Hugging Face's logo

ARTeLab
/
mbart-summarization-fanpage

Summarization
Transformers
PyTorch
Safetensors
Italian
mbart
text2text-generation
Inference Endpoints
Model card Files Community
3
artelabsuper commited on
Commit
9e4b0d9
•
1 Parent(s): dc7aa67

train

Browse files
Files changed (14) hide show
  1. README.md +62 -0
  2. all_results.json +28 -0
  3. config.json +60 -0
  4. eval_results.json +13 -0
  5. generated_predictions.txt +0 -0
  6. predict_results.json +12 -0
  7. pytorch_model.bin +3 -0
  8. sentencepiece.bpe.model +3 -0
  9. special_tokens_map.json +1 -0
  10. tokenizer.json +0 -0
  11. tokenizer_config.json +1 -0
  12. train_results.json +8 -0
  13. trainer_state.json +3259 -0
  14. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - it_IT
4
+ - it_IT
5
+ tags:
6
+ - generated_from_trainer
7
+ metrics:
8
+ - rouge
9
+ model-index:
10
+ - name: summarization_mbart_fanpage4epoch
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # summarization_mbart_fanpage4epoch
18
+
19
+ This model is a fine-tuned version of [facebook/mbart-large-cc25](https://huggingface.co/facebook/mbart-large-cc25) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 2.1833
22
+ - Rouge1: 36.5027
23
+ - Rouge2: 17.4428
24
+ - Rougel: 26.1734
25
+ - Rougelsum: 30.2636
26
+ - Gen Len: 75.2413
27
+
28
+ ## Model description
29
+
30
+ More information needed
31
+
32
+ ## Intended uses & limitations
33
+
34
+ More information needed
35
+
36
+ ## Training and evaluation data
37
+
38
+ More information needed
39
+
40
+ ## Training procedure
41
+
42
+ ### Training hyperparameters
43
+
44
+ The following hyperparameters were used during training:
45
+ - learning_rate: 5e-05
46
+ - train_batch_size: 1
47
+ - eval_batch_size: 1
48
+ - seed: 42
49
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
+ - lr_scheduler_type: linear
51
+ - num_epochs: 4.0
52
+
53
+ ### Training results
54
+
55
+
56
+
57
+ ### Framework versions
58
+
59
+ - Transformers 4.15.0.dev0
60
+ - Pytorch 1.10.0+cu102
61
+ - Datasets 1.15.1
62
+ - Tokenizers 0.10.3
all_results.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "eval_gen_len": 75.2413,
4
+ "eval_loss": 2.183253526687622,
5
+ "eval_rouge1": 36.5027,
6
+ "eval_rouge2": 17.4428,
7
+ "eval_rougeL": 26.1734,
8
+ "eval_rougeLsum": 30.2636,
9
+ "eval_runtime": 11473.5385,
10
+ "eval_samples": 8436,
11
+ "eval_samples_per_second": 0.735,
12
+ "eval_steps_per_second": 0.735,
13
+ "predict_gen_len": 74.9905,
14
+ "predict_loss": 2.1760566234588623,
15
+ "predict_rouge1": 36.5907,
16
+ "predict_rouge2": 17.5075,
17
+ "predict_rougeL": 26.2012,
18
+ "predict_rougeLsum": 30.279,
19
+ "predict_runtime": 11361.1862,
20
+ "predict_samples": 8437,
21
+ "predict_samples_per_second": 0.743,
22
+ "predict_steps_per_second": 0.743,
23
+ "train_loss": 1.6431024962071243,
24
+ "train_runtime": 97099.2276,
25
+ "train_samples": 67492,
26
+ "train_samples_per_second": 2.78,
27
+ "train_steps_per_second": 2.78
28
+ }
config.json ADDED
@@ -0,0 +1,60 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/mbart-large-cc25",
3
+ "_num_labels": 3,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "gelu",
6
+ "add_bias_logits": false,
7
+ "add_final_layer_norm": true,
8
+ "architectures": [
9
+ "MBartForConditionalGeneration"
10
+ ],
11
+ "attention_dropout": 0.0,
12
+ "bos_token_id": 0,
13
+ "classif_dropout": 0.0,
14
+ "classifier_dropout": 0.0,
15
+ "d_model": 1024,
16
+ "decoder_attention_heads": 16,
17
+ "decoder_ffn_dim": 4096,
18
+ "decoder_layerdrop": 0.0,
19
+ "decoder_layers": 12,
20
+ "decoder_start_token_id": 250011,
21
+ "dropout": 0.1,
22
+ "encoder_attention_heads": 16,
23
+ "encoder_ffn_dim": 4096,
24
+ "encoder_layerdrop": 0.0,
25
+ "encoder_layers": 12,
26
+ "eos_token_id": 2,
27
+ "forced_eos_token_id": 2,
28
+ "id2label": {
29
+ "0": "LABEL_0",
30
+ "1": "LABEL_1",
31
+ "2": "LABEL_2"
32
+ },
33
+ "init_std": 0.02,
34
+ "is_encoder_decoder": true,
35
+ "label2id": {
36
+ "LABEL_0": 0,
37
+ "LABEL_1": 1,
38
+ "LABEL_2": 2
39
+ },
40
+ "max_length": 1024,
41
+ "max_position_embeddings": 1024,
42
+ "model_type": "mbart",
43
+ "normalize_before": true,
44
+ "normalize_embedding": true,
45
+ "num_beams": 5,
46
+ "num_hidden_layers": 12,
47
+ "output_past": true,
48
+ "pad_token_id": 1,
49
+ "scale_embedding": true,
50
+ "static_position_embeddings": false,
51
+ "task_specific_params": {
52
+ "translation_en_to_ro": {
53
+ "decoder_start_token_id": 250020
54
+ }
55
+ },
56
+ "torch_dtype": "float32",
57
+ "transformers_version": "4.15.0.dev0",
58
+ "use_cache": true,
59
+ "vocab_size": 250027
60
+ }
eval_results.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "eval_gen_len": 75.2413,
4
+ "eval_loss": 2.183253526687622,
5
+ "eval_rouge1": 36.5027,
6
+ "eval_rouge2": 17.4428,
7
+ "eval_rougeL": 26.1734,
8
+ "eval_rougeLsum": 30.2636,
9
+ "eval_runtime": 11473.5385,
10
+ "eval_samples": 8436,
11
+ "eval_samples_per_second": 0.735,
12
+ "eval_steps_per_second": 0.735
13
+ }
generated_predictions.txt ADDED
The diff for this file is too large to render. See raw diff
 
predict_results.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "predict_gen_len": 74.9905,
3
+ "predict_loss": 2.1760566234588623,
4
+ "predict_rouge1": 36.5907,
5
+ "predict_rouge2": 17.5075,
6
+ "predict_rougeL": 26.2012,
7
+ "predict_rougeLsum": 30.279,
8
+ "predict_runtime": 11361.1862,
9
+ "predict_samples": 8437,
10
+ "predict_samples_per_second": 0.743,
11
+ "predict_steps_per_second": 0.743
12
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:076d472321df30d3a901e1112a47fc000ff6f3e2df4866af72ce2df391968bcd
3
+ size 2444599289
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}, "additional_special_tokens": ["ar_AR", "cs_CZ", "de_DE", "en_XX", "es_XX", "et_EE", "fi_FI", "fr_XX", "gu_IN", "hi_IN", "it_IT", "ja_XX", "kk_KZ", "ko_KR", "lt_LT", "lv_LV", "my_MM", "ne_NP", "nl_XX", "ro_RO", "ru_RU", "si_LK", "tr_TR", "vi_VN", "zh_CN"]}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "cls_token": "<s>", "unk_token": "<unk>", "pad_token": "<pad>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": true, "__type": "AddedToken"}, "src_lang": null, "tgt_lang": null, "additional_special_tokens": null, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "facebook/mbart-large-cc25", "tokenizer_class": "MBartTokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 4.0,
3
+ "train_loss": 1.6431024962071243,
4
+ "train_runtime": 97099.2276,
5
+ "train_samples": 67492,
6
+ "train_samples_per_second": 2.78,
7
+ "train_steps_per_second": 2.78
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,3259 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.0,
5
+ "global_step": 269968,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 4.990739643216974e-05,
13
+ "loss": 2.8507,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.01,
18
+ "learning_rate": 4.9814792864339485e-05,
19
+ "loss": 2.7365,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.02,
24
+ "learning_rate": 4.972218929650922e-05,
25
+ "loss": 2.7102,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.03,
30
+ "learning_rate": 4.9629585728678954e-05,
31
+ "loss": 2.7208,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.04,
36
+ "learning_rate": 4.953698216084869e-05,
37
+ "loss": 2.7067,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.04,
42
+ "learning_rate": 4.9444378593018436e-05,
43
+ "loss": 2.6698,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.05,
48
+ "learning_rate": 4.9351775025188174e-05,
49
+ "loss": 2.6546,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.06,
54
+ "learning_rate": 4.925917145735791e-05,
55
+ "loss": 2.6446,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.07,
60
+ "learning_rate": 4.916656788952765e-05,
61
+ "loss": 2.6289,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.07,
66
+ "learning_rate": 4.9073964321697394e-05,
67
+ "loss": 2.6532,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.08,
72
+ "learning_rate": 4.8981360753867125e-05,
73
+ "loss": 2.5513,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.09,
78
+ "learning_rate": 4.888875718603686e-05,
79
+ "loss": 2.5434,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.1,
84
+ "learning_rate": 4.87961536182066e-05,
85
+ "loss": 2.5836,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.1,
90
+ "learning_rate": 4.8703550050376345e-05,
91
+ "loss": 2.5433,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.11,
96
+ "learning_rate": 4.861094648254608e-05,
97
+ "loss": 2.5794,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.12,
102
+ "learning_rate": 4.851834291471582e-05,
103
+ "loss": 2.5457,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.13,
108
+ "learning_rate": 4.842573934688556e-05,
109
+ "loss": 2.5119,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.13,
114
+ "learning_rate": 4.83331357790553e-05,
115
+ "loss": 2.525,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.14,
120
+ "learning_rate": 4.8240532211225034e-05,
121
+ "loss": 2.5471,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 0.15,
126
+ "learning_rate": 4.814792864339477e-05,
127
+ "loss": 2.5862,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 0.16,
132
+ "learning_rate": 4.805532507556451e-05,
133
+ "loss": 2.5391,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 0.16,
138
+ "learning_rate": 4.7962721507734254e-05,
139
+ "loss": 2.5123,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 0.17,
144
+ "learning_rate": 4.787011793990399e-05,
145
+ "loss": 2.5144,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 0.18,
150
+ "learning_rate": 4.777751437207373e-05,
151
+ "loss": 2.5659,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 0.19,
156
+ "learning_rate": 4.768491080424347e-05,
