+
"loss": 1.9912,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.0,
|
| 198 |
+
"eval_loss": 1.708198070526123,
|
| 199 |
+
"eval_runtime": 629.4341,
|
| 200 |
+
"eval_samples_per_second": 401.594,
|
| 201 |
+
"eval_steps_per_second": 25.1,
|
| 202 |
+
"step": 15799
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 1.01,
|
| 206 |
+
"learning_rate": 1.797455535160453e-05,
|
| 207 |
+
"loss": 1.9373,
|
| 208 |
+
"step": 16000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 1.04,
|
| 212 |
+
"learning_rate": 1.7911260206342176e-05,
|
| 213 |
+
"loss": 1.8308,
|
| 214 |
+
"step": 16500
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 1.08,
|
| 218 |
+
"learning_rate": 1.7847965061079817e-05,
|
| 219 |
+
"loss": 1.8442,
|
| 220 |
+
"step": 17000
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.11,
|
| 224 |
+
"learning_rate": 1.778466991581746e-05,
|
| 225 |
+
"loss": 1.8558,
|
| 226 |
+
"step": 17500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.14,
|
| 230 |
+
"learning_rate": 1.77213747705551e-05,
|
| 231 |
+
"loss": 1.8564,
|
| 232 |
+
"step": 18000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 1.17,
|
| 236 |
+
"learning_rate": 1.765807962529274e-05,
|
| 237 |
+
"loss": 1.8578,
|
| 238 |
+
"step": 18500
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 1.2,
|
| 242 |
+
"learning_rate": 1.7594784480030382e-05,
|
| 243 |
+
"loss": 1.8312,
|
| 244 |
+
"step": 19000
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 1.23,
|
| 248 |
+
"learning_rate": 1.7531489334768024e-05,
|
| 249 |
+
"loss": 1.8784,
|
| 250 |
+
"step": 19500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.27,
|
| 254 |
+
"learning_rate": 1.746819418950567e-05,
|
| 255 |
+
"loss": 1.8497,
|
| 256 |
+
"step": 20000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.3,
|
| 260 |
+
"learning_rate": 1.740489904424331e-05,
|
| 261 |
+
"loss": 1.8528,
|
| 262 |
+
"step": 20500
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.33,
|
| 266 |
+
"learning_rate": 1.734160389898095e-05,
|
| 267 |
+
"loss": 1.8645,
|
| 268 |
+
"step": 21000
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 1.36,
|
| 272 |
+
"learning_rate": 1.7278308753718592e-05,
|
| 273 |
+
"loss": 1.8563,
|
| 274 |
+
"step": 21500
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.39,
|
| 278 |
+
"learning_rate": 1.7215013608456234e-05,
|
| 279 |
+
"loss": 1.8616,
|
| 280 |
+
"step": 22000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 1.42,
|
| 284 |
+
"learning_rate": 1.7151718463193875e-05,
|
| 285 |
+
"loss": 1.8699,
|
| 286 |
+
"step": 22500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.46,
|
| 290 |
+
"learning_rate": 1.7088423317931516e-05,
|
| 291 |
+
"loss": 1.8583,
|
| 292 |
+
"step": 23000
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 1.49,
|
| 296 |
+
"learning_rate": 1.7025128172669158e-05,
|
| 297 |
+
"loss": 1.868,
|
| 298 |
+
"step": 23500
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.52,
|
| 302 |
+
"learning_rate": 1.69618330274068e-05,
|
| 303 |
+
"loss": 1.8534,
|
| 304 |
+
"step": 24000
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.55,
|
| 308 |
+
"learning_rate": 1.689853788214444e-05,
|
| 309 |
+
"loss": 1.8557,
|
| 310 |
+
"step": 24500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.58,
|
| 314 |
+
"learning_rate": 1.683524273688208e-05,
|
| 315 |
+
"loss": 1.8709,
|
| 316 |
+
"step": 25000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.61,
|
| 320 |
+
"learning_rate": 1.6771947591619723e-05,
|
| 321 |
+
"loss": 1.8544,
|
| 322 |
+
"step": 25500
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 1.65,
|
| 326 |
+
"learning_rate": 1.6708652446357364e-05,
|
| 327 |
+
"loss": 1.8803,
|
| 328 |
+
"step": 26000
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 1.68,
|
| 332 |
+
"learning_rate": 1.6645357301095006e-05,
|
| 333 |
+
"loss": 1.8573,
|
| 334 |
+
"step": 26500
|
| 335 |
+
},
|
| 336 |
+
{
|
| 337 |
+
"epoch": 1.71,
|
| 338 |
+
"learning_rate": 1.658206215583265e-05,
|
| 339 |
+
"loss": 1.8668,
|
| 340 |
+
"step": 27000
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 1.74,
|
| 344 |
+
"learning_rate": 1.6518767010570292e-05,
|
| 345 |
+
"loss": 1.8592,
|
| 346 |
+
"step": 27500
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.77,
|
| 350 |
+
"learning_rate": 1.6455471865307933e-05,
|
| 351 |
+
"loss": 1.8551,
|
| 352 |
+
"step": 28000
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 1.8,
|
| 356 |
+
"learning_rate": 1.6392176720045574e-05,
|
| 357 |
+
"loss": 1.8504,
|
| 358 |
+
"step": 28500
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 1.84,
|
| 362 |
+
"learning_rate": 1.6328881574783216e-05,
|
| 363 |
+
"loss": 1.8578,
|
| 364 |
+
"step": 29000
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 1.87,
|
| 368 |
+
"learning_rate": 1.6265586429520857e-05,
|
| 369 |
+
"loss": 1.8614,
|
| 370 |
+
"step": 29500
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 1.9,
|
| 374 |
+
"learning_rate": 1.62022912842585e-05,
|
| 375 |
+
"loss": 1.8592,
|
| 376 |
+
"step": 30000
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 1.93,
|
| 380 |
+
"learning_rate": 1.613899613899614e-05,
|
| 381 |
+
"loss": 1.854,
|
| 382 |
+
"step": 30500
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 1.96,
|
| 386 |
+
"learning_rate": 1.607570099373378e-05,
|
| 387 |
+
"loss": 1.8536,
|
| 388 |
+
"step": 31000
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.99,
|
| 392 |
+
"learning_rate": 1.6012405848471422e-05,
|
| 393 |
+
"loss": 1.8687,
|
| 394 |
+
"step": 31500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 2.0,
|
| 398 |
+
"eval_loss": 1.5787432193756104,
|
| 399 |
+
"eval_runtime": 629.6856,
|
| 400 |
+
"eval_samples_per_second": 401.434,
|
| 401 |
+
"eval_steps_per_second": 25.09,
|
| 402 |
+
"step": 31598
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 2.03,
|
| 406 |
+
"learning_rate": 1.5949110703209064e-05,
|
| 407 |
+
"loss": 1.7515,
|
| 408 |
+
"step": 32000
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 2.06,
|
| 412 |
+
"learning_rate": 1.5885815557946705e-05,
|
| 413 |
+
"loss": 1.7233,
|
| 414 |
+
"step": 32500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 2.09,
|
| 418 |
+
"learning_rate": 1.5822520412684346e-05,
|
| 419 |
+
"loss": 1.754,
|
| 420 |
