Hugging Face's logo

ARTeLab
/
mbart-summarization-mlsum

Summarization
Transformers
PyTorch
Safetensors
Italian
mbart
text2text-generation
Inference Endpoints
Model card Files Community
2
artelabsuper commited on
Commit
6424b22
·
1 Parent(s): 1ac048e

trained model

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 +1555 -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_mlsum
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_mlsum
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: 3.3336
22
+ - Rouge1: 19.3489
23
+ - Rouge2: 6.4028
24
+ - Rougel: 16.3497
25
+ - Rougelsum: 16.5387
26
+ - Gen Len: 33.5945
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": 33.5945,
4
+ "eval_loss": 3.3335702419281006,
5
+ "eval_rouge1": 19.3489,
6
+ "eval_rouge2": 6.4028,
7
+ "eval_rougeL": 16.3497,
8
+ "eval_rougeLsum": 16.5387,
9
+ "eval_runtime": 2520.0092,
10
+ "eval_samples": 4000,
11
+ "eval_samples_per_second": 1.587,
12
+ "eval_steps_per_second": 1.587,
13
+ "predict_gen_len": 33.5382,
14
+ "predict_loss": 3.2747132778167725,
15
+ "predict_rouge1": 19.8687,
16
+ "predict_rouge2": 6.6948,
17
+ "predict_rougeL": 16.8896,
18
+ "predict_rougeLsum": 17.0323,
19
+ "predict_runtime": 2521.7344,
20
+ "predict_samples": 4000,
21
+ "predict_samples_per_second": 1.586,
22
+ "predict_steps_per_second": 1.586,
23
+ "train_loss": 2.0082999333540426,
24
+ "train_runtime": 37501.9411,
25
+ "train_samples": 31997,
26
+ "train_samples_per_second": 3.413,
27
+ "train_steps_per_second": 3.413
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": 33.5945,
4
+ "eval_loss": 3.3335702419281006,
5
+ "eval_rouge1": 19.3489,
6
+ "eval_rouge2": 6.4028,
7
+ "eval_rougeL": 16.3497,
8
+ "eval_rougeLsum": 16.5387,
9
+ "eval_runtime": 2520.0092,
10
+ "eval_samples": 4000,
11
+ "eval_samples_per_second": 1.587,
12
+ "eval_steps_per_second": 1.587
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": 33.5382,
3
+ "predict_loss": 3.2747132778167725,
4
+ "predict_rouge1": 19.8687,
5
+ "predict_rouge2": 6.6948,
6
+ "predict_rougeL": 16.8896,
7
+ "predict_rougeLsum": 17.0323,
8
+ "predict_runtime": 2521.7344,
9
+ "predict_samples": 4000,
10
+ "predict_samples_per_second": 1.586,
11
+ "predict_steps_per_second": 1.586
12
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5620ecf7a04f9a3a233163ac428f89883750a719e79c7877fc39b5b6413f68b1
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": 2.0082999333540426,
4
+ "train_runtime": 37501.9411,
5
+ "train_samples": 31997,
6
+ "train_samples_per_second": 3.413,
7
+ "train_steps_per_second": 3.413
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1555 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.0,
5
+ "global_step": 127988,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.02,
12
+ "learning_rate": 4.980466918773635e-05,
13
+ "loss": 3.8391,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.03,
18
+ "learning_rate": 4.96093383754727e-05,
19
+ "loss": 3.5549,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.05,
24
+ "learning_rate": 4.941400756320905e-05,
25
+ "loss": 3.5792,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.06,
30