157
+ "loss": 2.5536,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 0.19,
162
+ "learning_rate": 4.7592307236413205e-05,
163
+ "loss": 2.4725,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 0.2,
168
+ "learning_rate": 4.749970366858294e-05,
169
+ "loss": 2.4336,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 0.21,
174
+ "learning_rate": 4.740710010075268e-05,
175
+ "loss": 2.5392,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 0.21,
180
+ "learning_rate": 4.7314496532922425e-05,
181
+ "loss": 2.4867,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 0.22,
186
+ "learning_rate": 4.722189296509216e-05,
187
+ "loss": 2.4775,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 0.23,
192
+ "learning_rate": 4.71292893972619e-05,
193
+ "loss": 2.5399,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 0.24,
198
+ "learning_rate": 4.703668582943164e-05,
199
+ "loss": 2.4984,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 0.24,
204
+ "learning_rate": 4.6944082261601376e-05,
205
+ "loss": 2.5248,
206
+ "step": 16500
207
+ },
208
+ {
209
+ "epoch": 0.25,
210
+ "learning_rate": 4.6851478693771114e-05,
211
+ "loss": 2.4343,
212
+ "step": 17000
213
+ },
214
+ {
215
+ "epoch": 0.26,
216
+ "learning_rate": 4.675887512594085e-05,
217
+ "loss": 2.4473,
218
+ "step": 17500
219
+ },
220
+ {
221
+ "epoch": 0.27,
222
+ "learning_rate": 4.666627155811059e-05,
223
+ "loss": 2.455,
224
+ "step": 18000
225
+ },
226
+ {
227
+ "epoch": 0.27,
228
+ "learning_rate": 4.6573667990280334e-05,
229
+ "loss": 2.4335,
230
+ "step": 18500
231
+ },
232
+ {
233
+ "epoch": 0.28,
234
+ "learning_rate": 4.648106442245007e-05,
235
+ "loss": 2.4983,
236
+ "step": 19000
237
+ },
238
+ {
239
+ "epoch": 0.29,
240
+ "learning_rate": 4.638846085461981e-05,
241
+ "loss": 2.4334,
242
+ "step": 19500
243
+ },
244
+ {
245
+ "epoch": 0.3,
246
+ "learning_rate": 4.629585728678955e-05,
247
+ "loss": 2.4077,
248
+ "step": 20000
249
+ },
250
+ {
251
+ "epoch": 0.3,
252
+ "learning_rate": 4.6203253718959285e-05,
253
+ "loss": 2.4161,
254
+ "step": 20500
255
+ },
256
+ {
257
+ "epoch": 0.31,
258
+ "learning_rate": 4.611065015112902e-05,
259
+ "loss": 2.4169,
260
+ "step": 21000
261
+ },
262
+ {
263
+ "epoch": 0.32,
264
+ "learning_rate": 4.601804658329876e-05,
265
+ "loss": 2.4123,
266
+ "step": 21500
267
+ },
268
+ {
269
+ "epoch": 0.33,
270
+ "learning_rate": 4.59254430154685e-05,
271
+ "loss": 2.4486,
272
+ "step": 22000
273
+ },
274
+ {
275
+ "epoch": 0.33,
276
+ "learning_rate": 4.583283944763824e-05,
277
+ "loss": 2.3579,
278
+ "step": 22500
279
+ },
280
+ {
281
+ "epoch": 0.34,
282
+ "learning_rate": 4.574023587980798e-05,
283
+ "loss": 2.355,
284
+ "step": 23000
285
+ },
286
+ {
287
+ "epoch": 0.35,
288
+ "learning_rate": 4.564763231197772e-05,
289
+ "loss": 2.3947,
290
+ "step": 23500
291
+ },
292
+ {
293
+ "epoch": 0.36,
294
+ "learning_rate": 4.555502874414746e-05,
295
+ "loss": 2.4513,
296
+ "step": 24000
297
+ },
298
+ {
299
+ "epoch": 0.36,
300
+ "learning_rate": 4.5462425176317195e-05,
301
+ "loss": 2.4414,
302
+ "step": 24500
303
+ },
304
+ {
305
+ "epoch": 0.37,
306
+ "learning_rate": 4.536982160848693e-05,
307
+ "loss": 2.4163,
308
+ "step": 25000
309
+ },
310
+ {
311
+ "epoch": 0.38,
312
+ "learning_rate": 4.527721804065667e-05,
313
+ "loss": 2.4108,
314
+ "step": 25500
315
+ },
316
+ {
317
+ "epoch": 0.39,
318
+ "learning_rate": 4.518461447282641e-05,
319
+ "loss": 2.3692,
320
+ "step": 26000
321
+ },
322
+ {
323
+ "epoch": 0.39,
324
+ "learning_rate": 4.509201090499615e-05,
325
+ "loss": 2.4022,
326
+ "step": 26500
327
+ },
328
+ {
329
+ "epoch": 0.4,
330
+ "learning_rate": 4.499940733716589e-05,
331
+ "loss": 2.3209,
332
+ "step": 27000
333
+ },
334
+ {
335
+ "epoch": 0.41,
336
+ "learning_rate": 4.490680376933563e-05,
337
+ "loss": 2.4336,
338
+ "step": 27500
339
+ },
340
+ {
341
+ "epoch": 0.41,
342
+ "learning_rate": 4.4814200201505366e-05,
343
+ "loss": 2.3763,
344
+ "step": 28000
345
+ },
346
+ {
347
+ "epoch": 0.42,
348
+ "learning_rate": 4.4721596633675104e-05,
349
+ "loss": 2.4228,
350
+ "step": 28500
351
+ },
352
+ {
353
+ "epoch": 0.43,
354
+ "learning_rate": 4.462899306584484e-05,
355
+ "loss": 2.4199,
356
+ "step": 29000
357
+ },
358
+ {
359
+ "epoch": 0.44,
360
+ "learning_rate": 4.453638949801458e-05,
361
+ "loss": 2.3636,
362
+ "step": 29500
363
+ },
364
+ {
365
+ "epoch": 0.44,
366
+ "learning_rate": 4.4443785930184324e-05,
367
+ "loss": 2.4332,
368
+ "step": 30000
369
+ },
370
+ {
371
+ "epoch": 0.45,
372
+ "learning_rate": 4.435118236235406e-05,
373
+ "loss": 2.4513,
374
+ "step": 30500
375
+ },
376
+ {
377
+ "epoch": 0.46,
378
+ "learning_rate": 4.42585787945238e-05,
379
+ "loss": 2.3083,
380
+ "step": 31000
381
+ },
382
+ {
383
+ "epoch": 0.47,
384
+ "learning_rate": 4.416597522669354e-05,
385
+ "loss": 2.3773,
386
+ "step": 31500
387
+ },
388
+ {
389
+ "epoch": 0.47,
390
+ "learning_rate": 4.4073371658863275e-05,
391
+ "loss": 2.3017,
392
+ "step": 32000
393
+ },
394
+ {
395
+ "epoch": 0.48,
396
+ "learning_rate": 4.398076809103301e-05,
397
+ "loss": 2.3439,
398
+ "step": 32500
399
+ },
400
+ {
401
+ "epoch": 0.49,
402
+ "learning_rate": 4.388816452320275e-05,
403
+ "loss": 2.3497,
404
+ "step": 33000
405
+ },
406
+ {
407
+ "epoch": 0.5,
408
+ "learning_rate": 4.379556095537249e-05,
409
+ "loss": 2.3305,
410
+ "step": 33500
411
+ },
412
+ {
413
+ "epoch": 0.5,
414
+ "learning_rate": 4.370295738754223e-05,
415
+ "loss": 2.3532,
416
+ "step": 34000
417
+ },
418
+ {
419
+ "epoch": 0.51,
420
+ "learning_rate": 4.361035381971197e-05,
421
+ "loss": 2.2562,
422
+ "step": 34500
423
+ },
424
+ {
425
+ "epoch": 0.52,
426
+ "learning_rate": 4.351775025188171e-05,
427
+ "loss": 2.3071,
428
+ "step": 35000
429
+ },
430
+ {
431
+ "epoch": 0.53,
432
+ "learning_rate": 4.3425146684051446e-05,
433
+ "loss": 2.3573,
434
+ "step": 35500
435
+ },
436
+ {
437
+ "epoch": 0.53,
438
+ "learning_rate": 4.3332543116221184e-05,
439
+ "loss": 2.3095,
440
+ "step": 36000
441
+ },
442
+ {
443
+ "epoch": 0.54,
444
+ "learning_rate": 4.323993954839092e-05,
445
+ "loss": 2.3321,
446
+ "step": 36500
447
+ },
448
+ {
449
+ "epoch": 0.55,
450
+ "learning_rate": 4.314733598056066e-05,
451
+ "loss": 2.2776,
452
+ "step": 37000
453
+ },
454
+ {
455
+ "epoch": 0.56,
456
+ "learning_rate": 4.30547324127304e-05,
457
+ "loss": 2.2757,
458
+ "step": 37500
459
+ },
460
+ {
461
+ "epoch": 0.56,
462
+ "learning_rate": 4.296212884490014e-05,
463
+ "loss": 2.3101,
464
+ "step": 38000
465
+ },
466
+ {
467
+ "epoch": 0.57,
468
+ "learning_rate": 4.286952527706988e-05,
469
+ "loss": 2.3099,
470
+ "step": 38500
471
+ },
472
+ {
473
+ "epoch": 0.58,
474
+ "learning_rate": 4.277692170923962e-05,
475
+ "loss": 2.3121,
476
+ "step": 39000
477
+ },
478
+ {
479
+ "epoch": 0.59,
480
+ "learning_rate": 4.2684318141409355e-05,
481
+ "loss": 2.3046,
482
+ "step": 39500
483
+ },
484
+ {
485
+ "epoch": 0.59,
486
+ "learning_rate": 4.259171457357909e-05,
487
+ "loss": 2.2414,
488
+ "step": 40000
489
+ },
490
+ {
491
+ "epoch": 0.6,
492
+ "learning_rate": 4.249911100574883e-05,
493
+ "loss": 2.2979,
494
+ "step": 40500
495
+ },
496
+ {
497
+ "epoch": 0.61,
498
+ "learning_rate": 4.240650743791857e-05,
499
+ "loss": 2.2733,
500
+ "step": 41000
501
+ },
502
+ {
503
+ "epoch": 0.61,
504
+ "learning_rate": 4.2313903870088306e-05,
505
+ "loss": 2.3319,
506
+ "step": 41500
507
+ },
508
+ {
509
+ "epoch": 0.62,
510
+ "learning_rate": 4.222130030225805e-05,
511
+ "loss": 2.3224,
512
+ "step": 42000
513
+ },
514
+ {
515
+ "epoch": 0.63,
516
+ "learning_rate": 4.212869673442779e-05,
517
+ "loss": 2.2873,
518
+ "step": 42500
519
+ },
520
+ {
521
+ "epoch": 0.64,
522
+ "learning_rate": 4.2036093166597526e-05,
523
+ "loss": 2.3584,
524
+ "step": 43000
525
+ },
526
+ {
527
+ "epoch": 0.64,
528
+ "learning_rate": 4.1943489598767264e-05,
529
+ "loss": 2.265,
530
+ "step": 43500
531
+ },
532
+ {
533
+ "epoch": 0.65,
534
+ "learning_rate": 4.1850886030937e-05,
535
+ "loss": 2.2914,
536
+ "step": 44000
537
+ },
538
+ {
539
+ "epoch": 0.66,
540
+ "learning_rate": 4.175828246310674e-05,
541
+ "loss": 2.2585,
542
+ "step": 44500
543
+ },
544
+ {
545
+ "epoch": 0.67,
546
+ "learning_rate": 4.166567889527648e-05,
547
+ "loss": 2.2645,
548
+ "step": 45000
549
+ },
550
+ {
551
+ "epoch": 0.67,
552
+ "learning_rate": 4.157307532744622e-05,
553
+ "loss": 2.3289,
554
+ "step": 45500
555
+ },
556
+ {
557
+ "epoch": 0.68,
558
+ "learning_rate": 4.148047175961596e-05,
559
+ "loss": 2.2665,
560
+ "step": 46000
561
+ },
562
+ {
563
+ "epoch": 0.69,
564
+ "learning_rate": 4.13878681917857e-05,
565
+ "loss": 2.2714,
566
+ "step": 46500
567
+ },
568
+ {
569
+ "epoch": 0.7,
570
+ "learning_rate": 4.129526462395543e-05,
571
+ "loss": 2.2622,
572
+ "step": 47000
573
+ },
574
+ {
575
+ "epoch": 0.7,
576
+ "learning_rate": 4.120266105612517e-05,
577
+ "loss": 2.2456,
578
+ "step": 47500
579
+ },
580
+ {
581
+ "epoch": 0.71,
582
+ "learning_rate": 4.111005748829491e-05,
583
+ "loss": 2.3016,
584
+ "step": 48000
585
+ },
586
+ {
587
+ "epoch": 0.72,
588
+ "learning_rate": 4.101745392046465e-05,
589
+ "loss": 2.2608,
590
+ "step": 48500
591
+ },
592
+ {
593
+ "epoch": 0.73,
594
+ "learning_rate": 4.092485035263439e-05,
595
+ "loss": 2.2983,
596
+ "step": 49000
597
+ },
598
+ {
599
+ "epoch": 0.73,
600
+ "learning_rate": 4.083224678480413e-05,
601
+ "loss": 2.2725,
602
+ "step": 49500
603
+ },
604
+ {
605
+ "epoch": 0.74,
606
+ "learning_rate": 4.073964321697387e-05,
607
+ "loss": 2.2304,
608
+ "step": 50000
609
+ },
610
+ {
611
+ "epoch": 0.75,
612
+ "learning_rate": 4.064703964914361e-05,
613
+ "loss": 2.3037,
614
+ "step": 50500
615
+ },
616
+ {
617
+ "epoch": 0.76,
618
+ "learning_rate": 4.055443608131334e-05,