+
"step": 33000
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 2.12,
|
| 424 |
+
"learning_rate": 1.5759225267421988e-05,
|
| 425 |
+
"loss": 1.7302,
|
| 426 |
+
"step": 33500
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 2.15,
|
| 430 |
+
"learning_rate": 1.5695930122159632e-05,
|
| 431 |
+
"loss": 1.7369,
|
| 432 |
+
"step": 34000
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 2.18,
|
| 436 |
+
"learning_rate": 1.5632634976897274e-05,
|
| 437 |
+
"loss": 1.7294,
|
| 438 |
+
"step": 34500
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.22,
|
| 442 |
+
"learning_rate": 1.5569339831634915e-05,
|
| 443 |
+
"loss": 1.7317,
|
| 444 |
+
"step": 35000
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 2.25,
|
| 448 |
+
"learning_rate": 1.5506044686372556e-05,
|
| 449 |
+
"loss": 1.7457,
|
| 450 |
+
"step": 35500
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 2.28,
|
| 454 |
+
"learning_rate": 1.5442749541110198e-05,
|
| 455 |
+
"loss": 1.758,
|
| 456 |
+
"step": 36000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 2.31,
|
| 460 |
+
"learning_rate": 1.537945439584784e-05,
|
| 461 |
+
"loss": 1.7442,
|
| 462 |
+
"step": 36500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 2.34,
|
| 466 |
+
"learning_rate": 1.531615925058548e-05,
|
| 467 |
+
"loss": 1.7449,
|
| 468 |
+
"step": 37000
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 2.37,
|
| 472 |
+
"learning_rate": 1.5252864105323122e-05,
|
| 473 |
+
"loss": 1.7502,
|
| 474 |
+
"step": 37500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.41,
|
| 478 |
+
"learning_rate": 1.5189568960060765e-05,
|
| 479 |
+
"loss": 1.7529,
|
| 480 |
+
"step": 38000
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 2.44,
|
| 484 |
+
"learning_rate": 1.5126273814798406e-05,
|
| 485 |
+
"loss": 1.7675,
|
| 486 |
+
"step": 38500
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 2.47,
|
| 490 |
+
"learning_rate": 1.5062978669536047e-05,
|
| 491 |
+
"loss": 1.7537,
|
| 492 |
+
"step": 39000
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 2.5,
|
| 496 |
+
"learning_rate": 1.4999683524273689e-05,
|
| 497 |
+
"loss": 1.7546,
|
| 498 |
+
"step": 39500
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 2.53,
|
| 502 |
+
"learning_rate": 1.493638837901133e-05,
|
| 503 |
+
"loss": 1.7409,
|
| 504 |
+
"step": 40000
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 2.56,
|
| 508 |
+
"learning_rate": 1.4873093233748971e-05,
|
| 509 |
+
"loss": 1.7599,
|
| 510 |
+
"step": 40500
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.6,
|
| 514 |
+
"learning_rate": 1.4809798088486613e-05,
|
| 515 |
+
"loss": 1.7467,
|
| 516 |
+
"step": 41000
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 2.63,
|
| 520 |
+
"learning_rate": 1.4746502943224257e-05,
|
| 521 |
+
"loss": 1.7426,
|
| 522 |
+
"step": 41500
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 2.66,
|
| 526 |
+
"learning_rate": 1.4683207797961899e-05,
|
| 527 |
+
"loss": 1.7421,
|
| 528 |
+
"step": 42000
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 2.69,
|
| 532 |
+
"learning_rate": 1.461991265269954e-05,
|
| 533 |
+
"loss": 1.7572,
|
| 534 |
+
"step": 42500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 2.72,
|
| 538 |
+
"learning_rate": 1.4556617507437181e-05,
|
| 539 |
+
"loss": 1.7489,
|
| 540 |
+
"step": 43000
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 2.75,
|
| 544 |
+
"learning_rate": 1.4493322362174823e-05,
|
| 545 |
+
"loss": 1.7482,
|
| 546 |
+
"step": 43500
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 2.78,
|
| 550 |
+
"learning_rate": 1.4430027216912464e-05,
|
| 551 |
+
"loss": 1.7578,
|
| 552 |
+
"step": 44000
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 2.82,
|
| 556 |
+
"learning_rate": 1.4366732071650105e-05,
|
| 557 |
+
"loss": 1.7608,
|
| 558 |
+
"step": 44500
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 2.85,
|
| 562 |
+
"learning_rate": 1.4303436926387748e-05,
|
| 563 |
+
"loss": 1.7623,
|
| 564 |
+
"step": 45000
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 2.88,
|
| 568 |
+
"learning_rate": 1.424014178112539e-05,
|
| 569 |
+
"loss": 1.7534,
|
| 570 |
+
"step": 45500
|
| 571 |
+
},
|
| 572 |
+
{
|
| 573 |
+
"epoch": 2.91,
|
| 574 |
+
"learning_rate": 1.4176846635863031e-05,
|
| 575 |
+
"loss": 1.7513,
|
| 576 |
+
"step": 46000
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 2.94,
|
| 580 |
+
"learning_rate": 1.4113551490600672e-05,
|
| 581 |
+
"loss": 1.7539,
|
| 582 |
+
"step": 46500
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 2.97,
|
| 586 |
+
"learning_rate": 1.4050256345338314e-05,
|
| 587 |
+
"loss": 1.7529,
|
| 588 |
+
"step": 47000
|
| 589 |
+
},
|
| 590 |
+
{
|
| 591 |
+
"epoch": 3.0,
|
| 592 |
+
"eval_loss": 1.4882566928863525,
|
| 593 |
+
"eval_runtime": 671.3515,
|
| 594 |
+
"eval_samples_per_second": 376.52,
|
| 595 |
+
"eval_steps_per_second": 23.533,
|
| 596 |
+
"step": 47397
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 3.01,
|
| 600 |
+
"learning_rate": 1.3986961200075955e-05,
|
| 601 |
+
"loss": 1.7233,
|
| 602 |
+
"step": 47500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 3.04,
|
| 606 |
+
"learning_rate": 1.3923666054813596e-05,
|
| 607 |
+
"loss": 1.6255,
|
| 608 |
+
"step": 48000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 3.07,
|
| 612 |
+
"learning_rate": 1.386037090955124e-05,
|
| 613 |
+
"loss": 1.6566,
|
| 614 |
+
"step": 48500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 3.1,
|
| 618 |
+
"learning_rate": 1.379707576428888e-05,
|
| 619 |
+
"loss": 1.6442,
|
| 620 |
+
"step": 49000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 3.13,
|
| 624 |
+
"learning_rate": 1.3733780619026522e-05,
|
| 625 |
+
"loss": 1.6439,
|
| 626 |
+
"step": 49500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 3.16,
|
| 630 |
+
"learning_rate": 1.3670485473764163e-05,
|
| 631 |
+
"loss": 1.6438,
|
| 632 |
+
"step": 50000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 3.2,
|
| 636 |
+
"learning_rate": 1.3607190328501805e-05,
|
| 637 |
+
"loss": 1.6527,
|
| 638 |
+
"step": 50500
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 3.23,
|
| 642 |
+
"learning_rate": 1.3543895183239446e-05,
|
| 643 |
+
"loss": 1.6426,
|
| 644 |
+
"step": 51000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 3.26,