+ "learning_rate": 4.9218676750945406e-05,
31
+ "loss": 3.4807,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.08,
36
+ "learning_rate": 4.9023345938681755e-05,
37
+ "loss": 5.5711,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.09,
42
+ "learning_rate": 4.8828015126418104e-05,
43
+ "loss": 3.4924,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.11,
48
+ "learning_rate": 4.863268431415445e-05,
49
+ "loss": 3.3686,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.13,
54
+ "learning_rate": 4.84373535018908e-05,
55
+ "loss": 3.3857,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.14,
60
+ "learning_rate": 4.824202268962715e-05,
61
+ "loss": 3.3547,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.16,
66
+ "learning_rate": 4.80466918773635e-05,
67
+ "loss": 3.3625,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.17,
72
+ "learning_rate": 4.7851361065099856e-05,
73
+ "loss": 3.3164,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.19,
78
+ "learning_rate": 4.7656030252836205e-05,
79
+ "loss": 3.301,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.2,
84
+ "learning_rate": 4.7460699440572554e-05,
85
+ "loss": 3.1746,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.22,
90
+ "learning_rate": 4.7265368628308904e-05,
91
+ "loss": 3.2548,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.23,
96
+ "learning_rate": 4.707003781604525e-05,
97
+ "loss": 3.1595,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.25,
102
+ "learning_rate": 4.687470700378161e-05,
103
+ "loss": 3.2356,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.27,
108
+ "learning_rate": 4.667937619151796e-05,
109
+ "loss": 3.2533,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.28,
114
+ "learning_rate": 4.648404537925431e-05,
115
+ "loss": 3.2142,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.3,
120
+ "learning_rate": 4.628871456699066e-05,
121
+ "loss": 3.1645,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 0.31,
126
+ "learning_rate": 4.609338375472701e-05,
127
+ "loss": 3.2479,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 0.33,
132
+ "learning_rate": 4.589805294246336e-05,
133
+ "loss": 3.2195,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 0.34,
138
+ "learning_rate": 4.570272213019971e-05,
139
+ "loss": 3.1258,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 0.36,
144
+ "learning_rate": 4.550739131793606e-05,
145
+ "loss": 3.1771,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 0.38,
150
+ "learning_rate": 4.531206050567241e-05,
151
+ "loss": 3.2118,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 0.39,
156
+ "learning_rate": 4.511672969340876e-05,
157
+ "loss": 3.101,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 0.41,
162
+ "learning_rate": 4.492139888114511e-05,
163
+ "loss": 3.129,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 0.42,
168
+ "learning_rate": 4.472606806888146e-05,
169
+ "loss": 3.1021,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 0.44,
174
+ "learning_rate": 4.453073725661781e-05,
175
+ "loss": 3.1185,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 0.45,
180