619
+ "loss": 2.2449,
620
+ "step": 51000
621
+ },
622
+ {
623
+ "epoch": 0.76,
624
+ "learning_rate": 4.046183251348308e-05,
625
+ "loss": 2.256,
626
+ "step": 51500
627
+ },
628
+ {
629
+ "epoch": 0.77,
630
+ "learning_rate": 4.036922894565282e-05,
631
+ "loss": 2.3002,
632
+ "step": 52000
633
+ },
634
+ {
635
+ "epoch": 0.78,
636
+ "learning_rate": 4.027662537782256e-05,
637
+ "loss": 2.2611,
638
+ "step": 52500
639
+ },
640
+ {
641
+ "epoch": 0.79,
642
+ "learning_rate": 4.0184021809992296e-05,
643
+ "loss": 2.2351,
644
+ "step": 53000
645
+ },
646
+ {
647
+ "epoch": 0.79,
648
+ "learning_rate": 4.009141824216204e-05,
649
+ "loss": 2.2351,
650
+ "step": 53500
651
+ },
652
+ {
653
+ "epoch": 0.8,
654
+ "learning_rate": 3.999881467433178e-05,
655
+ "loss": 2.2658,
656
+ "step": 54000
657
+ },
658
+ {
659
+ "epoch": 0.81,
660
+ "learning_rate": 3.990621110650151e-05,
661
+ "loss": 2.2394,
662
+ "step": 54500
663
+ },
664
+ {
665
+ "epoch": 0.81,
666
+ "learning_rate": 3.981360753867125e-05,
667
+ "loss": 2.2527,
668
+ "step": 55000
669
+ },
670
+ {
671
+ "epoch": 0.82,
672
+ "learning_rate": 3.972100397084099e-05,
673
+ "loss": 2.2554,
674
+ "step": 55500
675
+ },
676
+ {
677
+ "epoch": 0.83,
678
+ "learning_rate": 3.962840040301073e-05,
679
+ "loss": 2.2116,
680
+ "step": 56000
681
+ },
682
+ {
683
+ "epoch": 0.84,
684
+ "learning_rate": 3.953579683518047e-05,
685
+ "loss": 2.2459,
686
+ "step": 56500
687
+ },
688
+ {
689
+ "epoch": 0.84,
690
+ "learning_rate": 3.9443193267350205e-05,
691
+ "loss": 2.2093,
692
+ "step": 57000
693
+ },
694
+ {
695
+ "epoch": 0.85,
696
+ "learning_rate": 3.935058969951995e-05,
697
+ "loss": 2.2384,
698
+ "step": 57500
699
+ },
700
+ {
701
+ "epoch": 0.86,
702
+ "learning_rate": 3.925798613168968e-05,
703
+ "loss": 2.222,
704
+ "step": 58000
705
+ },
706
+ {
707
+ "epoch": 0.87,
708
+ "learning_rate": 3.916538256385942e-05,
709
+ "loss": 2.1714,
710
+ "step": 58500
711
+ },
712
+ {
713
+ "epoch": 0.87,
714
+ "learning_rate": 3.907277899602916e-05,
715
+ "loss": 2.2678,
716
+ "step": 59000
717
+ },
718
+ {
719
+ "epoch": 0.88,
720
+ "learning_rate": 3.89801754281989e-05,
721
+ "loss": 2.259,
722
+ "step": 59500
723
+ },
724
+ {
725
+ "epoch": 0.89,
726
+ "learning_rate": 3.888757186036864e-05,
727
+ "loss": 2.1827,
728
+ "step": 60000
729
+ },
730
+ {
731
+ "epoch": 0.9,
732
+ "learning_rate": 3.8794968292538376e-05,
733
+ "loss": 2.2199,
734
+ "step": 60500
735
+ },
736
+ {
737
+ "epoch": 0.9,
738
+ "learning_rate": 3.870236472470812e-05,
739
+ "loss": 2.1635,
740
+ "step": 61000
741
+ },
742
+ {
743
+ "epoch": 0.91,
744
+ "learning_rate": 3.860976115687786e-05,
745
+ "loss": 2.126,
746
+ "step": 61500
747
+ },
748
+ {
749
+ "epoch": 0.92,
750
+ "learning_rate": 3.851715758904759e-05,
751
+ "loss": 2.277,
752
+ "step": 62000
753
+ },
754
+ {
755
+ "epoch": 0.93,
756
+ "learning_rate": 3.842455402121733e-05,
757
+ "loss": 2.1814,
758
+ "step": 62500
759
+ },
760
+ {
761
+ "epoch": 0.93,
762
+ "learning_rate": 3.833195045338707e-05,
763
+ "loss": 2.2185,
764
+ "step": 63000
765
+ },
766
+ {
767
+ "epoch": 0.94,
768
+ "learning_rate": 3.823934688555681e-05,
769
+ "loss": 2.1325,
770
+ "step": 63500
771
+ },
772
+ {
773
+ "epoch": 0.95,
774
+ "learning_rate": 3.814674331772655e-05,
775
+ "loss": 2.1752,
776
+ "step": 64000
777
+ },
778
+ {
779
+ "epoch": 0.96,
780
+ "learning_rate": 3.8054139749896285e-05,
781
+ "loss": 2.2558,
782
+ "step": 64500
783
+ },
784
+ {
785
+ "epoch": 0.96,
786
+ "learning_rate": 3.796153618206603e-05,
787
+ "loss": 2.1914,
788
+ "step": 65000
789
+ },
790
+ {
791
+ "epoch": 0.97,
792
+ "learning_rate": 3.786893261423576e-05,
793
+ "loss": 2.1828,
794
+ "step": 65500
795
+ },
796
+ {
797
+ "epoch": 0.98,
798
+ "learning_rate": 3.77763290464055e-05,
799
+ "loss": 2.1864,
800
+ "step": 66000
801
+ },
802
+ {
803
+ "epoch": 0.99,
804
+ "learning_rate": 3.7683725478575236e-05,
805
+ "loss": 2.2374,
806
+ "step": 66500
807
+ },
808
+ {
809
+ "epoch": 0.99,
810
+ "learning_rate": 3.759112191074498e-05,
811
+ "loss": 2.2294,
812
+ "step": 67000
813
+ },
814
+ {
815
+ "epoch": 1.0,
816
+ "learning_rate": 3.749851834291472e-05,
817
+ "loss": 2.1764,
818
+ "step": 67500
819
+ },
820
+ {
821
+ "epoch": 1.01,
822
+ "learning_rate": 3.7405914775084456e-05,
823
+ "loss": 1.8288,
824
+ "step": 68000
825
+ },
826
+ {
827
+ "epoch": 1.01,
828
+ "learning_rate": 3.7313311207254194e-05,
829
+ "loss": 1.8335,
830
+ "step": 68500
831
+ },
832
+ {
833
+ "epoch": 1.02,
834
+ "learning_rate": 3.722070763942393e-05,
835
+ "loss": 1.789,
836
+ "step": 69000
837
+ },
838
+ {
839
+ "epoch": 1.03,
840
+ "learning_rate": 3.712810407159367e-05,
841
+ "loss": 1.8011,
842
+ "step": 69500
843
+ },
844
+ {
845
+ "epoch": 1.04,
846
+ "learning_rate": 3.703550050376341e-05,
847
+ "loss": 1.8464,
848
+ "step": 70000
849
+ },
850
+ {
851
+ "epoch": 1.04,
852
+ "learning_rate": 3.6942896935933145e-05,
853
+ "loss": 1.9109,
854
+ "step": 70500
855
+ },
856
+ {
857
+ "epoch": 1.05,
858
+ "learning_rate": 3.685029336810289e-05,
859
+ "loss": 1.8988,
860
+ "step": 71000
861
+ },
862
+ {
863
+ "epoch": 1.06,
864
+ "learning_rate": 3.675768980027263e-05,
865
+ "loss": 1.8336,
866
+ "step": 71500
867
+ },
868
+ {
869
+ "epoch": 1.07,
870
+ "learning_rate": 3.6665086232442365e-05,
871
+ "loss": 1.8093,
872
+ "step": 72000
873
+ },
874
+ {
875
+ "epoch": 1.07,
876
+ "learning_rate": 3.657248266461211e-05,
877
+ "loss": 1.8718,
878
+ "step": 72500
879
+ },
880
+ {
881
+ "epoch": 1.08,
882
+ "learning_rate": 3.647987909678184e-05,
883
+ "loss": 1.8582,
884
+ "step": 73000
885
+ },
886
+ {
887
+ "epoch": 1.09,
888
+ "learning_rate": 3.638727552895158e-05,
889
+ "loss": 1.8773,
890
+ "step": 73500
891
+ },
892
+ {
893
+ "epoch": 1.1,
894
+ "learning_rate": 3.6294671961121317e-05,
895
+ "loss": 1.845,
896
+ "step": 74000
897
+ },
898
+ {
899
+ "epoch": 1.1,
900
+ "learning_rate": 3.620206839329106e-05,
901
+ "loss": 1.8208,
902
+ "step": 74500
903
+ },
904
+ {
905
+ "epoch": 1.11,
906
+ "learning_rate": 3.61094648254608e-05,
907
+ "loss": 1.866,
908
+ "step": 75000
909
+ },
910
+ {
911
+ "epoch": 1.12,
912
+ "learning_rate": 3.601686125763054e-05,
913
+ "loss": 1.8354,
914
+ "step": 75500
915
+ },
916
+ {
917
+ "epoch": 1.13,
918
+ "learning_rate": 3.5924257689800274e-05,
919
+ "loss": 1.874,
920
+ "step": 76000
921
+ },
922
+ {
923
+ "epoch": 1.13,
924
+ "learning_rate": 3.583165412197001e-05,
925
+ "loss": 1.8157,
926
+ "step": 76500
927
+ },
928
+ {
929
+ "epoch": 1.14,
930
+ "learning_rate": 3.573905055413975e-05,
931
+ "loss": 1.8941,
932
+ "step": 77000
933
+ },
934
+ {
935
+ "epoch": 1.15,
936
+ "learning_rate": 3.564644698630949e-05,
937
+ "loss": 1.7856,
938
+ "step": 77500
939
+ },
940
+ {
941
+ "epoch": 1.16,
942
+ "learning_rate": 3.5553843418479226e-05,
943
+ "loss": 1.8355,
944
+ "step": 78000
945
+ },
946
+ {
947
+ "epoch": 1.16,
948
+ "learning_rate": 3.546123985064897e-05,
949
+ "loss": 1.8542,
950
+ "step": 78500
951
+ },
952
+ {
953
+ "epoch": 1.17,
954
+ "learning_rate": 3.536863628281871e-05,
955
+ "loss": 1.8734,
956
+ "step": 79000
957
+ },
958
+ {
959
+ "epoch": 1.18,
960
+ "learning_rate": 3.5276032714988446e-05,
961
+ "loss": 1.8471,
962
+ "step": 79500
963
+ },
964
+ {
965
+ "epoch": 1.19,
966
+ "learning_rate": 3.5183429147158183e-05,
967
+ "loss": 1.8843,
968
+ "step": 80000
969
+ },
970
+ {
971
+ "epoch": 1.19,
972
+ "learning_rate": 3.509082557932792e-05,
973
+ "loss": 1.8549,
974
+ "step": 80500
975
+ },
976
+ {
977
+ "epoch": 1.2,
978
+ "learning_rate": 3.499822201149766e-05,
979
+ "loss": 1.8794,
980
+ "step": 81000
981
+ },
982
+ {
983
+ "epoch": 1.21,
984
+ "learning_rate": 3.49056184436674e-05,
985
+ "loss": 1.8123,
986
+ "step": 81500
987
+ },
988
+ {
989
+ "epoch": 1.21,
990
+ "learning_rate": 3.4813014875837135e-05,
991
+ "loss": 1.9222,
992
+ "step": 82000
993
+ },
994
+ {
995
+ "epoch": 1.22,
996
+ "learning_rate": 3.472041130800688e-05,
997
+ "loss": 1.8118,
998
+ "step": 82500
999
+ },
1000
+ {
1001
+ "epoch": 1.23,
1002
+ "learning_rate": 3.462780774017662e-05,
1003
+ "loss": 1.8444,
1004
+ "step": 83000
1005
+ },
1006
+ {
1007
+ "epoch": 1.24,
1008
+ "learning_rate": 3.4535204172346355e-05,
1009
+ "loss": 1.8205,
1010
+ "step": 83500
1011
+ },
1012
+ {
1013
+ "epoch": 1.24,
1014
+ "learning_rate": 3.444260060451609e-05,
1015
+ "loss": 1.832,
1016
+ "step": 84000
1017
+ },
1018
+ {
1019
+ "epoch": 1.25,
1020
+ "learning_rate": 3.434999703668583e-05,
1021
+ "loss": 1.8681,
1022
+ "step": 84500
1023
+ },
1024
+ {
1025
+ "epoch": 1.26,
1026
+ "learning_rate": 3.425739346885557e-05,
1027
+ "loss": 1.8315,
1028
+ "step": 85000
1029
+ },
1030
+ {
1031
+ "epoch": 1.27,
1032
+ "learning_rate": 3.4164789901025306e-05,
1033
+ "loss": 1.8455,
1034
+ "step": 85500
1035
+ },
1036
+ {
1037
+ "epoch": 1.27,
1038
+ "learning_rate": 3.407218633319505e-05,
1039
+ "loss": 1.8116,
1040
+ "step": 86000
1041
+ },
1042
+ {
1043
+ "epoch": 1.28,
1044
+ "learning_rate": 3.397958276536479e-05,
1045
+ "loss": 1.8428,
1046
+ "step": 86500
1047
+ },
1048
+ {
1049
+ "epoch": 1.29,
1050
+ "learning_rate": 3.3886979197534526e-05,
1051
+ "loss": 1.8807,
1052
+ "step": 87000
1053
+ },
1054
+ {
1055
+ "epoch": 1.3,
1056
+ "learning_rate": 3.3794375629704264e-05,
1057
+ "loss": 1.848,
1058
+ "step": 87500
1059
+ },
1060
+ {
1061
+ "epoch": 1.3,
1062
+ "learning_rate": 3.3701772061874e-05,
1063
+ "loss": 1.8763,
1064
+ "step": 88000
1065
+ },
1066
+ {
1067
+ "epoch": 1.31,
1068
+ "learning_rate": 3.360916849404374e-05,
1069
+ "loss": 1.8613,
1070
+ "step": 88500
1071
+ },
1072
+ {
1073
+ "epoch": 1.32,
1074
+ "learning_rate": 3.351656492621348e-05,
1075
+ "loss": 1.887,
1076
+ "step": 89000
1077
+ },
1078
+ {
1079
+ "epoch": 1.33,