|
| 648 |
+
"learning_rate": 1.3480600037977087e-05,
|
| 649 |
+
"loss": 1.6802,
|
| 650 |
+
"step": 51500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 3.29,
|
| 654 |
+
"learning_rate": 1.341730489271473e-05,
|
| 655 |
+
"loss": 1.6568,
|
| 656 |
+
"step": 52000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 3.32,
|
| 660 |
+
"learning_rate": 1.3354009747452372e-05,
|
| 661 |
+
"loss": 1.6657,
|
| 662 |
+
"step": 52500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 3.35,
|
| 666 |
+
"learning_rate": 1.3290714602190013e-05,
|
| 667 |
+
"loss": 1.6734,
|
| 668 |
+
"step": 53000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 3.39,
|
| 672 |
+
"learning_rate": 1.3227419456927654e-05,
|
| 673 |
+
"loss": 1.655,
|
| 674 |
+
"step": 53500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 3.42,
|
| 678 |
+
"learning_rate": 1.3164124311665296e-05,
|
| 679 |
+
"loss": 1.6831,
|
| 680 |
+
"step": 54000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 3.45,
|
| 684 |
+
"learning_rate": 1.3100829166402937e-05,
|
| 685 |
+
"loss": 1.6532,
|
| 686 |
+
"step": 54500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 3.48,
|
| 690 |
+
"learning_rate": 1.3037534021140578e-05,
|
| 691 |
+
"loss": 1.6649,
|
| 692 |
+
"step": 55000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 3.51,
|
| 696 |
+
"learning_rate": 1.2974238875878221e-05,
|
| 697 |
+
"loss": 1.6643,
|
| 698 |
+
"step": 55500
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 3.54,
|
| 702 |
+
"learning_rate": 1.2910943730615863e-05,
|
| 703 |
+
"loss": 1.6749,
|
| 704 |
+
"step": 56000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 3.58,
|
| 708 |
+
"learning_rate": 1.2847648585353504e-05,
|
| 709 |
+
"loss": 1.6802,
|
| 710 |
+
"step": 56500
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 3.61,
|
| 714 |
+
"learning_rate": 1.2784353440091145e-05,
|
| 715 |
+
"loss": 1.6753,
|
| 716 |
+
"step": 57000
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 3.64,
|
| 720 |
+
"learning_rate": 1.2721058294828787e-05,
|
| 721 |
+
"loss": 1.6759,
|
| 722 |
+
"step": 57500
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 3.67,
|
| 726 |
+
"learning_rate": 1.2657763149566428e-05,
|
| 727 |
+
"loss": 1.6756,
|
| 728 |
+
"step": 58000
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 3.7,
|
| 732 |
+
"learning_rate": 1.259446800430407e-05,
|
| 733 |
+
"loss": 1.6733,
|
| 734 |
+
"step": 58500
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 3.73,
|
| 738 |
+
"learning_rate": 1.253117285904171e-05,
|
| 739 |
+
"loss": 1.671,
|
| 740 |
+
"step": 59000
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 3.77,
|
| 744 |
+
"learning_rate": 1.2467877713779355e-05,
|
| 745 |
+
"loss": 1.6697,
|
| 746 |
+
"step": 59500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 3.8,
|
| 750 |
+
"learning_rate": 1.2404582568516997e-05,
|
| 751 |
+
"loss": 1.668,
|
| 752 |
+
"step": 60000
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 3.83,
|
| 756 |
+
"learning_rate": 1.2341287423254638e-05,
|
| 757 |
+
"loss": 1.6689,
|
| 758 |
+
"step": 60500
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 3.86,
|
| 762 |
+
"learning_rate": 1.227799227799228e-05,
|
| 763 |
+
"loss": 1.6874,
|
| 764 |
+
"step": 61000
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 3.89,
|
| 768 |
+
"learning_rate": 1.221469713272992e-05,
|
| 769 |
+
"loss": 1.6926,
|
| 770 |
+
"step": 61500
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 3.92,
|
| 774 |
+
"learning_rate": 1.2151401987467562e-05,
|
| 775 |
+
"loss": 1.6819,
|
| 776 |
+
"step": 62000
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 3.96,
|
| 780 |
+
"learning_rate": 1.2088106842205203e-05,
|
| 781 |
+
"loss": 1.6599,
|
| 782 |
+
"step": 62500
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 3.99,
|
| 786 |
+
"learning_rate": 1.2024811696942846e-05,
|
| 787 |
+
"loss": 1.6886,
|
| 788 |
+
"step": 63000
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 4.0,
|
| 792 |
+
"eval_loss": 1.417983055114746,
|
| 793 |
+
"eval_runtime": 634.8433,
|
| 794 |
+
"eval_samples_per_second": 398.172,
|
| 795 |
+
"eval_steps_per_second": 24.886,
|
| 796 |
+
"step": 63196
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 4.02,
|
| 800 |
+
"learning_rate": 1.1961516551680488e-05,
|
| 801 |
+
"loss": 1.6122,
|
| 802 |
+
"step": 63500
|
| 803 |
+
},
|
| 804 |
+
{
|
| 805 |
+
"epoch": 4.05,
|
| 806 |
+
"learning_rate": 1.1898221406418129e-05,
|
| 807 |
+
"loss": 1.578,
|
| 808 |
+
"step": 64000
|
| 809 |
+
},
|
| 810 |
+
{
|
| 811 |
+
"epoch": 4.08,
|
| 812 |
+
"learning_rate": 1.183492626115577e-05,
|
| 813 |
+
"loss": 1.5662,
|
| 814 |
+
"step": 64500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 4.11,
|
| 818 |
+
"learning_rate": 1.1771631115893412e-05,
|
| 819 |
+
"loss": 1.5732,
|
| 820 |
+
"step": 65000
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 4.15,
|
| 824 |
+
"learning_rate": 1.1708335970631053e-05,
|
| 825 |
+
"loss": 1.5726,
|
| 826 |
+
"step": 65500
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 4.18,
|
| 830 |
+
"learning_rate": 1.1645040825368694e-05,
|
| 831 |
+
"loss": 1.5868,
|
| 832 |
+
"step": 66000
|
| 833 |
+
},
|
| 834 |
+
{
|
| 835 |
+
"epoch": 4.21,
|
| 836 |
+
"learning_rate": 1.1581745680106337e-05,
|
| 837 |
+
"loss": 1.5781,
|
| 838 |
+
"step": 66500
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 4.24,
|
| 842 |
+
"learning_rate": 1.1518450534843979e-05,
|
| 843 |
+
"loss": 1.5965,
|
| 844 |
+
"step": 67000
|
| 845 |
+
},
|
| 846 |
+
{
|
| 847 |
+
"epoch": 4.27,
|
| 848 |
+
"learning_rate": 1.145515538958162e-05,
|
| 849 |
+
"loss": 1.5934,
|
| 850 |
+
"step": 67500
|
| 851 |
+
},
|
| 852 |
+
{
|
| 853 |
+
"epoch": 4.3,
|
| 854 |
+
"learning_rate": 1.1391860244319261e-05,
|
| 855 |
+
"loss": 1.5791,
|
| 856 |
+
"step": 68000
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 4.34,
|
| 860 |
+
"learning_rate": 1.1328565099056903e-05,
|
| 861 |
+
"loss": 1.6037,
|
| 862 |
+
"step": 68500
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 4.37,
|
| 866 |
+
"learning_rate": 1.1265269953794544e-05,
|
| 867 |
+
"loss": 1.6046,
|
| 868 |
+
"step": 69000
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 4.4,
|
| 872 |
+
"learning_rate": 1.1201974808532185e-05,