+ "learning_rate": 4.433540644435416e-05,
181
+ "loss": 3.1291,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 0.47,
186
+ "learning_rate": 4.414007563209051e-05,
187
+ "loss": 3.0354,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 0.48,
192
+ "learning_rate": 4.394474481982686e-05,
193
+ "loss": 3.0164,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 0.5,
198
+ "learning_rate": 4.374941400756321e-05,
199
+ "loss": 3.0837,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 0.52,
204
+ "learning_rate": 4.3554083195299564e-05,
205
+ "loss": 3.0235,
206
+ "step": 16500
207
+ },
208
+ {
209
+ "epoch": 0.53,
210
+ "learning_rate": 4.335875238303591e-05,
211
+ "loss": 3.0046,
212
+ "step": 17000
213
+ },
214
+ {
215
+ "epoch": 0.55,
216
+ "learning_rate": 4.316342157077226e-05,
217
+ "loss": 3.0702,
218
+ "step": 17500
219
+ },
220
+ {
221
+ "epoch": 0.56,
222
+ "learning_rate": 4.296809075850861e-05,
223
+ "loss": 3.0582,
224
+ "step": 18000
225
+ },
226
+ {
227
+ "epoch": 0.58,
228
+ "learning_rate": 4.277275994624496e-05,
229
+ "loss": 3.039,
230
+ "step": 18500
231
+ },
232
+ {
233
+ "epoch": 0.59,
234
+ "learning_rate": 4.257742913398131e-05,
235
+ "loss": 3.1094,
236
+ "step": 19000
237
+ },
238
+ {
239
+ "epoch": 0.61,
240
+ "learning_rate": 4.238209832171766e-05,
241
+ "loss": 2.9548,
242
+ "step": 19500
243
+ },
244
+ {
245
+ "epoch": 0.63,
246
+ "learning_rate": 4.2186767509454014e-05,
247
+ "loss": 3.0493,
248
+ "step": 20000
249
+ },
250
+ {
251
+ "epoch": 0.64,
252
+ "learning_rate": 4.1991436697190363e-05,
253
+ "loss": 3.0473,
254
+ "step": 20500
255
+ },
256
+ {
257
+ "epoch": 0.66,
258
+ "learning_rate": 4.179610588492671e-05,
259
+ "loss": 3.0363,
260
+ "step": 21000
261
+ },
262
+ {
263
+ "epoch": 0.67,
264
+ "learning_rate": 4.160077507266306e-05,
265
+ "loss": 2.9445,
266
+ "step": 21500
267
+ },
268
+ {
269
+ "epoch": 0.69,
270
+ "learning_rate": 4.140544426039941e-05,
271
+ "loss": 2.9368,
272
+ "step": 22000
273
+ },
274
+ {
275
+ "epoch": 0.7,
276
+ "learning_rate": 4.121011344813576e-05,
277
+ "loss": 2.9829,
278
+ "step": 22500
279
+ },
280
+ {
281
+ "epoch": 0.72,
282
+ "learning_rate": 4.1014782635872116e-05,
283
+ "loss": 3.0427,
284
+ "step": 23000
285
+ },
286
+ {
287
+ "epoch": 0.73,
288
+ "learning_rate": 4.0819451823608465e-05,
289
+ "loss": 2.974,
290
+ "step": 23500
291
+ },
292
+ {
293
+ "epoch": 0.75,
294
+ "learning_rate": 4.062412101134482e-05,
295
+ "loss": 2.9314,
296
+ "step": 24000
297
+ },
298
+ {
299
+ "epoch": 0.77,
300
+ "learning_rate": 4.042879019908117e-05,
301
+ "loss": 2.9788,
302
+ "step": 24500
303
+ },
304
+ {
305
+ "epoch": 0.78,
306
+ "learning_rate": 4.023345938681752e-05,
307
+ "loss": 2.9794,
308
+ "step": 25000
309
+ },
310
+ {
311
+ "epoch": 0.8,
312
+ "learning_rate": 4.003812857455387e-05,
313
+ "loss": 2.9851,
314
+ "step": 25500
315
+ },
316
+ {
317
+ "epoch": 0.81,
318
+ "learning_rate": 3.984279776229022e-05,
319
+ "loss": 2.9453,
320
+ "step": 26000
321
+ },
322
+ {
323
+ "epoch": 0.83,
324
+ "learning_rate": 3.9647466950026566e-05,
325
+ "loss": 3.0433,
326
+ "step": 26500
327
+ },
328
+ {
329
+ "epoch": 0.84,