1080
+ "learning_rate": 3.3423961358383215e-05,
1081
+ "loss": 1.8228,
1082
+ "step": 89500
1083
+ },
1084
+ {
1085
+ "epoch": 1.33,
1086
+ "learning_rate": 3.333135779055296e-05,
1087
+ "loss": 1.8495,
1088
+ "step": 90000
1089
+ },
1090
+ {
1091
+ "epoch": 1.34,
1092
+ "learning_rate": 3.32387542227227e-05,
1093
+ "loss": 1.8136,
1094
+ "step": 90500
1095
+ },
1096
+ {
1097
+ "epoch": 1.35,
1098
+ "learning_rate": 3.3146150654892435e-05,
1099
+ "loss": 1.8473,
1100
+ "step": 91000
1101
+ },
1102
+ {
1103
+ "epoch": 1.36,
1104
+ "learning_rate": 3.305354708706217e-05,
1105
+ "loss": 1.8911,
1106
+ "step": 91500
1107
+ },
1108
+ {
1109
+ "epoch": 1.36,
1110
+ "learning_rate": 3.296094351923191e-05,
1111
+ "loss": 1.8469,
1112
+ "step": 92000
1113
+ },
1114
+ {
1115
+ "epoch": 1.37,
1116
+ "learning_rate": 3.286833995140165e-05,
1117
+ "loss": 1.8637,
1118
+ "step": 92500
1119
+ },
1120
+ {
1121
+ "epoch": 1.38,
1122
+ "learning_rate": 3.2775736383571386e-05,
1123
+ "loss": 1.8833,
1124
+ "step": 93000
1125
+ },
1126
+ {
1127
+ "epoch": 1.39,
1128
+ "learning_rate": 3.2683132815741124e-05,
1129
+ "loss": 1.7691,
1130
+ "step": 93500
1131
+ },
1132
+ {
1133
+ "epoch": 1.39,
1134
+ "learning_rate": 3.259052924791087e-05,
1135
+ "loss": 1.8607,
1136
+ "step": 94000
1137
+ },
1138
+ {
1139
+ "epoch": 1.4,
1140
+ "learning_rate": 3.2497925680080606e-05,
1141
+ "loss": 1.8724,
1142
+ "step": 94500
1143
+ },
1144
+ {
1145
+ "epoch": 1.41,
1146
+ "learning_rate": 3.2405322112250344e-05,
1147
+ "loss": 1.857,
1148
+ "step": 95000
1149
+ },
1150
+ {
1151
+ "epoch": 1.41,
1152
+ "learning_rate": 3.231271854442008e-05,
1153
+ "loss": 1.8516,
1154
+ "step": 95500
1155
+ },
1156
+ {
1157
+ "epoch": 1.42,
1158
+ "learning_rate": 3.222011497658982e-05,
1159
+ "loss": 1.8776,
1160
+ "step": 96000
1161
+ },
1162
+ {
1163
+ "epoch": 1.43,
1164
+ "learning_rate": 3.212751140875956e-05,
1165
+ "loss": 1.856,
1166
+ "step": 96500
1167
+ },
1168
+ {
1169
+ "epoch": 1.44,
1170
+ "learning_rate": 3.2034907840929295e-05,
1171
+ "loss": 1.8006,
1172
+ "step": 97000
1173
+ },
1174
+ {
1175
+ "epoch": 1.44,
1176
+ "learning_rate": 3.194230427309903e-05,
1177
+ "loss": 1.8157,
1178
+ "step": 97500
1179
+ },
1180
+ {
1181
+ "epoch": 1.45,
1182
+ "learning_rate": 3.184970070526878e-05,
1183
+ "loss": 1.888,
1184
+ "step": 98000
1185
+ },
1186
+ {
1187
+ "epoch": 1.46,
1188
+ "learning_rate": 3.1757097137438515e-05,
1189
+ "loss": 1.7996,
1190
+ "step": 98500
1191
+ },
1192
+ {
1193
+ "epoch": 1.47,
1194
+ "learning_rate": 3.166449356960825e-05,
1195
+ "loss": 1.8884,
1196
+ "step": 99000
1197
+ },
1198
+ {
1199
+ "epoch": 1.47,
1200
+ "learning_rate": 3.1571890001777984e-05,
1201
+ "loss": 1.8581,
1202
+ "step": 99500
1203
+ },
1204
+ {
1205
+ "epoch": 1.48,
1206
+ "learning_rate": 3.147928643394773e-05,
1207
+ "loss": 1.8047,
1208
+ "step": 100000
1209
+ },
1210
+ {
1211
+ "epoch": 1.49,
1212
+ "learning_rate": 3.1386682866117467e-05,
1213
+ "loss": 1.782,
1214
+ "step": 100500
1215
+ },
1216
+ {
1217
+ "epoch": 1.5,
1218
+ "learning_rate": 3.1294079298287204e-05,
1219
+ "loss": 1.8976,
1220
+ "step": 101000
1221
+ },
1222
+ {
1223
+ "epoch": 1.5,
1224
+ "learning_rate": 3.120147573045695e-05,
1225
+ "loss": 1.8875,
1226
+ "step": 101500
1227
+ },
1228
+ {
1229
+ "epoch": 1.51,
1230
+ "learning_rate": 3.110887216262669e-05,
1231
+ "loss": 1.8775,
1232
+ "step": 102000
1233
+ },
1234
+ {
1235
+ "epoch": 1.52,
1236
+ "learning_rate": 3.1016268594796424e-05,
1237
+ "loss": 1.8253,
1238
+ "step": 102500
1239
+ },
1240
+ {
1241
+ "epoch": 1.53,
1242
+ "learning_rate": 3.0923665026966155e-05,
1243
+ "loss": 1.8766,
1244
+ "step": 103000
1245
+ },
1246
+ {
1247
+ "epoch": 1.53,
1248
+ "learning_rate": 3.08310614591359e-05,
1249
+ "loss": 1.7809,
1250
+ "step": 103500
1251
+ },
1252
+ {
1253
+ "epoch": 1.54,
1254
+ "learning_rate": 3.073845789130564e-05,
1255
+ "loss": 1.8724,
1256
+ "step": 104000
1257
+ },
1258
+ {
1259
+ "epoch": 1.55,
1260
+ "learning_rate": 3.0645854323475376e-05,
1261
+ "loss": 1.87,
1262
+ "step": 104500
1263
+ },
1264
+ {
1265
+ "epoch": 1.56,
1266
+ "learning_rate": 3.055325075564511e-05,
1267
+ "loss": 1.888,
1268
+ "step": 105000
1269
+ },
1270
+ {
1271
+ "epoch": 1.56,
1272
+ "learning_rate": 3.0460647187814855e-05,
1273
+ "loss": 1.804,
1274
+ "step": 105500
1275
+ },
1276
+ {
1277
+ "epoch": 1.57,
1278
+ "learning_rate": 3.0368043619984592e-05,
1279
+ "loss": 1.8045,
1280
+ "step": 106000
1281
+ },
1282
+ {
1283
+ "epoch": 1.58,
1284
+ "learning_rate": 3.027544005215433e-05,
1285
+ "loss": 1.8211,
1286
+ "step": 106500
1287
+ },
1288
+ {
1289
+ "epoch": 1.59,
1290
+ "learning_rate": 3.0182836484324068e-05,
1291
+ "loss": 1.8786,
1292
+ "step": 107000
1293
+ },
1294
+ {
1295
+ "epoch": 1.59,
1296
+ "learning_rate": 3.009023291649381e-05,
1297
+ "loss": 1.863,
1298
+ "step": 107500
1299
+ },
1300
+ {
1301
+ "epoch": 1.6,
1302
+ "learning_rate": 2.9997629348663547e-05,
1303
+ "loss": 1.8576,
1304
+ "step": 108000
1305
+ },
1306
+ {
1307
+ "epoch": 1.61,
1308
+ "learning_rate": 2.9905025780833285e-05,
1309
+ "loss": 1.871,
1310
+ "step": 108500
1311
+ },
1312
+ {
1313
+ "epoch": 1.62,
1314
+ "learning_rate": 2.9812422213003022e-05,
1315
+ "loss": 1.8222,
1316
+ "step": 109000
1317
+ },
1318
+ {
1319
+ "epoch": 1.62,
1320
+ "learning_rate": 2.9719818645172764e-05,
1321
+ "loss": 1.8619,
1322
+ "step": 109500
1323
+ },
1324
+ {
1325
+ "epoch": 1.63,
1326
+ "learning_rate": 2.96272150773425e-05,
1327
+ "loss": 1.8899,
1328
+ "step": 110000
1329
+ },
1330
+ {
1331
+ "epoch": 1.64,
1332
+ "learning_rate": 2.953461150951224e-05,
1333
+ "loss": 1.8915,
1334
+ "step": 110500
1335
+ },
1336
+ {
1337
+ "epoch": 1.64,
1338
+ "learning_rate": 2.9442007941681977e-05,
1339
+ "loss": 1.8638,
1340
+ "step": 111000
1341
+ },
1342
+ {
1343
+ "epoch": 1.65,
1344
+ "learning_rate": 2.9349404373851718e-05,
1345
+ "loss": 1.8411,
1346
+ "step": 111500
1347
+ },
1348
+ {
1349
+ "epoch": 1.66,
1350
+ "learning_rate": 2.9256800806021456e-05,
1351
+ "loss": 1.8337,
1352
+ "step": 112000
1353
+ },
1354
+ {
1355
+ "epoch": 1.67,
1356
+ "learning_rate": 2.9164197238191194e-05,
1357
+ "loss": 1.8295,
1358
+ "step": 112500
1359
+ },
1360
+ {
1361
+ "epoch": 1.67,
1362
+ "learning_rate": 2.907159367036093e-05,
1363
+ "loss": 1.8165,
1364
+ "step": 113000
1365
+ },
1366
+ {
1367
+ "epoch": 1.68,
1368
+ "learning_rate": 2.8978990102530673e-05,
1369
+ "loss": 1.7852,
1370
+ "step": 113500
1371
+ },
1372
+ {
1373
+ "epoch": 1.69,
1374
+ "learning_rate": 2.888638653470041e-05,
1375
+ "loss": 1.806,
1376
+ "step": 114000
1377
+ },
1378
+ {
1379
+ "epoch": 1.7,
1380
+ "learning_rate": 2.8793782966870148e-05,
1381
+ "loss": 1.836,
1382
+ "step": 114500
1383
+ },
1384
+ {
1385
+ "epoch": 1.7,
1386
+ "learning_rate": 2.870117939903989e-05,
1387
+ "loss": 1.8174,
1388
+ "step": 115000
1389
+ },
1390
+ {
1391
+ "epoch": 1.71,
1392
+ "learning_rate": 2.8608575831209627e-05,
1393
+ "loss": 1.8736,
1394
+ "step": 115500
1395
+ },
1396
+ {
1397
+ "epoch": 1.72,
1398
+ "learning_rate": 2.8515972263379365e-05,
1399
+ "loss": 1.8421,
1400
+ "step": 116000
1401
+ },
1402
+ {
1403
+ "epoch": 1.73,
1404
+ "learning_rate": 2.8423368695549103e-05,
1405
+ "loss": 1.8666,
1406
+ "step": 116500
1407
+ },
1408
+ {
1409
+ "epoch": 1.73,
1410
+ "learning_rate": 2.8330765127718844e-05,
1411
+ "loss": 1.8455,
1412
+ "step": 117000
1413
+ },
1414
+ {
1415
+ "epoch": 1.74,
1416
+ "learning_rate": 2.8238161559888582e-05,
1417
+ "loss": 1.8277,
1418
+ "step": 117500
1419
+ },
1420
+ {
1421
+ "epoch": 1.75,
1422
+ "learning_rate": 2.814555799205832e-05,
1423
+ "loss": 1.7909,
1424
+ "step": 118000
1425
+ },
1426
+ {
1427
+ "epoch": 1.76,
1428
+ "learning_rate": 2.8052954424228057e-05,
1429
+ "loss": 1.7783,
1430
+ "step": 118500
1431
+ },
1432
+ {
1433
+ "epoch": 1.76,
1434
+ "learning_rate": 2.79603508563978e-05,
1435
+ "loss": 1.8208,
1436
+ "step": 119000
1437
+ },
1438
+ {
1439
+ "epoch": 1.77,
1440
+ "learning_rate": 2.7867747288567536e-05,
1441
+ "loss": 1.8484,
1442
+ "step": 119500
1443
+ },
1444
+ {
1445
+ "epoch": 1.78,
1446
+ "learning_rate": 2.7775143720737274e-05,
1447
+ "loss": 1.824,
1448
+ "step": 120000
1449
+ },
1450
+ {
1451
+ "epoch": 1.79,
1452
+ "learning_rate": 2.7682540152907012e-05,
1453
+ "loss": 1.8373,
1454
+ "step": 120500
1455
+ },
1456
+ {
1457
+ "epoch": 1.79,
1458
+ "learning_rate": 2.7589936585076753e-05,
1459
+ "loss": 1.8155,
1460
+ "step": 121000
1461
+ },
1462
+ {
1463
+ "epoch": 1.8,
1464
+ "learning_rate": 2.749733301724649e-05,
1465
+ "loss": 1.7846,
1466
+ "step": 121500
1467
+ },
1468
+ {
1469
+ "epoch": 1.81,
1470
+ "learning_rate": 2.740472944941623e-05,
1471
+ "loss": 1.8153,
1472
+ "step": 122000
1473
+ },
1474
+ {
1475
+ "epoch": 1.82,
1476
+ "learning_rate": 2.7312125881585966e-05,
1477
+ "loss": 1.8557,
1478
+ "step": 122500
1479
+ },
1480
+ {
1481
+ "epoch": 1.82,
1482
+ "learning_rate": 2.7219522313755708e-05,
1483
+ "loss": 1.8372,
1484
+ "step": 123000
1485
+ },
1486
+ {
1487
+ "epoch": 1.83,
1488
+ "learning_rate": 2.7126918745925445e-05,
1489
+ "loss": 1.8148,
1490
+ "step": 123500
1491
+ },
1492
+ {
1493
+ "epoch": 1.84,
1494
+ "learning_rate": 2.7034315178095183e-05,
1495
+ "loss": 1.7911,
1496
+ "step": 124000
1497
+ },
1498
+ {
1499
+ "epoch": 1.84,
1500
+ "learning_rate": 2.694171161026492e-05,
1501
+ "loss": 1.8467,
1502
+ "step": 124500
1503
+ },
1504
+ {
1505
+ "epoch": 1.85,
1506
+ "learning_rate": 2.6849108042434662e-05,
1507
+ "loss": 1.8121,
1508
+ "step": 125000
1509
+ },
1510
+ {
1511
+ "epoch": 1.86,
1512
+ "learning_rate": 2.67565044746044e-05,
1513
+ "loss": 1.8034,
1514
+ "step": 125500
1515
+ },
1516
+ {
1517
+ "epoch": 1.87,
1518