|
| 873 |
+
"loss": 1.5903,
|
| 874 |
+
"step": 69500
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 4.43,
|
| 878 |
+
"learning_rate": 1.1138679663269828e-05,
|
| 879 |
+
"loss": 1.5837,
|
| 880 |
+
"step": 70000
|
| 881 |
+
},
|
| 882 |
+
{
|
| 883 |
+
"epoch": 4.46,
|
| 884 |
+
"learning_rate": 1.107538451800747e-05,
|
| 885 |
+
"loss": 1.6162,
|
| 886 |
+
"step": 70500
|
| 887 |
+
},
|
| 888 |
+
{
|
| 889 |
+
"epoch": 4.49,
|
| 890 |
+
"learning_rate": 1.1012089372745111e-05,
|
| 891 |
+
"loss": 1.5988,
|
| 892 |
+
"step": 71000
|
| 893 |
+
},
|
| 894 |
+
{
|
| 895 |
+
"epoch": 4.53,
|
| 896 |
+
"learning_rate": 1.0948794227482752e-05,
|
| 897 |
+
"loss": 1.6082,
|
| 898 |
+
"step": 71500
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 4.56,
|
| 902 |
+
"learning_rate": 1.0885499082220394e-05,
|
| 903 |
+
"loss": 1.5832,
|
| 904 |
+
"step": 72000
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 4.59,
|
| 908 |
+
"learning_rate": 1.0822203936958035e-05,
|
| 909 |
+
"loss": 1.6153,
|
| 910 |
+
"step": 72500
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 4.62,
|
| 914 |
+
"learning_rate": 1.0758908791695676e-05,
|
| 915 |
+
"loss": 1.6178,
|
| 916 |
+
"step": 73000
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 4.65,
|
| 920 |
+
"learning_rate": 1.0695613646433321e-05,
|
| 921 |
+
"loss": 1.5981,
|
| 922 |
+
"step": 73500
|
| 923 |
+
},
|
| 924 |
+
{
|
| 925 |
+
"epoch": 4.68,
|
| 926 |
+
"learning_rate": 1.0632318501170963e-05,
|
| 927 |
+
"loss": 1.6135,
|
| 928 |
+
"step": 74000
|
| 929 |
+
},
|
| 930 |
+
{
|
| 931 |
+
"epoch": 4.72,
|
| 932 |
+
"learning_rate": 1.0569023355908604e-05,
|
| 933 |
+
"loss": 1.6122,
|
| 934 |
+
"step": 74500
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 4.75,
|
| 938 |
+
"learning_rate": 1.0505728210646245e-05,
|
| 939 |
+
"loss": 1.5929,
|
| 940 |
+
"step": 75000
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 4.78,
|
| 944 |
+
"learning_rate": 1.0442433065383886e-05,
|
| 945 |
+
"loss": 1.6069,
|
| 946 |
+
"step": 75500
|
| 947 |
+
},
|
| 948 |
+
{
|
| 949 |
+
"epoch": 4.81,
|
| 950 |
+
"learning_rate": 1.0379137920121528e-05,
|
| 951 |
+
"loss": 1.6025,
|
| 952 |
+
"step": 76000
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 4.84,
|
| 956 |
+
"learning_rate": 1.0315842774859167e-05,
|
| 957 |
+
"loss": 1.6284,
|
| 958 |
+
"step": 76500
|
| 959 |
+
},
|
| 960 |
+
{
|
| 961 |
+
"epoch": 4.87,
|
| 962 |
+
"learning_rate": 1.0252547629596812e-05,
|
| 963 |
+
"loss": 1.6134,
|
| 964 |
+
"step": 77000
|
| 965 |
+
},
|
| 966 |
+
{
|
| 967 |
+
"epoch": 4.91,
|
| 968 |
+
"learning_rate": 1.0189252484334454e-05,
|
| 969 |
+
"loss": 1.6092,
|
| 970 |
+
"step": 77500
|
| 971 |
+
},
|
| 972 |
+
{
|
| 973 |
+
"epoch": 4.94,
|
| 974 |
+
"learning_rate": 1.0125957339072095e-05,
|
| 975 |
+
"loss": 1.6194,
|
| 976 |
+
"step": 78000
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 4.97,
|
| 980 |
+
"learning_rate": 1.0062662193809736e-05,
|
| 981 |
+
"loss": 1.6227,
|
| 982 |
+
"step": 78500
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 5.0,
|
| 986 |
+
"eval_loss": 1.3593807220458984,
|
| 987 |
+
"eval_runtime": 634.5713,
|
| 988 |
+
"eval_samples_per_second": 398.343,
|
| 989 |
+
"eval_steps_per_second": 24.897,
|
| 990 |
+
"step": 78995
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 5.0,
|
| 994 |
+
"learning_rate": 9.999367048547378e-06,
|
| 995 |
+
"loss": 1.6451,
|
| 996 |
+
"step": 79000
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 5.03,
|
| 1000 |
+
"learning_rate": 9.936071903285019e-06,
|
| 1001 |
+
"loss": 1.5186,
|
| 1002 |
+
"step": 79500
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 5.06,
|
| 1006 |
+
"learning_rate": 9.87277675802266e-06,
|
| 1007 |
+
"loss": 1.5124,
|
| 1008 |
+
"step": 80000
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 5.1,
|
| 1012 |
+
"learning_rate": 9.809481612760301e-06,
|
| 1013 |
+
"loss": 1.5223,
|
| 1014 |
+
"step": 80500
|
| 1015 |
+
},
|
| 1016 |
+
{
|
| 1017 |
+
"epoch": 5.13,
|
| 1018 |
+
"learning_rate": 9.746186467497943e-06,
|
| 1019 |
+
"loss": 1.5234,
|
| 1020 |
+
"step": 81000
|
| 1021 |
+
},
|
| 1022 |
+
{
|
| 1023 |
+
"epoch": 5.16,
|
| 1024 |
+
"learning_rate": 9.682891322235586e-06,
|
| 1025 |
+
"loss": 1.5298,
|
| 1026 |
+
"step": 81500
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 5.19,
|
| 1030 |
+
"learning_rate": 9.619596176973227e-06,
|
| 1031 |
+
"loss": 1.5259,
|
| 1032 |
+
"step": 82000
|
| 1033 |
+
},
|
| 1034 |
+
{
|
| 1035 |
+
"epoch": 5.22,
|
| 1036 |
+
"learning_rate": 9.556301031710869e-06,
|
| 1037 |
+
"loss": 1.5463,
|
| 1038 |
+
"step": 82500
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 5.25,
|
| 1042 |
+
"learning_rate": 9.49300588644851e-06,
|
| 1043 |
+
"loss": 1.5367,
|
| 1044 |
+
"step": 83000
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 5.29,
|
| 1048 |
+
"learning_rate": 9.429710741186153e-06,
|
| 1049 |
+
"loss": 1.543,
|
| 1050 |
+
"step": 83500
|
| 1051 |
+
},
|
| 1052 |
+
{
|
| 1053 |
+
"epoch": 5.32,
|
| 1054 |
+
"learning_rate": 9.366415595923794e-06,
|
| 1055 |
+
"loss": 1.5379,
|
| 1056 |
+
"step": 84000
|
| 1057 |
+
},
|
| 1058 |
+
{
|
| 1059 |
+
"epoch": 5.35,
|
| 1060 |
+
"learning_rate": 9.303120450661436e-06,
|
| 1061 |
+
"loss": 1.5215,
|
| 1062 |
+
"step": 84500
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 5.38,
|
| 1066 |
+
"learning_rate": 9.239825305399077e-06,
|
| 1067 |
+
"loss": 1.5339,
|
| 1068 |
+
"step": 85000
|
| 1069 |
+
},
|
| 1070 |
+
{
|
| 1071 |
+
"epoch": 5.41,
|
| 1072 |
+
"learning_rate": 9.176530160136718e-06,
|
| 1073 |
+
"loss": 1.5588,
|
| 1074 |
+
"step": 85500
|
| 1075 |
+
},
|
| 1076 |
+
{
|
| 1077 |
+
"epoch": 5.44,
|
| 1078 |
+
"learning_rate": 9.11323501487436e-06,
|
| 1079 |
+
"loss": 1.5522,
|
| 1080 |
+
"step": 86000
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 5.48,
|
| 1084 |
+
"learning_rate": 9.049939869612e-06,
|
| 1085 |
+
"loss": 1.5516,
|
| 1086 |
+
"step": 86500
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 5.51,
|
| 1090 |
+
"learning_rate": 8.986644724349644e-06,
|
| 1091 |
+
"loss": 1.5503,
|
| 1092 |
+
"step": 87000
|
| 1093 |
+
},
|
| 1094 |
+
{
|
| 1095 |
+
"epoch": 5.54,
|
| 1096 |
+
"learning_rate": 8.923349579087285e-06,
|
| 1097 |
+
"loss": 1.5459,
|
| 1098 |
+