330
+ "learning_rate": 3.9452136137762916e-05,
331
+ "loss": 2.956,
332
+ "step": 27000
333
+ },
334
+ {
335
+ "epoch": 0.86,
336
+ "learning_rate": 3.925680532549927e-05,
337
+ "loss": 2.9388,
338
+ "step": 27500
339
+ },
340
+ {
341
+ "epoch": 0.88,
342
+ "learning_rate": 3.906147451323562e-05,
343
+ "loss": 2.9308,
344
+ "step": 28000
345
+ },
346
+ {
347
+ "epoch": 0.89,
348
+ "learning_rate": 3.886614370097197e-05,
349
+ "loss": 2.9588,
350
+ "step": 28500
351
+ },
352
+ {
353
+ "epoch": 0.91,
354
+ "learning_rate": 3.867081288870832e-05,
355
+ "loss": 2.9588,
356
+ "step": 29000
357
+ },
358
+ {
359
+ "epoch": 0.92,
360
+ "learning_rate": 3.847548207644467e-05,
361
+ "loss": 2.8812,
362
+ "step": 29500
363
+ },
364
+ {
365
+ "epoch": 0.94,
366
+ "learning_rate": 3.828015126418102e-05,
367
+ "loss": 2.9103,
368
+ "step": 30000
369
+ },
370
+ {
371
+ "epoch": 0.95,
372
+ "learning_rate": 3.8084820451917366e-05,
373
+ "loss": 2.8671,
374
+ "step": 30500
375
+ },
376
+ {
377
+ "epoch": 0.97,
378
+ "learning_rate": 3.788948963965372e-05,
379
+ "loss": 2.868,
380
+ "step": 31000
381
+ },
382
+ {
383
+ "epoch": 0.98,
384
+ "learning_rate": 3.769415882739007e-05,
385
+ "loss": 2.9499,
386
+ "step": 31500
387
+ },
388
+ {
389
+ "epoch": 1.0,
390
+ "learning_rate": 3.749882801512642e-05,
391
+ "loss": 2.9272,
392
+ "step": 32000
393
+ },
394
+ {
395
+ "epoch": 1.02,
396
+ "learning_rate": 3.730349720286277e-05,
397
+ "loss": 2.3079,
398
+ "step": 32500
399
+ },
400
+ {
401
+ "epoch": 1.03,
402
+ "learning_rate": 3.710816639059912e-05,
403
+ "loss": 2.3296,
404
+ "step": 33000
405
+ },
406
+ {
407
+ "epoch": 1.05,
408
+ "learning_rate": 3.691283557833547e-05,
409
+ "loss": 2.3636,
410
+ "step": 33500
411
+ },
412
+ {
413
+ "epoch": 1.06,
414
+ "learning_rate": 3.671750476607182e-05,
415
+ "loss": 2.3879,
416
+ "step": 34000
417
+ },
418
+ {
419
+ "epoch": 1.08,
420
+ "learning_rate": 3.652217395380817e-05,
421
+ "loss": 2.3653,
422
+ "step": 34500
423
+ },
424
+ {
425
+ "epoch": 1.09,
426
+ "learning_rate": 3.632684314154452e-05,
427
+ "loss": 2.2813,
428
+ "step": 35000
429
+ },
430
+ {
431
+ "epoch": 1.11,
432
+ "learning_rate": 3.613151232928087e-05,
433
+ "loss": 2.3776,
434
+ "step": 35500
435
+ },
436
+ {
437
+ "epoch": 1.13,
438
+ "learning_rate": 3.593618151701722e-05,
439
+ "loss": 2.3478,
440
+ "step": 36000
441
+ },
442
+ {
443
+ "epoch": 1.14,
444
+ "learning_rate": 3.574085070475357e-05,
445
+ "loss": 2.3453,
446
+ "step": 36500
447
+ },
448
+ {
449
+ "epoch": 1.16,
450
+ "learning_rate": 3.554551989248992e-05,
451
+ "loss": 2.3145,
452
+ "step": 37000
453
+ },
454
+ {
455
+ "epoch": 1.17,
456
+ "learning_rate": 3.535018908022627e-05,
457
+ "loss": 2.3296,
458
+ "step": 37500
459
+ },
460
+ {
461
+ "epoch": 1.19,
462
+ "learning_rate": 3.515485826796262e-05,
463
+ "loss": 2.3543,
464
+ "step": 38000
465
+ },
466
+ {
467
+ "epoch": 1.2,
468
+ "learning_rate": 3.495952745569898e-05,
469
+ "loss": 2.289,
470
+ "step": 38500
471
+ },
472
+ {
473
+ "epoch": 1.22,
474
+ "learning_rate": 3.476419664343533e-05,
475
+ "loss": 2.4305,
476
+ "step": 39000
477
+ },
478
+ {
479