+ "learning_rate": 2.6663900906774138e-05,
1519
+ "loss": 1.7933,
1520
+ "step": 126000
1521
+ },
1522
+ {
1523
+ "epoch": 1.87,
1524
+ "learning_rate": 2.6571297338943872e-05,
1525
+ "loss": 1.8018,
1526
+ "step": 126500
1527
+ },
1528
+ {
1529
+ "epoch": 1.88,
1530
+ "learning_rate": 2.6478693771113617e-05,
1531
+ "loss": 1.7972,
1532
+ "step": 127000
1533
+ },
1534
+ {
1535
+ "epoch": 1.89,
1536
+ "learning_rate": 2.6386090203283354e-05,
1537
+ "loss": 1.8592,
1538
+ "step": 127500
1539
+ },
1540
+ {
1541
+ "epoch": 1.9,
1542
+ "learning_rate": 2.6293486635453092e-05,
1543
+ "loss": 1.8192,
1544
+ "step": 128000
1545
+ },
1546
+ {
1547
+ "epoch": 1.9,
1548
+ "learning_rate": 2.6200883067622833e-05,
1549
+ "loss": 1.828,
1550
+ "step": 128500
1551
+ },
1552
+ {
1553
+ "epoch": 1.91,
1554
+ "learning_rate": 2.610827949979257e-05,
1555
+ "loss": 1.8501,
1556
+ "step": 129000
1557
+ },
1558
+ {
1559
+ "epoch": 1.92,
1560
+ "learning_rate": 2.601567593196231e-05,
1561
+ "loss": 1.8408,
1562
+ "step": 129500
1563
+ },
1564
+ {
1565
+ "epoch": 1.93,
1566
+ "learning_rate": 2.5923072364132047e-05,
1567
+ "loss": 1.8387,
1568
+ "step": 130000
1569
+ },
1570
+ {
1571
+ "epoch": 1.93,
1572
+ "learning_rate": 2.5830468796301788e-05,
1573
+ "loss": 1.8467,
1574
+ "step": 130500
1575
+ },
1576
+ {
1577
+ "epoch": 1.94,
1578
+ "learning_rate": 2.5737865228471526e-05,
1579
+ "loss": 1.8348,
1580
+ "step": 131000
1581
+ },
1582
+ {
1583
+ "epoch": 1.95,
1584
+ "learning_rate": 2.5645261660641263e-05,
1585
+ "loss": 1.796,
1586
+ "step": 131500
1587
+ },
1588
+ {
1589
+ "epoch": 1.96,
1590
+ "learning_rate": 2.5552658092810998e-05,
1591
+ "loss": 1.804,
1592
+ "step": 132000
1593
+ },
1594
+ {
1595
+ "epoch": 1.96,
1596
+ "learning_rate": 2.5460054524980742e-05,
1597
+ "loss": 1.7959,
1598
+ "step": 132500
1599
+ },
1600
+ {
1601
+ "epoch": 1.97,
1602
+ "learning_rate": 2.536745095715048e-05,
1603
+ "loss": 1.8213,
1604
+ "step": 133000
1605
+ },
1606
+ {
1607
+ "epoch": 1.98,
1608
+ "learning_rate": 2.5274847389320218e-05,
1609
+ "loss": 1.7823,
1610
+ "step": 133500
1611
+ },
1612
+ {
1613
+ "epoch": 1.99,
1614
+ "learning_rate": 2.5182243821489952e-05,
1615
+ "loss": 1.7913,
1616
+ "step": 134000
1617
+ },
1618
+ {
1619
+ "epoch": 1.99,
1620
+ "learning_rate": 2.5089640253659697e-05,
1621
+ "loss": 1.7693,
1622
+ "step": 134500
1623
+ },
1624
+ {
1625
+ "epoch": 2.0,
1626
+ "learning_rate": 2.4997036685829435e-05,
1627
+ "loss": 1.797,
1628
+ "step": 135000
1629
+ },
1630
+ {
1631
+ "epoch": 2.01,
1632
+ "learning_rate": 2.4904433117999172e-05,
1633
+ "loss": 1.3749,
1634
+ "step": 135500
1635
+ },
1636
+ {
1637
+ "epoch": 2.02,
1638
+ "learning_rate": 2.481182955016891e-05,
1639
+ "loss": 1.3449,
1640
+ "step": 136000
1641
+ },
1642
+ {
1643
+ "epoch": 2.02,
1644
+ "learning_rate": 2.4719225982338648e-05,
1645
+ "loss": 1.387,
1646
+ "step": 136500
1647
+ },
1648
+ {
1649
+ "epoch": 2.03,
1650
+ "learning_rate": 2.462662241450839e-05,
1651
+ "loss": 1.3564,
1652
+ "step": 137000
1653
+ },
1654
+ {
1655
+ "epoch": 2.04,
1656
+ "learning_rate": 2.4534018846678124e-05,
1657
+ "loss": 1.3306,
1658
+ "step": 137500
1659
+ },
1660
+ {
1661
+ "epoch": 2.04,
1662
+ "learning_rate": 2.4441415278847865e-05,
1663
+ "loss": 1.3972,
1664
+ "step": 138000
1665
+ },
1666
+ {
1667
+ "epoch": 2.05,
1668
+ "learning_rate": 2.4348811711017603e-05,
1669
+ "loss": 1.3669,
1670
+ "step": 138500
1671
+ },
1672
+ {
1673
+ "epoch": 2.06,
1674
+ "learning_rate": 2.4256208143187344e-05,
1675
+ "loss": 1.4143,
1676
+ "step": 139000
1677
+ },
1678
+ {
1679
+ "epoch": 2.07,
1680
+ "learning_rate": 2.4163604575357078e-05,
1681
+ "loss": 1.3784,
1682
+ "step": 139500
1683
+ },
1684
+ {
1685
+ "epoch": 2.07,
1686
+ "learning_rate": 2.407100100752682e-05,
1687
+ "loss": 1.3759,
1688
+ "step": 140000
1689
+ },
1690
+ {
1691
+ "epoch": 2.08,
1692
+ "learning_rate": 2.3978397439696557e-05,
1693
+ "loss": 1.3384,
1694
+ "step": 140500
1695
+ },
1696
+ {
1697
+ "epoch": 2.09,
1698
+ "learning_rate": 2.3885793871866298e-05,
1699
+ "loss": 1.3865,
1700
+ "step": 141000
1701
+ },
1702
+ {
1703
+ "epoch": 2.1,
1704
+ "learning_rate": 2.3793190304036033e-05,
1705
+ "loss": 1.3423,
1706
+ "step": 141500
1707
+ },
1708
+ {
1709
+ "epoch": 2.1,
1710
+ "learning_rate": 2.3700586736205774e-05,
1711
+ "loss": 1.3576,
1712
+ "step": 142000
1713
+ },
1714
+ {
1715
+ "epoch": 2.11,
1716
+ "learning_rate": 2.360798316837551e-05,
1717
+ "loss": 1.3832,
1718
+ "step": 142500
1719
+ },
1720
+ {
1721
+ "epoch": 2.12,
1722
+ "learning_rate": 2.351537960054525e-05,
1723
+ "loss": 1.3412,
1724
+ "step": 143000
1725
+ },
1726
+ {
1727
+ "epoch": 2.13,
1728
+ "learning_rate": 2.342277603271499e-05,
1729
+ "loss": 1.4008,
1730
+ "step": 143500
1731
+ },
1732
+ {
1733
+ "epoch": 2.13,
1734
+ "learning_rate": 2.333017246488473e-05,
1735
+ "loss": 1.4198,
1736
+ "step": 144000
1737
+ },
1738
+ {
1739
+ "epoch": 2.14,
1740
+ "learning_rate": 2.323756889705447e-05,
1741
+ "loss": 1.3535,
1742
+ "step": 144500
1743
+ },
1744
+ {
1745
+ "epoch": 2.15,
1746
+ "learning_rate": 2.3144965329224204e-05,
1747
+ "loss": 1.4097,
1748
+ "step": 145000
1749
+ },
1750
+ {
1751
+ "epoch": 2.16,
1752
+ "learning_rate": 2.3052361761393945e-05,
1753
+ "loss": 1.409,
1754
+ "step": 145500
1755
+ },
1756
+ {
1757
+ "epoch": 2.16,
1758
+ "learning_rate": 2.2959758193563683e-05,
1759
+ "loss": 1.4151,
1760
+ "step": 146000
1761
+ },
1762
+ {
1763
+ "epoch": 2.17,
1764
+ "learning_rate": 2.2867154625733424e-05,
1765
+ "loss": 1.4083,
1766
+ "step": 146500
1767
+ },
1768
+ {
1769
+ "epoch": 2.18,
1770
+ "learning_rate": 2.277455105790316e-05,
1771
+ "loss": 1.4173,
1772
+ "step": 147000
1773
+ },
1774
+ {
1775
+ "epoch": 2.19,
1776
+ "learning_rate": 2.26819474900729e-05,
1777
+ "loss": 1.463,
1778
+ "step": 147500
1779
+ },
1780
+ {
1781
+ "epoch": 2.19,
1782
+ "learning_rate": 2.2589343922242637e-05,
1783
+ "loss": 1.4193,
1784
+ "step": 148000
1785
+ },
1786
+ {
1787
+ "epoch": 2.2,
1788
+ "learning_rate": 2.2496740354412375e-05,
1789
+ "loss": 1.4219,
1790
+ "step": 148500
1791
+ },
1792
+ {
1793
+ "epoch": 2.21,
1794
+ "learning_rate": 2.2404136786582113e-05,
1795
+ "loss": 1.3641,
1796
+ "step": 149000
1797
+ },
1798
+ {
1799
+ "epoch": 2.22,
1800
+ "learning_rate": 2.2311533218751854e-05,
1801
+ "loss": 1.3905,
1802
+ "step": 149500
1803
+ },
1804
+ {
1805
+ "epoch": 2.22,
1806
+ "learning_rate": 2.2218929650921592e-05,
1807
+ "loss": 1.4293,
1808
+ "step": 150000
1809
+ },
1810
+ {
1811
+ "epoch": 2.23,
1812
+ "learning_rate": 2.212632608309133e-05,
1813
+ "loss": 1.4548,
1814
+ "step": 150500
1815
+ },
1816
+ {
1817
+ "epoch": 2.24,
1818
+ "learning_rate": 2.2033722515261068e-05,
1819
+ "loss": 1.4201,
1820
+ "step": 151000
1821
+ },
1822
+ {
1823
+ "epoch": 2.24,
1824
+ "learning_rate": 2.194111894743081e-05,
1825
+ "loss": 1.3813,
1826
+ "step": 151500
1827
+ },
1828
+ {
1829
+ "epoch": 2.25,
1830
+ "learning_rate": 2.1848515379600546e-05,
1831
+ "loss": 1.3631,
1832
+ "step": 152000
1833
+ },
1834
+ {
1835
+ "epoch": 2.26,
1836
+ "learning_rate": 2.1755911811770284e-05,
1837
+ "loss": 1.4613,
1838
+ "step": 152500
1839
+ },
1840
+ {
1841
+ "epoch": 2.27,
1842
+ "learning_rate": 2.1663308243940022e-05,
1843
+ "loss": 1.37,
1844
+ "step": 153000
1845
+ },
1846
+ {
1847
+ "epoch": 2.27,
1848
+ "learning_rate": 2.1570704676109763e-05,
1849
+ "loss": 1.4528,
1850
+ "step": 153500
1851
+ },
1852
+ {
1853
+ "epoch": 2.28,
1854
+ "learning_rate": 2.14781011082795e-05,
1855
+ "loss": 1.3796,
1856
+ "step": 154000
1857
+ },
1858
+ {
1859
+ "epoch": 2.29,
1860
+ "learning_rate": 2.138549754044924e-05,
1861
+ "loss": 1.4019,
1862
+ "step": 154500
1863
+ },
1864
+ {
1865
+ "epoch": 2.3,
1866
+ "learning_rate": 2.1292893972618977e-05,
1867
+ "loss": 1.3995,
1868
+ "step": 155000
1869
+ },
1870
+ {
1871
+ "epoch": 2.3,
1872
+ "learning_rate": 2.1200290404788718e-05,
1873
+ "loss": 1.3776,
1874
+ "step": 155500
1875
+ },
1876
+ {
1877
+ "epoch": 2.31,
1878
+ "learning_rate": 2.1107686836958456e-05,
1879
+ "loss": 1.4263,
1880
+ "step": 156000
1881
+ },
1882
+ {
1883
+ "epoch": 2.32,
1884
+ "learning_rate": 2.1015083269128193e-05,
1885
+ "loss": 1.395,
1886
+ "step": 156500
1887
+ },
1888
+ {
1889
+ "epoch": 2.33,
1890
+ "learning_rate": 2.092247970129793e-05,
1891
+ "loss": 1.3474,
1892
+ "step": 157000
1893
+ },
1894
+ {
1895
+ "epoch": 2.33,
1896
+ "learning_rate": 2.0829876133467672e-05,
1897
+ "loss": 1.3742,
1898
+ "step": 157500
1899
+ },
1900
+ {
1901
+ "epoch": 2.34,
1902
+ "learning_rate": 2.073727256563741e-05,
1903
+ "loss": 1.4322,
1904
+ "step": 158000
1905
+ },
1906
+ {
1907
+ "epoch": 2.35,
1908
+ "learning_rate": 2.0644668997807148e-05,
1909
+ "loss": 1.3475,
1910
+ "step": 158500
1911
+ },
1912
+ {
1913
+ "epoch": 2.36,
1914
+ "learning_rate": 2.055206542997689e-05,
1915
+ "loss": 1.3618,
1916
+ "step": 159000
1917
+ },
1918
+ {
1919
+ "epoch": 2.36,
1920
+ "learning_rate": 2.0459461862146627e-05,
1921
+ "loss": 1.3992,
1922
+ "step": 159500
1923
+ },
1924
+ {
1925
+ "epoch": 2.37,
1926
+ "learning_rate": 2.0366858294316365e-05,
1927
+ "loss": 1.4042,
1928
+ "step": 160000
1929
+ },
1930
+ {
1931
+ "epoch": 2.38,
1932
+ "learning_rate": 2.0274254726486102e-05,
1933
+ "loss": 1.4052,
1934
+ "step": 160500
1935
+ },
1936
+ {
1937
+ "epoch": 2.39,
1938
+ "learning_rate": 2.0181651158655844e-05,
1939
+ "loss": 1.4609,
1940
+ "step": 161000
1941
+ },
1942
+ {
1943
+ "epoch": 2.39,
1944
+ "learning_rate": 2.008904759082558e-05,
1945
+ "loss": 1.4281,
1946
+ "step": 161500
1947
+ },
1948
+ {
1949
+ "epoch": 2.4,
1950
+ "learning_rate": 1.999644402299532e-05,
1951
+ "loss": 1.4165,
1952
+ "step": 162000
1953
+ },
1954
+ {
1955
+ "epoch": 2.41,
1956