"step": 87500
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 5.57,
|
| 1102 |
+
"learning_rate": 8.860054433824927e-06,
|
| 1103 |
+
"loss": 1.5437,
|
| 1104 |
+
"step": 88000
|
| 1105 |
+
},
|
| 1106 |
+
{
|
| 1107 |
+
"epoch": 5.6,
|
| 1108 |
+
"learning_rate": 8.796759288562568e-06,
|
| 1109 |
+
"loss": 1.5452,
|
| 1110 |
+
"step": 88500
|
| 1111 |
+
},
|
| 1112 |
+
{
|
| 1113 |
+
"epoch": 5.63,
|
| 1114 |
+
"learning_rate": 8.73346414330021e-06,
|
| 1115 |
+
"loss": 1.5434,
|
| 1116 |
+
"step": 89000
|
| 1117 |
+
},
|
| 1118 |
+
{
|
| 1119 |
+
"epoch": 5.66,
|
| 1120 |
+
"learning_rate": 8.67016899803785e-06,
|
| 1121 |
+
"loss": 1.5633,
|
| 1122 |
+
"step": 89500
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 5.7,
|
| 1126 |
+
"learning_rate": 8.606873852775492e-06,
|
| 1127 |
+
"loss": 1.5535,
|
| 1128 |
+
"step": 90000
|
| 1129 |
+
},
|
| 1130 |
+
{
|
| 1131 |
+
"epoch": 5.73,
|
| 1132 |
+
"learning_rate": 8.543578707513135e-06,
|
| 1133 |
+
"loss": 1.5692,
|
| 1134 |
+
"step": 90500
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 5.76,
|
| 1138 |
+
"learning_rate": 8.480283562250776e-06,
|
| 1139 |
+
"loss": 1.5609,
|
| 1140 |
+
"step": 91000
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 5.79,
|
| 1144 |
+
"learning_rate": 8.416988416988418e-06,
|
| 1145 |
+
"loss": 1.5529,
|
| 1146 |
+
"step": 91500
|
| 1147 |
+
},
|
| 1148 |
+
{
|
| 1149 |
+
"epoch": 5.82,
|
| 1150 |
+
"learning_rate": 8.353693271726059e-06,
|
| 1151 |
+
"loss": 1.5602,
|
| 1152 |
+
"step": 92000
|
| 1153 |
+
},
|
| 1154 |
+
{
|
| 1155 |
+
"epoch": 5.85,
|
| 1156 |
+
"learning_rate": 8.290398126463702e-06,
|
| 1157 |
+
"loss": 1.5547,
|
| 1158 |
+
"step": 92500
|
| 1159 |
+
},
|
| 1160 |
+
{
|
| 1161 |
+
"epoch": 5.89,
|
| 1162 |
+
"learning_rate": 8.227102981201343e-06,
|
| 1163 |
+
"loss": 1.5557,
|
| 1164 |
+
"step": 93000
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 5.92,
|
| 1168 |
+
"learning_rate": 8.163807835938985e-06,
|
| 1169 |
+
"loss": 1.5488,
|
| 1170 |
+
"step": 93500
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 5.95,
|
| 1174 |
+
"learning_rate": 8.100512690676626e-06,
|
| 1175 |
+
"loss": 1.5736,
|
| 1176 |
+
"step": 94000
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 5.98,
|
| 1180 |
+
"learning_rate": 8.037217545414267e-06,
|
| 1181 |
+
"loss": 1.559,
|
| 1182 |
+
"step": 94500
|
| 1183 |
+
},
|
| 1184 |
+
{
|
| 1185 |
+
"epoch": 6.0,
|
| 1186 |
+
"eval_loss": 1.3149573802947998,
|
| 1187 |
+
"eval_runtime": 678.6783,
|
| 1188 |
+
"eval_samples_per_second": 372.455,
|
| 1189 |
+
"eval_steps_per_second": 23.279,
|
| 1190 |
+
"step": 94794
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 6.01,
|
| 1194 |
+
"learning_rate": 7.973922400151909e-06,
|
| 1195 |
+
"loss": 1.5248,
|
| 1196 |
+
"step": 95000
|
| 1197 |
+
},
|
| 1198 |
+
{
|
| 1199 |
+
"epoch": 6.04,
|
| 1200 |
+
"learning_rate": 7.91062725488955e-06,
|
| 1201 |
+
"loss": 1.4873,
|
| 1202 |
+
"step": 95500
|
| 1203 |
+
},
|
| 1204 |
+
{
|
| 1205 |
+
"epoch": 6.08,
|
| 1206 |
+
"learning_rate": 7.847332109627193e-06,
|
| 1207 |
+
"loss": 1.4885,
|
| 1208 |
+
"step": 96000
|
| 1209 |
+
},
|
| 1210 |
+
{
|
| 1211 |
+
"epoch": 6.11,
|
| 1212 |
+
"learning_rate": 7.784036964364834e-06,
|
| 1213 |
+
"loss": 1.4882,
|
| 1214 |
+
"step": 96500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 6.14,
|
| 1218 |
+
"learning_rate": 7.720741819102476e-06,
|
| 1219 |
+
"loss": 1.499,
|
| 1220 |
+
"step": 97000
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 6.17,
|
| 1224 |
+
"learning_rate": 7.657446673840117e-06,
|
| 1225 |
+
"loss": 1.493,
|
| 1226 |
+
"step": 97500
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 6.2,
|
| 1230 |
+
"learning_rate": 7.594151528577759e-06,
|
| 1231 |
+
"loss": 1.4864,
|
| 1232 |
+
"step": 98000
|
| 1233 |
+
},
|
| 1234 |
+
{
|
| 1235 |
+
"epoch": 6.23,
|
| 1236 |
+
"learning_rate": 7.5308563833154e-06,
|
| 1237 |
+
"loss": 1.4889,
|
| 1238 |
+
"step": 98500
|
| 1239 |
+
},
|
| 1240 |
+
{
|
| 1241 |
+
"epoch": 6.27,
|
| 1242 |
+
"learning_rate": 7.467561238053042e-06,
|
| 1243 |
+
"loss": 1.5047,
|
| 1244 |
+
"step": 99000
|
| 1245 |
+
},
|
| 1246 |
+
{
|
| 1247 |
+
"epoch": 6.3,
|
| 1248 |
+
"learning_rate": 7.404266092790684e-06,
|
| 1249 |
+
"loss": 1.4828,
|
| 1250 |
+
"step": 99500
|
| 1251 |
+
},
|
| 1252 |
+
{
|
| 1253 |
+
"epoch": 6.33,
|
| 1254 |
+
"learning_rate": 7.340970947528325e-06,
|
| 1255 |
+
"loss": 1.4884,
|
| 1256 |
+
"step": 100000
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 6.36,
|
| 1260 |
+
"learning_rate": 7.2776758022659665e-06,
|
| 1261 |
+
"loss": 1.4981,
|
| 1262 |
+
"step": 100500
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 6.39,
|
| 1266 |
+
"learning_rate": 7.214380657003608e-06,
|
| 1267 |
+
"loss": 1.494,
|
| 1268 |
+
"step": 101000
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 6.42,
|
| 1272 |
+
"learning_rate": 7.15108551174125e-06,
|
| 1273 |
+
"loss": 1.4798,
|
| 1274 |
+
"step": 101500
|
| 1275 |
+
},
|
| 1276 |
+
{
|
| 1277 |
+
"epoch": 6.46,
|
| 1278 |
+
"learning_rate": 7.087790366478891e-06,
|
| 1279 |
+
"loss": 1.498,
|
| 1280 |
+
"step": 102000
|
| 1281 |
+
},
|
| 1282 |
+
{
|
| 1283 |
+
"epoch": 6.49,
|
| 1284 |
+
"learning_rate": 7.024495221216533e-06,
|
| 1285 |
+
"loss": 1.496,
|
| 1286 |
+
"step": 102500
|
| 1287 |
+
},
|
| 1288 |
+
{
|
| 1289 |
+
"epoch": 6.52,
|
| 1290 |
+
"learning_rate": 6.961200075954176e-06,
|
| 1291 |
+
"loss": 1.5097,
|
| 1292 |
+
"step": 103000
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 6.55,
|
| 1296 |
+
"learning_rate": 6.897904930691817e-06,
|
| 1297 |
+
"loss": 1.5032,
|
| 1298 |
+
"step": 103500
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 6.58,
|
| 1302 |
+
"learning_rate": 6.8346097854294576e-06,
|
| 1303 |
+
"loss": 1.5001,
|
| 1304 |
+
"step": 104000
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 6.61,
|
| 1308 |
+
"learning_rate": 6.771314640167099e-06,
|
| 1309 |
+
"loss": 1.5097,
|
| 1310 |
+
"step": 104500
|
| 1311 |
+
},
|
| 1312 |
+
{
|
| 1313 |
+
"epoch": 6.65,
|
| 1314 |
+
"learning_rate": 6.708019494904742e-06,
|
| 1315 |
+
"loss": 1.5065,
|
| 1316 |
+
"step": 105000
|
| 1317 |
+
},
|
| 1318 |
+
{
|
| 1319 |
+
"epoch": 6.68,
|
| 1320 |
+
"learning_rate": 6.644724349642383e-06,
|
| 1321 |
+
"loss": 1.4961,
|
| 1322 |
+
"step": 105500
|
| 1323 |
+
},
|
| 1324 |
+
{
|
| 1325 |
+
"epoch": 6.71,
|
| 1326 |
+
"learning_rate": 6.5814292043800246e-06,
|
| 1327 |
+
"loss": 1.5092,
|
| 1328 |
+
"step": 106000
|
| 1329 |
+
},
|
| 1330 |
+
{
|
| 1331 |
+
"epoch": 6.74,
|
| 1332 |
+
"learning_rate": 6.518134059117667e-06,
|
| 1333 |
+
"loss": 1.5079,
|
| 1334 |
+
"step": 106500
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 6.77,
|
| 1338 |
+
"learning_rate": 6.454838913855308e-06,
|
| 1339 |
+
"loss": 1.513,
|
| 1340 |
+
"step": 107000
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 6.8,
|
| 1344 |
+
"learning_rate": 6.391543768592949e-06,
|
| 1345 |
+
"loss": 1.5076,
|
| 1346 |
+
"step": 107500
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 6.84,
|
| 1350 |
+
"learning_rate": 6.328248623330591e-06,
|
| 1351 |
+
"loss": 1.5123,
|
| 1352 |
+
"step": 108000
|
| 1353 |
+
},
|
| 1354 |
+
{
|
| 1355 |
+
"epoch": 6.87,
|
| 1356 |
+
"learning_rate": 6.264953478068233e-06,
|
| 1357 |
+
"loss": 1.5117,
|
| 1358 |
+
"step": 108500
|
| 1359 |
+
},
|
| 1360 |
+
{
|
| 1361 |
+
"epoch": 6.9,
|
| 1362 |
+
"learning_rate": 6.201658332805874e-06,
|
| 1363 |
+
"loss": 1.5056,
|
| 1364 |
+
"step": 109000
|
| 1365 |
+
},
|
| 1366 |
+
{
|
| 1367 |
+
"epoch": 6.93,
|
| 1368 |
+
"learning_rate": 6.1383631875435156e-06,
|
| 1369 |
+
"loss": 1.517,
|
| 1370 |
+
"step": 109500
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 6.96,
|
| 1374 |
+
"learning_rate": 6.075068042281157e-06,
|
| 1375 |
+
"loss": 1.515,
|
| 1376 |
+
"step": 110000
|
| 1377 |
+
},
|
| 1378 |
+
{
|
| 1379 |
+
"epoch": 6.99,
|
| 1380 |
+
"learning_rate": 6.011772897018799e-06,
|
| 1381 |
+
"loss": 1.5193,
|
| 1382 |
+
"step": 110500
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 7.0,
|
| 1386 |
+
"eval_loss": 1.2794440984725952,
|
| 1387 |
+
"eval_runtime": 637.2277,
|
| 1388 |
+
"eval_samples_per_second": 396.682,
|
| 1389 |
+
"eval_steps_per_second": 24.793,
|
| 1390 |
+
"step": 110593
|
| 1391 |
+
},
|
| 1392 |
+
{
|
| 1393 |
+
"epoch": 7.03,
|
| 1394 |
+
"learning_rate": 5.94847775175644e-06,
|
| 1395 |
+
"loss": 1.4557,
|
| 1396 |
+
"step": 111000
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 7.06,
|
| 1400 |
+
"learning_rate": 5.885182606494082e-06,
|
| 1401 |
+
"loss": 1.4395,
|
| 1402 |
+
"step": 111500
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 7.09,
|
| 1406 |
+
"learning_rate": 5.821887461231725e-06,
|
| 1407 |
+
"loss": 1.4518,
|
| 1408 |
+
"step": 112000
|
| 1409 |
+
},
|
| 1410 |
+
{
|
| 1411 |
+
"epoch": 7.12,
|
| 1412 |
+
"learning_rate": 5.758592315969366e-06,
|
| 1413 |
+
"loss": 1.4513,
|
| 1414 |
+
"step": 112500
|
| 1415 |
+
},
|
| 1416 |
+
{
|
| 1417 |
+
"epoch": 7.15,
|
| 1418 |
+
"learning_rate": 5.695297170707007e-06,
|
| 1419 |
+
"loss": 1.454,
|
| 1420 |
+
"step": 113000
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 7.18,
|
| 1424 |
+
"learning_rate": 5.632002025444649e-06,
|
| 1425 |
+
"loss": 1.4597,
|
| 1426 |
+
"step": 113500
|
| 1427 |
+
},
|
| 1428 |
+
{
|
| 1429 |
+
"epoch": 7.22,
|
| 1430 |
+
"learning_rate": 5.568706880182291e-06,
|
| 1431 |
+
"loss": 1.4383,
|
| 1432 |
+
"step": 114000
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 7.25,
|
| 1436 |
+
"learning_rate": 5.505411734919932e-06,
|
| 1437 |
+
"loss": 1.4529,
|
| 1438 |
+
"step": 114500
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 7.28,
|
| 1442 |
+
"learning_rate": 5.442116589657574e-06,
|
| 1443 |
+
"loss": 1.4706,
|
| 1444 |
+
"step": 115000
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 7.31,
|
| 1448 |
+
"learning_rate": 5.378821444395216e-06,
|
| 1449 |
+
"loss": 1.4576,
|
| 1450 |
+
"step": 115500
|
| 1451 |
+
},
|
| 1452 |
+
{
|
| 1453 |
+
"epoch": 7.34,
|
| 1454 |
+
"learning_rate": 5.315526299132857e-06,
|
| 1455 |
+
"loss": 1.4681,
|
| 1456 |
+
"step": 116000
|
| 1457 |
+
},
|
| 1458 |
+
{
|
| 1459 |
+
"epoch": 7.37,
|
| 1460 |
+
"learning_rate": 5.252231153870498e-06,
|
| 1461 |
+
"loss": 1.4537,
|
| 1462 |
+
"step": 116500
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 7.41,
|
| 1466 |
+
"learning_rate": 5.18893600860814e-06,
|
| 1467 |
+
"loss": 1.4583,
|
| 1468 |
+
"step": 117000
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 7.44,
|
| 1472 |
+
"learning_rate": 5.125640863345782e-06,
|
| 1473 |
+
"loss": 1.4645,
|
| 1474 |
+
"step": 117500
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 7.47,
|
| 1478 |
+
"learning_rate": 5.062345718083423e-06,
|
| 1479 |
+
"loss": 1.455,
|
| 1480 |
+
"step": 118000
|
| 1481 |
+
},
|
| 1482 |
+
{
|
| 1483 |
+
"epoch": 7.5,
|
| 1484 |
+
"learning_rate": 4.999050572821065e-06,
|
| 1485 |
+
"loss": 1.4821,
|
| 1486 |
+
"step": 118500
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 7.53,
|
| 1490 |
+
"learning_rate": 4.935755427558707e-06,
|
| 1491 |
+
"loss": 1.4605,
|
| 1492 |
+
"step": 119000
|
| 1493 |
+
},
|
| 1494 |
+
{
|
| 1495 |
+
"epoch": 7.56,
|
| 1496 |
+
"learning_rate": 4.872460282296348e-06,
|
| 1497 |
+
"loss": 1.4621,
|
| 1498 |
+
"step": 119500
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 7.6,
|
| 1502 |
+
"learning_rate": 4.8091651370339894e-06,
|
| 1503 |
+
"loss": 1.4601,
|
| 1504 |
+
"step": 120000
|
| 1505 |
+
},
|
| 1506 |
+
{
|
| 1507 |
+
"epoch": 7.63,
|
| 1508 |
+
"learning_rate": 4.745869991771632e-06,
|
| 1509 |
+
"loss": 1.4648,
|
| 1510 |
+
"step": 120500
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 7.66,
|
| 1514 |
+
"learning_rate": 4.682574846509273e-06,
|
| 1515 |
+
"loss": 1.4723,
|
| 1516 |
+
"step": 121000
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 7.69,
|
| 1520 |
+
"learning_rate": 4.619279701246915e-06,
|
| 1521 |
+
"loss": 1.4733,
|
| 1522 |
+
"step": 121500
|
| 1523 |
+
},
|
| 1524 |
+
{
|
| 1525 |
+
"epoch": 7.72,
|
| 1526 |
+
"learning_rate": 4.5559845559845564e-06,
|
| 1527 |
+
"loss": 1.4723,
|
| 1528 |
+
"step": 122000
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 7.75,
|
| 1532 |
+
"learning_rate": 4.492689410722198e-06,
|
| 1533 |
+
"loss": 1.4788,
|
| 1534 |
+
"step": 122500
|
| 1535 |
+
},
|
| 1536 |
+
{
|
| 1537 |
+
"epoch": 7.79,
|
| 1538 |
+
"learning_rate": 4.42939426545984e-06,
|
| 1539 |
+
"loss": 1.4665,
|
| 1540 |
+
"step": 123000
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 7.82,