+ "learning_rate": 1.9903840455165057e-05,
1957
+ "loss": 1.383,
1958
+ "step": 162500
1959
+ },
1960
+ {
1961
+ "epoch": 2.42,
1962
+ "learning_rate": 1.9811236887334798e-05,
1963
+ "loss": 1.3725,
1964
+ "step": 163000
1965
+ },
1966
+ {
1967
+ "epoch": 2.42,
1968
+ "learning_rate": 1.9718633319504536e-05,
1969
+ "loss": 1.3656,
1970
+ "step": 163500
1971
+ },
1972
+ {
1973
+ "epoch": 2.43,
1974
+ "learning_rate": 1.9626029751674274e-05,
1975
+ "loss": 1.4094,
1976
+ "step": 164000
1977
+ },
1978
+ {
1979
+ "epoch": 2.44,
1980
+ "learning_rate": 1.953342618384401e-05,
1981
+ "loss": 1.4369,
1982
+ "step": 164500
1983
+ },
1984
+ {
1985
+ "epoch": 2.44,
1986
+ "learning_rate": 1.9440822616013753e-05,
1987
+ "loss": 1.4073,
1988
+ "step": 165000
1989
+ },
1990
+ {
1991
+ "epoch": 2.45,
1992
+ "learning_rate": 1.9348219048183487e-05,
1993
+ "loss": 1.3785,
1994
+ "step": 165500
1995
+ },
1996
+ {
1997
+ "epoch": 2.46,
1998
+ "learning_rate": 1.9255615480353228e-05,
1999
+ "loss": 1.4189,
2000
+ "step": 166000
2001
+ },
2002
+ {
2003
+ "epoch": 2.47,
2004
+ "learning_rate": 1.9163011912522966e-05,
2005
+ "loss": 1.3908,
2006
+ "step": 166500
2007
+ },
2008
+ {
2009
+ "epoch": 2.47,
2010
+ "learning_rate": 1.9070408344692707e-05,
2011
+ "loss": 1.414,
2012
+ "step": 167000
2013
+ },
2014
+ {
2015
+ "epoch": 2.48,
2016
+ "learning_rate": 1.897780477686244e-05,
2017
+ "loss": 1.3761,
2018
+ "step": 167500
2019
+ },
2020
+ {
2021
+ "epoch": 2.49,
2022
+ "learning_rate": 1.8885201209032183e-05,
2023
+ "loss": 1.4145,
2024
+ "step": 168000
2025
+ },
2026
+ {
2027
+ "epoch": 2.5,
2028
+ "learning_rate": 1.879259764120192e-05,
2029
+ "loss": 1.425,
2030
+ "step": 168500
2031
+ },
2032
+ {
2033
+ "epoch": 2.5,
2034
+ "learning_rate": 1.869999407337166e-05,
2035
+ "loss": 1.4359,
2036
+ "step": 169000
2037
+ },
2038
+ {
2039
+ "epoch": 2.51,
2040
+ "learning_rate": 1.8607390505541396e-05,
2041
+ "loss": 1.3791,
2042
+ "step": 169500
2043
+ },
2044
+ {
2045
+ "epoch": 2.52,
2046
+ "learning_rate": 1.8514786937711137e-05,
2047
+ "loss": 1.3784,
2048
+ "step": 170000
2049
+ },
2050
+ {
2051
+ "epoch": 2.53,
2052
+ "learning_rate": 1.8422183369880875e-05,
2053
+ "loss": 1.4094,
2054
+ "step": 170500
2055
+ },
2056
+ {
2057
+ "epoch": 2.53,
2058
+ "learning_rate": 1.8329579802050613e-05,
2059
+ "loss": 1.4048,
2060
+ "step": 171000
2061
+ },
2062
+ {
2063
+ "epoch": 2.54,
2064
+ "learning_rate": 1.8236976234220354e-05,
2065
+ "loss": 1.4183,
2066
+ "step": 171500
2067
+ },
2068
+ {
2069
+ "epoch": 2.55,
2070
+ "learning_rate": 1.8144372666390092e-05,
2071
+ "loss": 1.3787,
2072
+ "step": 172000
2073
+ },
2074
+ {
2075
+ "epoch": 2.56,
2076
+ "learning_rate": 1.8051769098559833e-05,
2077
+ "loss": 1.4246,
2078
+ "step": 172500
2079
+ },
2080
+ {
2081
+ "epoch": 2.56,
2082
+ "learning_rate": 1.7959165530729567e-05,
2083
+ "loss": 1.3803,
2084
+ "step": 173000
2085
+ },
2086
+ {
2087
+ "epoch": 2.57,
2088
+ "learning_rate": 1.786656196289931e-05,
2089
+ "loss": 1.4329,
2090
+ "step": 173500
2091
+ },
2092
+ {
2093
+ "epoch": 2.58,
2094
+ "learning_rate": 1.7773958395069046e-05,
2095
+ "loss": 1.3802,
2096
+ "step": 174000
2097
+ },
2098
+ {
2099
+ "epoch": 2.59,
2100
+ "learning_rate": 1.7681354827238787e-05,
2101
+ "loss": 1.3925,
2102
+ "step": 174500
2103
+ },
2104
+ {
2105
+ "epoch": 2.59,
2106
+ "learning_rate": 1.7588751259408522e-05,
2107
+ "loss": 1.3817,
2108
+ "step": 175000
2109
+ },
2110
+ {
2111
+ "epoch": 2.6,
2112
+ "learning_rate": 1.7496147691578263e-05,
2113
+ "loss": 1.3918,
2114
+ "step": 175500
2115
+ },
2116
+ {
2117
+ "epoch": 2.61,
2118
+ "learning_rate": 1.7403544123748e-05,
2119
+ "loss": 1.4201,
2120
+ "step": 176000
2121
+ },
2122
+ {
2123
+ "epoch": 2.62,
2124
+ "learning_rate": 1.731094055591774e-05,
2125
+ "loss": 1.4103,
2126
+ "step": 176500
2127
+ },
2128
+ {
2129
+ "epoch": 2.62,
2130
+ "learning_rate": 1.7218336988087476e-05,
2131
+ "loss": 1.4312,
2132
+ "step": 177000
2133
+ },
2134
+ {
2135
+ "epoch": 2.63,
2136
+ "learning_rate": 1.7125733420257218e-05,
2137
+ "loss": 1.3746,
2138
+ "step": 177500
2139
+ },
2140
+ {
2141
+ "epoch": 2.64,
2142
+ "learning_rate": 1.7033129852426955e-05,
2143
+ "loss": 1.4273,
2144
+ "step": 178000
2145
+ },
2146
+ {
2147
+ "epoch": 2.64,
2148
+ "learning_rate": 1.6940526284596693e-05,
2149
+ "loss": 1.3886,
2150
+ "step": 178500
2151
+ },
2152
+ {
2153
+ "epoch": 2.65,
2154
+ "learning_rate": 1.684792271676643e-05,
2155
+ "loss": 1.3741,
2156
+ "step": 179000
2157
+ },
2158
+ {
2159
+ "epoch": 2.66,
2160
+ "learning_rate": 1.6755319148936172e-05,
2161
+ "loss": 1.3609,
2162
+ "step": 179500
2163
+ },
2164
+ {
2165
+ "epoch": 2.67,
2166
+ "learning_rate": 1.666271558110591e-05,
2167
+ "loss": 1.3623,
2168
+ "step": 180000
2169
+ },
2170
+ {
2171
+ "epoch": 2.67,
2172
+ "learning_rate": 1.6570112013275648e-05,
2173
+ "loss": 1.3889,
2174
+ "step": 180500
2175
+ },
2176
+ {
2177
+ "epoch": 2.68,
2178
+ "learning_rate": 1.6477508445445385e-05,
2179
+ "loss": 1.448,
2180
+ "step": 181000
2181
+ },
2182
+ {
2183
+ "epoch": 2.69,
2184
+ "learning_rate": 1.6384904877615127e-05,
2185
+ "loss": 1.3549,
2186
+ "step": 181500
2187
+ },
2188
+ {
2189
+ "epoch": 2.7,
2190
+ "learning_rate": 1.6292301309784864e-05,
2191
+ "loss": 1.4019,
2192
+ "step": 182000
2193
+ },
2194
+ {
2195
+ "epoch": 2.7,
2196
+ "learning_rate": 1.6199697741954602e-05,
2197
+ "loss": 1.3859,
2198
+ "step": 182500
2199
+ },
2200
+ {
2201
+ "epoch": 2.71,
2202
+ "learning_rate": 1.610709417412434e-05,
2203
+ "loss": 1.3908,
2204
+ "step": 183000
2205
+ },
2206
+ {
2207
+ "epoch": 2.72,
2208
+ "learning_rate": 1.601449060629408e-05,
2209
+ "loss": 1.3679,
2210
+ "step": 183500
2211
+ },
2212
+ {
2213
+ "epoch": 2.73,
2214
+ "learning_rate": 1.592188703846382e-05,
2215
+ "loss": 1.359,
2216
+ "step": 184000
2217
+ },
2218
+ {
2219
+ "epoch": 2.73,
2220
+ "learning_rate": 1.5829283470633557e-05,
2221
+ "loss": 1.3884,
2222
+ "step": 184500
2223
+ },
2224
+ {
2225
+ "epoch": 2.74,
2226
+ "learning_rate": 1.5736679902803294e-05,
2227
+ "loss": 1.3703,
2228
+ "step": 185000
2229
+ },
2230
+ {
2231
+ "epoch": 2.75,
2232
+ "learning_rate": 1.5644076334973036e-05,
2233
+ "loss": 1.3434,
2234
+ "step": 185500
2235
+ },
2236
+ {
2237
+ "epoch": 2.76,
2238
+ "learning_rate": 1.5551472767142773e-05,
2239
+ "loss": 1.3662,
2240
+ "step": 186000
2241
+ },
2242
+ {
2243
+ "epoch": 2.76,
2244
+ "learning_rate": 1.545886919931251e-05,
2245
+ "loss": 1.4051,
2246
+ "step": 186500
2247
+ },
2248
+ {
2249
+ "epoch": 2.77,
2250
+ "learning_rate": 1.5366265631482252e-05,
2251
+ "loss": 1.4064,
2252
+ "step": 187000
2253
+ },
2254
+ {
2255
+ "epoch": 2.78,
2256
+ "learning_rate": 1.527366206365199e-05,
2257
+ "loss": 1.3615,
2258
+ "step": 187500
2259
+ },
2260
+ {
2261
+ "epoch": 2.79,
2262
+ "learning_rate": 1.518105849582173e-05,
2263
+ "loss": 1.3496,
2264
+ "step": 188000
2265
+ },
2266
+ {
2267
+ "epoch": 2.79,
2268
+ "learning_rate": 1.5088454927991466e-05,
2269
+ "loss": 1.4558,
2270
+ "step": 188500
2271
+ },
2272
+ {
2273
+ "epoch": 2.8,
2274
+ "learning_rate": 1.4995851360161205e-05,
2275
+ "loss": 1.3331,
2276
+ "step": 189000
2277
+ },
2278
+ {
2279
+ "epoch": 2.81,
2280
+ "learning_rate": 1.4903247792330943e-05,
2281
+ "loss": 1.3776,
2282
+ "step": 189500
2283
+ },
2284
+ {
2285
+ "epoch": 2.82,
2286
+ "learning_rate": 1.4810644224500682e-05,
2287
+ "loss": 1.3912,
2288
+ "step": 190000
2289
+ },
2290
+ {
2291
+ "epoch": 2.82,
2292
+ "learning_rate": 1.471804065667042e-05,
2293
+ "loss": 1.3816,
2294
+ "step": 190500
2295
+ },
2296
+ {
2297
+ "epoch": 2.83,
2298
+ "learning_rate": 1.462543708884016e-05,
2299
+ "loss": 1.3897,
2300
+ "step": 191000
2301
+ },
2302
+ {
2303
+ "epoch": 2.84,
2304
+ "learning_rate": 1.4532833521009898e-05,
2305
+ "loss": 1.3831,
2306
+ "step": 191500
2307
+ },
2308
+ {
2309
+ "epoch": 2.84,
2310
+ "learning_rate": 1.4440229953179637e-05,
2311
+ "loss": 1.3954,
2312
+ "step": 192000
2313
+ },
2314
+ {
2315
+ "epoch": 2.85,
2316
+ "learning_rate": 1.4347626385349375e-05,
2317
+ "loss": 1.383,
2318
+ "step": 192500
2319
+ },
2320
+ {
2321
+ "epoch": 2.86,
2322
+ "learning_rate": 1.4255022817519114e-05,
2323
+ "loss": 1.3408,
2324
+ "step": 193000
2325
+ },
2326
+ {
2327
+ "epoch": 2.87,
2328
+ "learning_rate": 1.4162419249688852e-05,
2329
+ "loss": 1.3996,
2330
+ "step": 193500
2331
+ },
2332
+ {
2333
+ "epoch": 2.87,
2334
+ "learning_rate": 1.4069815681858592e-05,
2335
+ "loss": 1.3653,
2336
+ "step": 194000
2337
+ },
2338
+ {
2339
+ "epoch": 2.88,
2340
+ "learning_rate": 1.397721211402833e-05,
2341
+ "loss": 1.4181,
2342
+ "step": 194500
2343
+ },
2344
+ {
2345
+ "epoch": 2.89,
2346
+ "learning_rate": 1.3884608546198069e-05,
2347
+ "loss": 1.3542,
2348
+ "step": 195000
2349
+ },
2350
+ {
2351
+ "epoch": 2.9,
2352
+ "learning_rate": 1.3792004978367807e-05,
2353
+ "loss": 1.3785,
2354
+ "step": 195500
2355
+ },
2356
+ {
2357
+ "epoch": 2.9,
2358
+ "learning_rate": 1.3699401410537546e-05,
2359
+ "loss": 1.4211,
2360
+ "step": 196000
2361
+ },
2362
+ {
2363
+ "epoch": 2.91,
2364
+ "learning_rate": 1.3606797842707284e-05,
2365
+ "loss": 1.3833,
2366
+ "step": 196500
2367
+ },
2368
+ {
2369
+ "epoch": 2.92,
2370
+ "learning_rate": 1.3514194274877023e-05,
2371
+ "loss": 1.3677,
2372
+ "step": 197000
2373
+ },
2374
+ {
2375
+ "epoch": 2.93,
2376
+ "learning_rate": 1.3421590707046761e-05,
2377
+ "loss": 1.4014,
2378
+ "step": 197500
2379
+ },
2380
+ {
2381
+ "epoch": 2.93,
2382
+ "learning_rate": 1.33289871392165e-05,
2383
+ "loss": 1.408,
2384
+ "step": 198000
2385
+ },
2386
+ {
2387
+ "epoch": 2.94,
2388
+ "learning_rate": 1.3236383571386238e-05,
2389
+ "loss": 1.3761,
2390
+ "step": 198500
2391
+ },
2392
+ {
2393