|
| 1544 |
+
"learning_rate": 4.366099120197481e-06,
|
| 1545 |
+
"loss": 1.4699,
|
| 1546 |
+
"step": 123500
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 7.85,
|
| 1550 |
+
"learning_rate": 4.3028039749351235e-06,
|
| 1551 |
+
"loss": 1.4908,
|
| 1552 |
+
"step": 124000
|
| 1553 |
+
},
|
| 1554 |
+
{
|
| 1555 |
+
"epoch": 7.88,
|
| 1556 |
+
"learning_rate": 4.239508829672764e-06,
|
| 1557 |
+
"loss": 1.4712,
|
| 1558 |
+
"step": 124500
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 7.91,
|
| 1562 |
+
"learning_rate": 4.176213684410406e-06,
|
| 1563 |
+
"loss": 1.4722,
|
| 1564 |
+
"step": 125000
|
| 1565 |
+
},
|
| 1566 |
+
{
|
| 1567 |
+
"epoch": 7.94,
|
| 1568 |
+
"learning_rate": 4.1129185391480474e-06,
|
| 1569 |
+
"loss": 1.4856,
|
| 1570 |
+
"step": 125500
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 7.98,
|
| 1574 |
+
"learning_rate": 4.04962339388569e-06,
|
| 1575 |
+
"loss": 1.4793,
|
| 1576 |
+
"step": 126000
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 8.0,
|
| 1580 |
+
"eval_loss": 1.2516653537750244,
|
| 1581 |
+
"eval_runtime": 654.6089,
|
| 1582 |
+
"eval_samples_per_second": 386.15,
|
| 1583 |
+
"eval_steps_per_second": 24.135,
|
| 1584 |
+
"step": 126392
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 8.01,
|
| 1588 |
+
"learning_rate": 3.986328248623331e-06,
|
| 1589 |
+
"loss": 1.4563,
|
| 1590 |
+
"step": 126500
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 8.04,
|
| 1594 |
+
"learning_rate": 3.923033103360972e-06,
|
| 1595 |
+
"loss": 1.4263,
|
| 1596 |
+
"step": 127000
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 8.07,
|
| 1600 |
+
"learning_rate": 3.8597379580986145e-06,
|
| 1601 |
+
"loss": 1.4301,
|
| 1602 |
+
"step": 127500
|
| 1603 |
+
},
|
| 1604 |
+
{
|
| 1605 |
+
"epoch": 8.1,
|
| 1606 |
+
"learning_rate": 3.7964428128362558e-06,
|
| 1607 |
+
"loss": 1.43,
|
| 1608 |
+
"step": 128000
|
| 1609 |
+
},
|
| 1610 |
+
{
|
| 1611 |
+
"epoch": 8.13,
|
| 1612 |
+
"learning_rate": 3.7331476675738975e-06,
|
| 1613 |
+
"loss": 1.4355,
|
| 1614 |
+
"step": 128500
|
| 1615 |
+
},
|
| 1616 |
+
{
|
| 1617 |
+
"epoch": 8.17,
|
| 1618 |
+
"learning_rate": 3.669852522311539e-06,
|
| 1619 |
+
"loss": 1.4384,
|
| 1620 |
+
"step": 129000
|
| 1621 |
+
},
|
| 1622 |
+
{
|
| 1623 |
+
"epoch": 8.2,
|
| 1624 |
+
"learning_rate": 3.6065573770491806e-06,
|
| 1625 |
+
"loss": 1.4398,
|
| 1626 |
+
"step": 129500
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 8.23,
|
| 1630 |
+
"learning_rate": 3.543262231786822e-06,
|
| 1631 |
+
"loss": 1.425,
|
| 1632 |
+
"step": 130000
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 8.26,
|
| 1636 |
+
"learning_rate": 3.4799670865244637e-06,
|
| 1637 |
+
"loss": 1.423,
|
| 1638 |
+
"step": 130500
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 8.29,
|
| 1642 |
+
"learning_rate": 3.416671941262105e-06,
|
| 1643 |
+
"loss": 1.4278,
|
| 1644 |
+
"step": 131000
|
| 1645 |
+
},
|
| 1646 |
+
{
|
| 1647 |
+
"epoch": 8.32,
|
| 1648 |
+
"learning_rate": 3.3533767959997472e-06,
|
| 1649 |
+
"loss": 1.4368,
|
| 1650 |
+
"step": 131500
|
| 1651 |
+
},
|
| 1652 |
+
{
|
| 1653 |
+
"epoch": 8.35,
|
| 1654 |
+
"learning_rate": 3.290081650737389e-06,
|
| 1655 |
+
"loss": 1.4351,
|
| 1656 |
+
"step": 132000
|
| 1657 |
+
},
|
| 1658 |
+
{
|
| 1659 |
+
"epoch": 8.39,
|
| 1660 |
+
"learning_rate": 3.2267865054750303e-06,
|
| 1661 |
+
"loss": 1.4351,
|
| 1662 |
+
"step": 132500
|
| 1663 |
+
},
|
| 1664 |
+
{
|
| 1665 |
+
"epoch": 8.42,
|
| 1666 |
+
"learning_rate": 3.163491360212672e-06,
|
| 1667 |
+
"loss": 1.4299,
|
| 1668 |
+
"step": 133000
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 8.45,
|
| 1672 |
+
"learning_rate": 3.1001962149503134e-06,
|
| 1673 |
+
"loss": 1.4265,
|
| 1674 |
+
"step": 133500
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 8.48,
|
| 1678 |
+
"learning_rate": 3.036901069687955e-06,
|
| 1679 |
+
"loss": 1.4468,
|
| 1680 |
+
"step": 134000
|
| 1681 |
+
},
|
| 1682 |
+
{
|
| 1683 |
+
"epoch": 8.51,
|
| 1684 |
+
"learning_rate": 2.9736059244255965e-06,
|
| 1685 |
+
"loss": 1.4389,
|
| 1686 |
+
"step": 134500
|
| 1687 |
+
},
|
| 1688 |
+
{
|
| 1689 |
+
"epoch": 8.54,
|
| 1690 |
+
"learning_rate": 2.9103107791632386e-06,
|
| 1691 |
+
"loss": 1.4199,
|
| 1692 |
+
"step": 135000
|
| 1693 |
+
},
|
| 1694 |
+
{
|
| 1695 |
+
"epoch": 8.58,
|
| 1696 |
+
"learning_rate": 2.84701563390088e-06,
|
| 1697 |
+
"loss": 1.4361,
|
| 1698 |
+
"step": 135500
|
| 1699 |
+
},
|
| 1700 |
+
{
|
| 1701 |
+
"epoch": 8.61,
|
| 1702 |
+
"learning_rate": 2.7837204886385217e-06,
|
| 1703 |
+
"loss": 1.4401,
|
| 1704 |
+
"step": 136000
|
| 1705 |
+
},
|
| 1706 |
+
{
|
| 1707 |
+
"epoch": 8.64,
|
| 1708 |
+
"learning_rate": 2.7204253433761635e-06,
|
| 1709 |
+
"loss": 1.4423,
|
| 1710 |
+
"step": 136500
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 8.67,
|
| 1714 |
+
"learning_rate": 2.657130198113805e-06,
|
| 1715 |
+
"loss": 1.4266,
|
| 1716 |
+
"step": 137000
|
| 1717 |
+
},
|
| 1718 |
+
{
|
| 1719 |
+
"epoch": 8.7,
|
| 1720 |
+
"learning_rate": 2.5938350528514466e-06,
|
| 1721 |
+
"loss": 1.4406,
|
| 1722 |
+
"step": 137500
|
| 1723 |
+
},
|
| 1724 |
+
{
|
| 1725 |
+
"epoch": 8.73,
|
| 1726 |
+
"learning_rate": 2.530539907589088e-06,
|
| 1727 |
+
"loss": 1.441,
|
| 1728 |
+
"step": 138000
|
| 1729 |
+
},
|
| 1730 |
+
{
|
| 1731 |
+
"epoch": 8.77,
|
| 1732 |
+
"learning_rate": 2.4672447623267296e-06,
|
| 1733 |
+
"loss": 1.4551,
|
| 1734 |
+
"step": 138500
|
| 1735 |
+
},
|
| 1736 |
+
{
|
| 1737 |
+
"epoch": 8.8,
|
| 1738 |
+
"learning_rate": 2.4039496170643714e-06,
|
| 1739 |
+
"loss": 1.4452,
|
| 1740 |
+
"step": 139000
|
| 1741 |
+
},
|
| 1742 |
+
{
|
| 1743 |
+
"epoch": 8.83,
|
| 1744 |
+
"learning_rate": 2.340654471802013e-06,
|
| 1745 |
+
"loss": 1.4392,
|
| 1746 |
+
"step": 139500
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 8.86,
|
| 1750 |
+
"learning_rate": 2.2773593265396545e-06,
|
| 1751 |
+
"loss": 1.4361,
|
| 1752 |
+
"step": 140000
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 8.89,
|
| 1756 |
+
"learning_rate": 2.2140641812772962e-06,
|
| 1757 |
+
"loss": 1.4313,
|
| 1758 |
+
"step": 140500
|
| 1759 |
+
},
|
| 1760 |
+
{
|
| 1761 |
+
"epoch": 8.92,
|
| 1762 |
+
"learning_rate": 2.1507690360149376e-06,
|
| 1763 |
+
"loss": 1.4323,
|
| 1764 |