+ "epoch": 2.95,
2394
+ "learning_rate": 1.3143780003555978e-05,
2395
+ "loss": 1.412,
2396
+ "step": 199000
2397
+ },
2398
+ {
2399
+ "epoch": 2.96,
2400
+ "learning_rate": 1.3051176435725716e-05,
2401
+ "loss": 1.3706,
2402
+ "step": 199500
2403
+ },
2404
+ {
2405
+ "epoch": 2.96,
2406
+ "learning_rate": 1.2958572867895455e-05,
2407
+ "loss": 1.3831,
2408
+ "step": 200000
2409
+ },
2410
+ {
2411
+ "epoch": 2.97,
2412
+ "learning_rate": 1.2865969300065195e-05,
2413
+ "loss": 1.3469,
2414
+ "step": 200500
2415
+ },
2416
+ {
2417
+ "epoch": 2.98,
2418
+ "learning_rate": 1.2773365732234932e-05,
2419
+ "loss": 1.3403,
2420
+ "step": 201000
2421
+ },
2422
+ {
2423
+ "epoch": 2.99,
2424
+ "learning_rate": 1.2680762164404672e-05,
2425
+ "loss": 1.3783,
2426
+ "step": 201500
2427
+ },
2428
+ {
2429
+ "epoch": 2.99,
2430
+ "learning_rate": 1.258815859657441e-05,
2431
+ "loss": 1.3897,
2432
+ "step": 202000
2433
+ },
2434
+ {
2435
+ "epoch": 3.0,
2436
+ "learning_rate": 1.2495555028744147e-05,
2437
+ "loss": 1.3974,
2438
+ "step": 202500
2439
+ },
2440
+ {
2441
+ "epoch": 3.01,
2442
+ "learning_rate": 1.2402951460913887e-05,
2443
+ "loss": 0.9791,
2444
+ "step": 203000
2445
+ },
2446
+ {
2447
+ "epoch": 3.02,
2448
+ "learning_rate": 1.2310347893083625e-05,
2449
+ "loss": 0.9923,
2450
+ "step": 203500
2451
+ },
2452
+ {
2453
+ "epoch": 3.02,
2454
+ "learning_rate": 1.2217744325253364e-05,
2455
+ "loss": 0.9658,
2456
+ "step": 204000
2457
+ },
2458
+ {
2459
+ "epoch": 3.03,
2460
+ "learning_rate": 1.2125140757423102e-05,
2461
+ "loss": 1.0327,
2462
+ "step": 204500
2463
+ },
2464
+ {
2465
+ "epoch": 3.04,
2466
+ "learning_rate": 1.2032537189592841e-05,
2467
+ "loss": 0.9956,
2468
+ "step": 205000
2469
+ },
2470
+ {
2471
+ "epoch": 3.04,
2472
+ "learning_rate": 1.193993362176258e-05,
2473
+ "loss": 0.9618,
2474
+ "step": 205500
2475
+ },
2476
+ {
2477
+ "epoch": 3.05,
2478
+ "learning_rate": 1.1847330053932319e-05,
2479
+ "loss": 0.9582,
2480
+ "step": 206000
2481
+ },
2482
+ {
2483
+ "epoch": 3.06,
2484
+ "learning_rate": 1.1754726486102056e-05,
2485
+ "loss": 1.002,
2486
+ "step": 206500
2487
+ },
2488
+ {
2489
+ "epoch": 3.07,
2490
+ "learning_rate": 1.1662122918271796e-05,
2491
+ "loss": 0.959,
2492
+ "step": 207000
2493
+ },
2494
+ {
2495
+ "epoch": 3.07,
2496
+ "learning_rate": 1.1569519350441535e-05,
2497
+ "loss": 0.9495,
2498
+ "step": 207500
2499
+ },
2500
+ {
2501
+ "epoch": 3.08,
2502
+ "learning_rate": 1.1476915782611273e-05,
2503
+ "loss": 1.016,
2504
+ "step": 208000
2505
+ },
2506
+ {
2507
+ "epoch": 3.09,
2508
+ "learning_rate": 1.1384312214781013e-05,
2509
+ "loss": 1.0179,
2510
+ "step": 208500
2511
+ },
2512
+ {
2513
+ "epoch": 3.1,
2514
+ "learning_rate": 1.129170864695075e-05,
2515
+ "loss": 0.9873,
2516
+ "step": 209000
2517
+ },
2518
+ {
2519
+ "epoch": 3.1,
2520
+ "learning_rate": 1.119910507912049e-05,
2521
+ "loss": 1.0094,
2522
+ "step": 209500
2523
+ },
2524
+ {
2525
+ "epoch": 3.11,
2526
+ "learning_rate": 1.1106501511290228e-05,
2527
+ "loss": 0.9773,
2528
+ "step": 210000
2529
+ },
2530
+ {
2531
+ "epoch": 3.12,
2532
+ "learning_rate": 1.1013897943459967e-05,
2533
+ "loss": 1.004,
2534
+ "step": 210500
2535
+ },
2536
+ {
2537
+ "epoch": 3.13,
2538
+ "learning_rate": 1.0921294375629705e-05,
2539
+ "loss": 0.9712,
2540
+ "step": 211000
2541
+ },
2542
+ {
2543
+ "epoch": 3.13,
2544
+ "learning_rate": 1.0828690807799444e-05,
2545
+ "loss": 0.9503,
2546
+ "step": 211500
2547
+ },
2548
+ {
2549
+ "epoch": 3.14,
2550
+ "learning_rate": 1.0736087239969182e-05,
2551
+ "loss": 1.0044,
2552
+ "step": 212000
2553
+ },
2554
+ {
2555
+ "epoch": 3.15,
2556
+ "learning_rate": 1.064348367213892e-05,
2557
+ "loss": 0.9697,
2558
+ "step": 212500
2559
+ },
2560
+ {
2561
+ "epoch": 3.16,
2562
+ "learning_rate": 1.055088010430866e-05,
2563
+ "loss": 0.9893,
2564
+ "step": 213000
2565
+ },
2566
+ {
2567
+ "epoch": 3.16,
2568
+ "learning_rate": 1.0458276536478397e-05,
2569
+ "loss": 1.028,
2570
+ "step": 213500
2571
+ },
2572
+ {
2573
+ "epoch": 3.17,
2574
+ "learning_rate": 1.0365672968648137e-05,
2575
+ "loss": 0.972,
2576
+ "step": 214000
2577
+ },
2578
+ {
2579
+ "epoch": 3.18,
2580
+ "learning_rate": 1.0273069400817875e-05,
2581
+ "loss": 0.975,
2582
+ "step": 214500
2583
+ },
2584
+ {
2585
+ "epoch": 3.19,
2586
+ "learning_rate": 1.0180465832987614e-05,
2587
+ "loss": 0.9553,
2588
+ "step": 215000
2589
+ },
2590
+ {
2591
+ "epoch": 3.19,
2592
+ "learning_rate": 1.0087862265157352e-05,
2593
+ "loss": 0.9977,
2594
+ "step": 215500
2595
+ },
2596
+ {
2597
+ "epoch": 3.2,
2598
+ "learning_rate": 9.995258697327091e-06,
2599
+ "loss": 1.0038,
2600
+ "step": 216000
2601
+ },
2602
+ {
2603
+ "epoch": 3.21,
2604
+ "learning_rate": 9.902655129496829e-06,
2605
+ "loss": 0.9956,
2606
+ "step": 216500
2607
+ },
2608
+ {
2609
+ "epoch": 3.22,
2610
+ "learning_rate": 9.810051561666569e-06,
2611
+ "loss": 0.975,
2612
+ "step": 217000
2613
+ },
2614
+ {
2615
+ "epoch": 3.22,
2616
+ "learning_rate": 9.717447993836306e-06,
2617
+ "loss": 0.9638,
2618
+ "step": 217500
2619
+ },
2620
+ {
2621
+ "epoch": 3.23,
2622
+ "learning_rate": 9.624844426006046e-06,
2623
+ "loss": 0.9551,
2624
+ "step": 218000
2625
+ },
2626
+ {
2627
+ "epoch": 3.24,
2628
+ "learning_rate": 9.532240858175784e-06,
2629
+ "loss": 0.9791,
2630
+ "step": 218500
2631
+ },
2632
+ {
2633
+ "epoch": 3.24,
2634
+ "learning_rate": 9.439637290345523e-06,
2635
+ "loss": 1.015,
2636
+ "step": 219000
2637
+ },
2638
+ {
2639
+ "epoch": 3.25,
2640
+ "learning_rate": 9.347033722515261e-06,
2641
+ "loss": 0.9465,
2642
+ "step": 219500
2643
+ },
2644
+ {
2645
+ "epoch": 3.26,
2646
+ "learning_rate": 9.254430154685e-06,
2647
+ "loss": 0.9856,
2648
+ "step": 220000
2649
+ },
2650
+ {
2651
+ "epoch": 3.27,
2652
+ "learning_rate": 9.161826586854738e-06,
2653
+ "loss": 0.9704,
2654
+ "step": 220500
2655
+ },
2656
+ {
2657
+ "epoch": 3.27,
2658
+ "learning_rate": 9.069223019024476e-06,
2659
+ "loss": 0.9467,
2660
+ "step": 221000
2661
+ },
2662
+ {
2663
+ "epoch": 3.28,
2664
+ "learning_rate": 8.976619451194217e-06,
2665
+ "loss": 0.9747,
2666
+ "step": 221500
2667
+ },
2668
+ {
2669
+ "epoch": 3.29,
2670
+ "learning_rate": 8.884015883363955e-06,
2671
+ "loss": 0.9979,
2672
+ "step": 222000
2673
+ },
2674
+ {
2675
+ "epoch": 3.3,
2676
+ "learning_rate": 8.791412315533694e-06,
2677
+ "loss": 0.9839,
2678
+ "step": 222500
2679
+ },
2680
+ {
2681
+ "epoch": 3.3,
2682
+ "learning_rate": 8.698808747703432e-06,
2683
+ "loss": 0.9905,
2684
+ "step": 223000
2685
+ },
2686
+ {
2687
+ "epoch": 3.31,
2688
+ "learning_rate": 8.606205179873172e-06,
2689
+ "loss": 1.0233,
2690
+ "step": 223500
2691
+ },
2692
+ {
2693
+ "epoch": 3.32,
2694
+ "learning_rate": 8.51360161204291e-06,
2695
+ "loss": 1.0214,
2696
+ "step": 224000
2697
+ },
2698
+ {
2699
+ "epoch": 3.33,
2700
+ "learning_rate": 8.420998044212649e-06,
2701
+ "loss": 0.9928,
2702
+ "step": 224500
2703
+ },
2704
+ {
2705
+ "epoch": 3.33,
2706
+ "learning_rate": 8.328394476382387e-06,
2707
+ "loss": 0.98,
2708
+ "step": 225000
2709
+ },
2710
+ {
2711
+ "epoch": 3.34,
2712
+ "learning_rate": 8.235790908552126e-06,
2713
+ "loss": 0.9924,
2714
+ "step": 225500
2715
+ },
2716
+ {
2717
+ "epoch": 3.35,
2718
+ "learning_rate": 8.143187340721864e-06,
2719
+ "loss": 1.0215,
2720
+ "step": 226000
2721
+ },
2722
+ {
2723
+ "epoch": 3.36,
2724
+ "learning_rate": 8.050583772891602e-06,
2725
+ "loss": 0.957,
2726
+ "step": 226500
2727
+ },
2728
+ {
2729
+ "epoch": 3.36,
2730
+ "learning_rate": 7.957980205061341e-06,
2731
+ "loss": 0.988,
2732
+ "step": 227000
2733
+ },
2734
+ {
2735
+ "epoch": 3.37,
2736
+ "learning_rate": 7.865376637231079e-06,
2737
+ "loss": 0.9771,
2738
+ "step": 227500
2739
+ },
2740
+ {
2741
+ "epoch": 3.38,
2742
+ "learning_rate": 7.772773069400818e-06,
2743
+ "loss": 0.9663,
2744
+ "step": 228000
2745
+ },
2746
+ {
2747
+ "epoch": 3.39,
2748
+ "learning_rate": 7.680169501570556e-06,
2749
+ "loss": 0.9971,
2750
+ "step": 228500
2751
+ },
2752
+ {
2753
+ "epoch": 3.39,
2754
+ "learning_rate": 7.587565933740296e-06,
2755
+ "loss": 0.9708,
2756
+ "step": 229000
2757
+ },
2758
+ {
2759
+ "epoch": 3.4,
2760
+ "learning_rate": 7.494962365910034e-06,
2761
+ "loss": 0.9741,
2762
+ "step": 229500
2763
+ },
2764
+ {
2765
+ "epoch": 3.41,
2766
+ "learning_rate": 7.402358798079773e-06,
2767
+ "loss": 0.9427,
2768
+ "step": 230000
2769
+ },
2770
+ {
2771
+ "epoch": 3.42,
2772
+ "learning_rate": 7.309755230249511e-06,
2773
+ "loss": 1.0002,
2774
+ "step": 230500
2775
+ },
2776
+ {
2777
+ "epoch": 3.42,
2778
+ "learning_rate": 7.2171516624192494e-06,
2779
+ "loss": 0.9685,
2780
+ "step": 231000
2781
+ },
2782
+ {
2783
+ "epoch": 3.43,
2784
+ "learning_rate": 7.124548094588988e-06,
2785
+ "loss": 0.9676,
2786
+ "step": 231500
2787
+ },
2788
+ {
2789
+ "epoch": 3.44,
2790
+ "learning_rate": 7.031944526758727e-06,
2791
+ "loss": 0.9527,
2792
+ "step": 232000
2793
+ },
2794
+ {
2795
+ "epoch": 3.44,
2796
+ "learning_rate": 6.939340958928465e-06,
2797
+ "loss": 0.9675,
2798
+ "step": 232500
2799
+ },
2800
+ {
2801
+ "epoch": 3.45,
2802
+ "learning_rate": 6.846737391098204e-06,
2803
+ "loss": 0.9852,
2804
+ "step": 233000
2805
+ },
2806
+ {
2807
+ "epoch": 3.46,
2808
+ "learning_rate": 6.754133823267943e-06,
2809
+ "loss": 0.9607,
2810
+ "step": 233500
2811
+ },
2812
+ {
2813
+ "epoch": 3.47,
2814
+ "learning_rate": 6.661530255437681e-06,
2815
+ "loss": 0.9526,
2816
+ "step": 234000
2817
+ },
2818
+ {
2819
+ "epoch": 3.47,
2820
+ "learning_rate": 6.56892668760742e-06,
2821
+ "loss": 0.9592,
2822
+ "step": 234500
2823
+ },
2824
+ {
2825
+ "epoch": 3.48,
2826
+ "learning_rate": 6.4763231197771585e-06,
2827
+ "loss": 0.9227,
2828
+ "step": 235000
2829
+ },
2830
+ {
2831
+ "epoch": 3.49,
2832
+ "learning_rate": 6.383719551946897e-06,
2833
+ "loss": 0.9842,
2834
+ "step": 235500
2835
+ },
2836
+ {
2837
+ "epoch": 3.5,
2838
+ "learning_rate": 6.291115984116637e-06,
2839
+ "loss": 0.9649,
2840
+ "step": 236000
2841
+ },
2842
+ {
2843
+ "epoch": 3.5,
2844
+ "learning_rate": 6.198512416286375e-06,
2845
+ "loss": 0.9231,
2846
+ "step": 236500
2847
+ },
2848
+ {
2849
+ "epoch": 3.51,
2850
+ "learning_rate": 6.105908848456114e-06,
2851
+ "loss": 0.9331,
2852
+ "step": 237000
2853
+ },
2854
+ {
2855
+ "epoch": 3.52,
2856
+ "learning_rate": 6.013305280625852e-06,
2857
+ "loss": 0.9748,
2858
+ "step": 237500
2859
+ },
2860
+ {
2861
+ "epoch": 3.53,
2862
+ "learning_rate": 5.92070171279559e-06,
2863
+ "loss": 0.9471,
2864
+ "step": 238000
2865
+ },
2866
+ {
2867
+ "epoch": 3.53,
2868
+ "learning_rate": 5.828098144965329e-06,
2869
+ "loss": 0.9819,
2870
+ "step": 238500
2871
+ },
2872
+ {
2873
+ "epoch": 3.54,
2874
+ "learning_rate": 5.7354945771350676e-06,
2875
+ "loss": 0.9898,
2876
+ "step": 239000
2877
+ },
2878
+ {
2879
+ "epoch": 3.55,
2880
+ "learning_rate": 5.642891009304807e-06,
2881
+ "loss": 0.9692,
2882
+ "step": 239500
2883
+ },
2884
+ {
2885
+ "epoch": 3.56,
2886
+ "learning_rate": 5.550287441474546e-06,
2887
+ "loss": 0.9939,
2888
+ "step": 240000
2889
+ },
2890
+ {
2891
+ "epoch": 3.56,
2892
+ "learning_rate": 5.457683873644284e-06,
2893
+ "loss": 0.9903,
2894
+ "step": 240500
2895
+ },
2896
+ {
2897
+ "epoch": 3.57,
2898
+ "learning_rate": 5.365080305814023e-06,
2899
+ "loss": 0.9671,
2900
+ "step": 241000
2901
+ },
2902
+ {
2903
+ "epoch": 3.58,
2904
+ "learning_rate": 5.2724767379837616e-06,
2905
+ "loss": 0.9527,
2906
+ "step": 241500
2907
+ },
2908
+ {
2909
+ "epoch": 3.59,
2910
+ "learning_rate": 5.1798731701535e-06,
2911
+ "loss": 0.9644,
2912
+ "step": 242000
2913
+ },
2914
+ {
2915
+ "epoch": 3.59,
2916
+ "learning_rate": 5.087269602323239e-06,
2917
+ "loss": 0.9809,
2918
+ "step": 242500
2919
+ },
2920
+ {
2921
+ "epoch": 3.6,
2922
+ "learning_rate": 4.9946660344929775e-06,
2923
+ "loss": 0.9401,
2924
+ "step": 243000
2925
+ },
2926
+ {
2927
+ "epoch": 3.61,
2928
+ "learning_rate": 4.902062466662716e-06,
2929
+ "loss": 0.9661,
2930
+ "step": 243500
2931
+ },
2932
+ {
2933
+ "epoch": 3.62,
2934
+ "learning_rate": 4.809458898832455e-06,
2935
+ "loss": 0.9696,
2936
+ "step": 244000
2937
+ },
2938
+ {
2939
+ "epoch": 3.62,
2940
+ "learning_rate": 4.7168553310021925e-06,
2941
+ "loss": 0.959,
2942
+ "step": 244500
2943
+ },
2944
+ {
2945
+ "epoch": 3.63,
2946
+ "learning_rate": 4.624251763171931e-06,
2947
+ "loss": 0.9661,
2948
+ "step": 245000
2949
+ },
2950
+ {
2951
+ "epoch": 3.64,
2952
+ "learning_rate": 4.53164819534167e-06,
2953
+ "loss": 0.9654,
2954
+ "step": 245500
2955
+ },
2956
+ {
2957
+ "epoch": 3.64,
2958
+ "learning_rate": 4.439044627511408e-06,
2959
+ "loss": 0.9804,
2960
+ "step": 246000
2961
+ },
2962
+ {
2963
+ "epoch": 3.65,
2964
+ "learning_rate": 4.346441059681148e-06,
2965
+ "loss": 0.9792,
2966
+ "step": 246500
2967
+ },
2968
+ {
2969
+ "epoch": 3.66,
2970
+ "learning_rate": 4.2538374918508865e-06,
2971
+ "loss": 0.8952,
2972
+ "step": 247000
2973
+ },
2974
+ {
2975
+ "epoch": 3.67,
2976
+ "learning_rate": 4.161233924020625e-06,
2977
+ "loss": 0.9645,
2978
+ "step": 247500
2979
+ },
2980
+ {
2981
+ "epoch": 3.67,
2982
+ "learning_rate": 4.068630356190364e-06,
2983
+ "loss": 0.9889,
2984
+ "step": 248000
2985
+ },
2986
+ {
2987
+ "epoch": 3.68,
2988
+ "learning_rate": 3.976026788360102e-06,
2989
+ "loss": 0.9172,
2990
+ "step": 248500
2991
+ },
2992
+ {
2993
+ "epoch": 3.69,
2994
+ "learning_rate": 3.883423220529841e-06,
2995
+ "loss": 0.9503,
2996
+ "step": 249000
2997
+ },
2998
+ {
2999
+ "epoch": 3.7,
3000
+ "learning_rate": 3.7908196526995792e-06,
3001
+ "loss": 0.9454,
3002
+ "step": 249500
3003
+ },
3004
+ {
3005
+ "epoch": 3.7,
3006
+ "learning_rate": 3.698216084869318e-06,
3007
+ "loss": 0.9557,
3008
+ "step": 250000
3009
+ },
3010
+ {
3011
+ "epoch": 3.71,
3012
+ "learning_rate": 3.6056125170390565e-06,
3013
+ "loss": 0.9833,
3014
+ "step": 250500
3015
+ },
3016
+ {
3017
+ "epoch": 3.72,
3018
+ "learning_rate": 3.513008949208795e-06,
3019
+ "loss": 0.9378,
3020
+ "step": 251000
3021
+ },
3022
+ {
3023
+ "epoch": 3.73,
3024
+ "learning_rate": 3.4204053813785338e-06,
3025
+ "loss": 0.9718,
3026
+ "step": 251500
3027
+ },
3028
+ {
3029
+ "epoch": 3.73,
3030
+ "learning_rate": 3.3278018135482724e-06,
3031
+ "loss": 0.9639,
3032
+ "step": 252000
3033
+ },
3034
+ {
3035
+ "epoch": 3.74,
3036
+ "learning_rate": 3.235198245718011e-06,
3037
+ "loss": 0.9254,
3038
+ "step": 252500
3039
+ },
3040
+ {
3041
+ "epoch": 3.75,
3042
+ "learning_rate": 3.1425946778877497e-06,
3043
+ "loss": 1.0007,
3044
+ "step": 253000
3045
+ },
3046
+ {
3047
+ "epoch": 3.76,
3048
+ "learning_rate": 3.0499911100574883e-06,
3049
+ "loss": 0.968,
3050
+ "step": 253500
3051
+ },
3052
+ {
3053
+ "epoch": 3.76,
3054
+ "learning_rate": 2.957387542227227e-06,
3055
+ "loss": 0.9663,
3056
+ "step": 254000
3057
+ },
3058
+ {
3059
+ "epoch": 3.77,
3060
+ "learning_rate": 2.8647839743969656e-06,
3061
+ "loss": 0.9838,
3062
+ "step": 254500
3063
+ },
3064
+ {
3065
+ "epoch": 3.78,
3066
+ "learning_rate": 2.772180406566704e-06,
3067
+ "loss": 0.9547,
3068
+ "step": 255000
3069
+ },
3070
+ {
3071
+ "epoch": 3.79,
3072
+ "learning_rate": 2.6795768387364432e-06,
3073
+ "loss": 0.9677,
3074
+ "step": 255500
3075
+ },
3076
+ {
3077
+ "epoch": 3.79,
3078
+ "learning_rate": 2.586973270906182e-06,
3079
+ "loss": 0.9753,
3080
+ "step": 256000
3081
+ },
3082
+ {
3083
+ "epoch": 3.8,
3084
+ "learning_rate": 2.49436970307592e-06,
3085
+ "loss": 0.9834,
3086
+ "step": 256500
3087
+ },
3088
+ {
3089
+ "epoch": 3.81,
3090
+ "learning_rate": 2.4017661352456587e-06,
3091
+ "loss": 0.9809,
3092
+ "step": 257000
3093
+ },
3094
+ {
3095
+ "epoch": 3.82,
3096
+ "learning_rate": 2.3091625674153974e-06,
3097
+ "loss": 0.9584,
3098
+ "step": 257500
3099
+ },
3100
+ {
3101
+ "epoch": 3.82,
3102
+ "learning_rate": 2.216558999585136e-06,
3103
+ "loss": 0.9321,
3104
+ "step": 258000
3105
+ },
3106
+ {
3107
+ "epoch": 3.83,
3108
+ "learning_rate": 2.1239554317548746e-06,
3109
+ "loss": 0.944,
3110
+ "step": 258500
3111
+ },
3112
+ {
3113
+ "epoch": 3.84,
3114
+ "learning_rate": 2.0313518639246137e-06,
3115
+ "loss": 0.9592,
3116
+ "step": 259000
3117
+ },
3118
+ {
3119
+ "epoch": 3.84,
3120
+ "learning_rate": 1.9387482960943523e-06,
3121
+ "loss": 0.96,
3122
+ "step": 259500
3123
+ },
3124
+ {
3125
+ "epoch": 3.85,
3126
+ "learning_rate": 1.8461447282640907e-06,
3127
+ "loss": 0.9804,
3128
+ "step": 260000
3129
+ },
3130
+ {
3131
+ "epoch": 3.86,
3132
+ "learning_rate": 1.7535411604338294e-06,
3133
+ "loss": 0.9617,
3134
+ "step": 260500
3135
+ },
3136
+ {
3137
+ "epoch": 3.87,
3138
+ "learning_rate": 1.6609375926035678e-06,
3139
+ "loss": 0.9234,
3140
+ "step": 261000
3141
+ },
3142
+ {
3143
+ "epoch": 3.87,
3144
+ "learning_rate": 1.5683340247733064e-06,
3145
+ "loss": 0.965,
3146
+ "step": 261500
3147
+ },
3148
+ {
3149
+ "epoch": 3.88,
3150
+ "learning_rate": 1.4757304569430452e-06,
3151
+ "loss": 0.9286,
3152
+ "step": 262000
3153
+ },
3154
+ {
3155
+ "epoch": 3.89,
3156
+ "learning_rate": 1.3831268891127839e-06,
3157
+ "loss": 0.9899,
3158
+ "step": 262500
3159
+ },
3160
+ {
3161
+ "epoch": 3.9,
3162
+ "learning_rate": 1.2905233212825223e-06,
3163
+ "loss": 0.9527,
3164
+ "step": 263000
3165
+ },
3166
+ {
3167
+ "epoch": 3.9,
3168
+ "learning_rate": 1.1979197534522611e-06,
3169
+ "loss": 0.9387,
3170
+ "step": 263500
3171
+ },
3172
+ {
3173
+ "epoch": 3.91,
3174
+ "learning_rate": 1.1053161856219998e-06,
3175
+ "loss": 0.9385,
3176
+ "step": 264000
3177
+ },
3178
+ {
3179
+ "epoch": 3.92,
3180
+ "learning_rate": 1.0127126177917382e-06,
3181
+ "loss": 0.976,
3182
+ "step": 264500
3183
+ },
3184
+ {
3185
+ "epoch": 3.93,
3186
+ "learning_rate": 9.20109049961477e-07,
3187
+ "loss": 0.9702,
3188
+ "step": 265000
3189
+ },
3190
+ {
3191
+ "epoch": 3.93,
3192
+ "learning_rate": 8.275054821312157e-07,
3193
+ "loss": 0.9343,
3194
+ "step": 265500
3195
+ },
3196
+ {
3197
+ "epoch": 3.94,
3198
+ "learning_rate": 7.349019143009542e-07,
3199
+ "loss": 0.9258,
3200
+ "step": 266000
3201
+ },
3202
+ {
3203
+ "epoch": 3.95,
3204
+ "learning_rate": 6.422983464706928e-07,
3205
+ "loss": 0.9513,
3206
+ "step": 266500
3207
+ },
3208
+ {
3209
+ "epoch": 3.96,
3210
+ "learning_rate": 5.496947786404315e-07,
3211
+ "loss": 0.9689,
3212
+ "step": 267000
3213
+ },
3214
+ {
3215
+ "epoch": 3.96,
3216
+ "learning_rate": 4.5709121081017015e-07,
3217
+ "loss": 0.9296,
3218
+ "step": 267500
3219
+ },
3220
+ {
3221
+ "epoch": 3.97,
3222
+ "learning_rate": 3.644876429799088e-07,
3223
+ "loss": 0.9628,
3224
+ "step": 268000
3225
+ },
3226
+ {
3227
+ "epoch": 3.98,
3228
+ "learning_rate": 2.7188407514964736e-07,
3229
+ "loss": 0.9392,
3230
+ "step": 268500
3231
+ },
3232
+ {
3233
+ "epoch": 3.99,
3234
+ "learning_rate": 1.7928050731938602e-07,
3235
+ "loss": 0.9467,
3236
+ "step": 269000
3237
+ },
3238
+ {
3239
+ "epoch": 3.99,
3240
+ "learning_rate": 8.667693948912465e-08,
3241
+ "loss": 0.9457,
3242
+ "step": 269500
3243
+ },
3244
+ {
3245
+ "epoch": 4.0,
3246
+ "step": 269968,
3247
+ "total_flos": 2.914340112407593e+17,
3248
+ "train_loss": 1.6431024962071243,
3249
+ "train_runtime": 97099.2276,
3250
+ "train_samples_per_second": 2.78,
3251
+ "train_steps_per_second": 2.78
3252
+ }
3253
+ ],
3254
+ "max_steps": 269968,
3255
+ "num_train_epochs": 4,
3256
+ "total_flos": 2.914340112407593e+17,
3257
+ "trial_name": null,
3258
+ "trial_params": null
3259
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cfd1e52f5142ee95a393cd26199c906828a61bcd34af9644567e255b5a65343
3
+ size 3119