+
"step": 141000
|
| 1765 |
+
},
|
| 1766 |
+
{
|
| 1767 |
+
"epoch": 8.96,
|
| 1768 |
+
"learning_rate": 2.0874738907525793e-06,
|
| 1769 |
+
"loss": 1.4266,
|
| 1770 |
+
"step": 141500
|
| 1771 |
+
},
|
| 1772 |
+
{
|
| 1773 |
+
"epoch": 8.99,
|
| 1774 |
+
"learning_rate": 2.024178745490221e-06,
|
| 1775 |
+
"loss": 1.4354,
|
| 1776 |
+
"step": 142000
|
| 1777 |
+
},
|
| 1778 |
+
{
|
| 1779 |
+
"epoch": 9.0,
|
| 1780 |
+
"eval_loss": 1.2341375350952148,
|
| 1781 |
+
"eval_runtime": 642.6304,
|
| 1782 |
+
"eval_samples_per_second": 393.347,
|
| 1783 |
+
"eval_steps_per_second": 24.585,
|
| 1784 |
+
"step": 142191
|
| 1785 |
+
},
|
| 1786 |
+
{
|
| 1787 |
+
"epoch": 9.02,
|
| 1788 |
+
"learning_rate": 1.960883600227863e-06,
|
| 1789 |
+
"loss": 1.4034,
|
| 1790 |
+
"step": 142500
|
| 1791 |
+
},
|
| 1792 |
+
{
|
| 1793 |
+
"epoch": 9.05,
|
| 1794 |
+
"learning_rate": 1.8975884549655044e-06,
|
| 1795 |
+
"loss": 1.3966,
|
| 1796 |
+
"step": 143000
|
| 1797 |
+
},
|
| 1798 |
+
{
|
| 1799 |
+
"epoch": 9.08,
|
| 1800 |
+
"learning_rate": 1.834293309703146e-06,
|
| 1801 |
+
"loss": 1.3921,
|
| 1802 |
+
"step": 143500
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 9.11,
|
| 1806 |
+
"learning_rate": 1.7709981644407874e-06,
|
| 1807 |
+
"loss": 1.396,
|
| 1808 |
+
"step": 144000
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 9.15,
|
| 1812 |
+
"learning_rate": 1.7077030191784292e-06,
|
| 1813 |
+
"loss": 1.411,
|
| 1814 |
+
"step": 144500
|
| 1815 |
+
},
|
| 1816 |
+
{
|
| 1817 |
+
"epoch": 9.18,
|
| 1818 |
+
"learning_rate": 1.6444078739160707e-06,
|
| 1819 |
+
"loss": 1.406,
|
| 1820 |
+
"step": 145000
|
| 1821 |
+
},
|
| 1822 |
+
{
|
| 1823 |
+
"epoch": 9.21,
|
| 1824 |
+
"learning_rate": 1.5811127286537123e-06,
|
| 1825 |
+
"loss": 1.407,
|
| 1826 |
+
"step": 145500
|
| 1827 |
+
},
|
| 1828 |
+
{
|
| 1829 |
+
"epoch": 9.24,
|
| 1830 |
+
"learning_rate": 1.5178175833913538e-06,
|
| 1831 |
+
"loss": 1.4182,
|
| 1832 |
+
"step": 146000
|
| 1833 |
+
},
|
| 1834 |
+
{
|
| 1835 |
+
"epoch": 9.27,
|
| 1836 |
+
"learning_rate": 1.4545224381289958e-06,
|
| 1837 |
+
"loss": 1.4116,
|
| 1838 |
+
"step": 146500
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 9.3,
|
| 1842 |
+
"learning_rate": 1.3912272928666373e-06,
|
| 1843 |
+
"loss": 1.4166,
|
| 1844 |
+
"step": 147000
|
| 1845 |
+
},
|
| 1846 |
+
{
|
| 1847 |
+
"epoch": 9.34,
|
| 1848 |
+
"learning_rate": 1.3279321476042789e-06,
|
| 1849 |
+
"loss": 1.4063,
|
| 1850 |
+
"step": 147500
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 9.37,
|
| 1854 |
+
"learning_rate": 1.2646370023419204e-06,
|
| 1855 |
+
"loss": 1.4025,
|
| 1856 |
+
"step": 148000
|
| 1857 |
+
},
|
| 1858 |
+
{
|
| 1859 |
+
"epoch": 9.4,
|
| 1860 |
+
"learning_rate": 1.2013418570795622e-06,
|
| 1861 |
+
"loss": 1.4061,
|
| 1862 |
+
"step": 148500
|
| 1863 |
+
},
|
| 1864 |
+
{
|
| 1865 |
+
"epoch": 9.43,
|
| 1866 |
+
"learning_rate": 1.1380467118172037e-06,
|
| 1867 |
+
"loss": 1.4066,
|
| 1868 |
+
"step": 149000
|
| 1869 |
+
},
|
| 1870 |
+
{
|
| 1871 |
+
"epoch": 9.46,
|
| 1872 |
+
"learning_rate": 1.0747515665548455e-06,
|
| 1873 |
+
"loss": 1.4152,
|
| 1874 |
+
"step": 149500
|
| 1875 |
+
},
|
| 1876 |
+
{
|
| 1877 |
+
"epoch": 9.49,
|
| 1878 |
+
"learning_rate": 1.011456421292487e-06,
|
| 1879 |
+
"loss": 1.417,
|
| 1880 |
+
"step": 150000
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 9.53,
|
| 1884 |
+
"learning_rate": 9.481612760301285e-07,
|
| 1885 |
+
"loss": 1.411,
|
| 1886 |
+
"step": 150500
|
| 1887 |
+
},
|
| 1888 |
+
{
|
| 1889 |
+
"epoch": 9.56,
|
| 1890 |
+
"learning_rate": 8.848661307677701e-07,
|
| 1891 |
+
"loss": 1.4162,
|
| 1892 |
+
"step": 151000
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 9.59,
|
| 1896 |
+
"learning_rate": 8.215709855054118e-07,
|
| 1897 |
+
"loss": 1.4195,
|
| 1898 |
+
"step": 151500
|
| 1899 |
+
},
|
| 1900 |
+
{
|
| 1901 |
+
"epoch": 9.62,
|
| 1902 |
+
"learning_rate": 7.582758402430535e-07,
|
| 1903 |
+
"loss": 1.4226,
|
| 1904 |
+
"step": 152000
|
| 1905 |
+
},
|
| 1906 |
+
{
|
| 1907 |
+
"epoch": 9.65,
|
| 1908 |
+
"learning_rate": 6.94980694980695e-07,
|
| 1909 |
+
"loss": 1.4239,
|
| 1910 |
+
"step": 152500
|
| 1911 |
+
},
|
| 1912 |
+
{
|
| 1913 |
+
"epoch": 9.68,
|
| 1914 |
+
"learning_rate": 6.316855497183366e-07,
|
| 1915 |
+
"loss": 1.4078,
|
| 1916 |
+
"step": 153000
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 9.72,
|
| 1920 |
+
"learning_rate": 5.683904044559782e-07,
|
| 1921 |
+
"loss": 1.4101,
|
| 1922 |
+
"step": 153500
|
| 1923 |
+
},
|
| 1924 |
+
{
|
| 1925 |
+
"epoch": 9.75,
|
| 1926 |
+
"learning_rate": 5.050952591936199e-07,
|
| 1927 |
+
"loss": 1.416,
|
| 1928 |
+
"step": 154000
|
| 1929 |
+
},
|
| 1930 |
+
{
|
| 1931 |
+
"epoch": 9.78,
|
| 1932 |
+
"learning_rate": 4.418001139312615e-07,
|
| 1933 |
+
"loss": 1.4182,
|
| 1934 |
+
"step": 154500
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 9.81,
|
| 1938 |
+
"learning_rate": 3.785049686689031e-07,
|
| 1939 |
+
"loss": 1.4196,
|
| 1940 |
+
"step": 155000
|
| 1941 |
+
},
|
| 1942 |
+
{
|
| 1943 |
+
"epoch": 9.84,
|
| 1944 |
+
"learning_rate": 3.1520982340654476e-07,
|
| 1945 |
+
"loss": 1.4132,
|
| 1946 |
+
"step": 155500
|
| 1947 |
+
},
|
| 1948 |
+
{
|
| 1949 |
+
"epoch": 9.87,
|
| 1950 |
+
"learning_rate": 2.5191467814418635e-07,
|
| 1951 |
+
"loss": 1.4138,
|
| 1952 |
+
"step": 156000
|
| 1953 |
+
},
|
| 1954 |
+
{
|
| 1955 |
+
"epoch": 9.91,
|
| 1956 |
+
"learning_rate": 1.88619532881828e-07,
|
| 1957 |
+
"loss": 1.4333,
|
| 1958 |
+
"step": 156500
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 9.94,
|
| 1962 |
+
"learning_rate": 1.253243876194696e-07,
|
| 1963 |
+
"loss": 1.413,
|
| 1964 |
+
"step": 157000
|
| 1965 |
+
},
|
| 1966 |
+
{
|
| 1967 |
+
"epoch": 9.97,
|
| 1968 |
+
"learning_rate": 6.202924235711122e-08,
|
| 1969 |
+
"loss": 1.4116,
|
| 1970 |
+
"step": 157500
|
| 1971 |
+
}
|
| 1972 |
+
],
|
| 1973 |
+
"max_steps": 157990,
|
| 1974 |
+
"num_train_epochs": 10,
|
| 1975 |
+
"total_flos": 3.820454731815322e+16,
|
| 1976 |
+
"trial_name": null,
|
| 1977 |
+
"trial_params": null
|
| 1978 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a7d624b7f61620cf3bd86485c3abbf851a0ba124edd2a7b3aec3c4a0d5076e32
|
| 3 |
+
size 4091
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|