+ "epoch": 1.23,
480
+ "learning_rate": 3.456886583117168e-05,
481
+ "loss": 2.3672,
482
+ "step": 39500
483
+ },
484
+ {
485
+ "epoch": 1.25,
486
+ "learning_rate": 3.4373535018908026e-05,
487
+ "loss": 2.3041,
488
+ "step": 40000
489
+ },
490
+ {
491
+ "epoch": 1.27,
492
+ "learning_rate": 3.4178204206644375e-05,
493
+ "loss": 2.3282,
494
+ "step": 40500
495
+ },
496
+ {
497
+ "epoch": 1.28,
498
+ "learning_rate": 3.3982873394380725e-05,
499
+ "loss": 2.3631,
500
+ "step": 41000
501
+ },
502
+ {
503
+ "epoch": 1.3,
504
+ "learning_rate": 3.3787542582117074e-05,
505
+ "loss": 2.2899,
506
+ "step": 41500
507
+ },
508
+ {
509
+ "epoch": 1.31,
510
+ "learning_rate": 3.359221176985343e-05,
511
+ "loss": 2.3022,
512
+ "step": 42000
513
+ },
514
+ {
515
+ "epoch": 1.33,
516
+ "learning_rate": 3.339688095758978e-05,
517
+ "loss": 2.3512,
518
+ "step": 42500
519
+ },
520
+ {
521
+ "epoch": 1.34,
522
+ "learning_rate": 3.320155014532613e-05,
523
+ "loss": 2.3304,
524
+ "step": 43000
525
+ },
526
+ {
527
+ "epoch": 1.36,
528
+ "learning_rate": 3.300621933306248e-05,
529
+ "loss": 2.3307,
530
+ "step": 43500
531
+ },
532
+ {
533
+ "epoch": 1.38,
534
+ "learning_rate": 3.2810888520798826e-05,
535
+ "loss": 2.3752,
536
+ "step": 44000
537
+ },
538
+ {
539
+ "epoch": 1.39,
540
+ "learning_rate": 3.2615557708535175e-05,
541
+ "loss": 2.2553,
542
+ "step": 44500
543
+ },
544
+ {
545
+ "epoch": 1.41,
546
+ "learning_rate": 3.2420226896271524e-05,
547
+ "loss": 2.3906,
548
+ "step": 45000
549
+ },
550
+ {
551
+ "epoch": 1.42,
552
+ "learning_rate": 3.222489608400788e-05,
553
+ "loss": 2.3277,
554
+ "step": 45500
555
+ },
556
+ {
557
+ "epoch": 1.44,
558
+ "learning_rate": 3.202956527174423e-05,
559
+ "loss": 2.3515,
560
+ "step": 46000
561
+ },
562
+ {
563
+ "epoch": 1.45,
564
+ "learning_rate": 3.183423445948058e-05,
565
+ "loss": 2.3409,
566
+ "step": 46500
567
+ },
568
+ {
569
+ "epoch": 1.47,
570
+ "learning_rate": 3.163890364721693e-05,
571
+ "loss": 2.3093,
572
+ "step": 47000
573
+ },
574
+ {
575
+ "epoch": 1.48,
576
+ "learning_rate": 3.1443572834953277e-05,
577
+ "loss": 2.2961,
578
+ "step": 47500
579
+ },
580
+ {
581
+ "epoch": 1.5,
582
+ "learning_rate": 3.1248242022689626e-05,
583
+ "loss": 2.2775,
584
+ "step": 48000
585
+ },
586
+ {
587
+ "epoch": 1.52,
588
+ "learning_rate": 3.1052911210425975e-05,
589
+ "loss": 2.2714,
590
+ "step": 48500
591
+ },
592
+ {
593
+ "epoch": 1.53,
594
+ "learning_rate": 3.085758039816233e-05,
595
+ "loss": 2.3634,
596
+ "step": 49000
597
+ },
598
+ {
599
+ "epoch": 1.55,
600
+ "learning_rate": 3.066224958589868e-05,
601
+ "loss": 2.353,
602
+ "step": 49500
603
+ },
604
+ {
605
+ "epoch": 1.56,
606
+ "learning_rate": 3.046691877363503e-05,
607
+ "loss": 2.3295,
608
+ "step": 50000
609
+ },
610
+ {
611
+ "epoch": 1.58,
612
+ "learning_rate": 3.0271587961371378e-05,
613
+ "loss": 2.3014,
614
+ "step": 50500
615
+ },
616
+ {
617
+ "epoch": 1.59,
618
+ "learning_rate": 3.0076257149107727e-05,
619
+ "loss": 2.319,
620
+ "step": 51000
621
+ },
622
+ {
623
+ "epoch": 1.61,
624
+ "learning_rate": 2.988092633684408e-05,
625
+ "loss": 2.3364,
626
+ "step": 51500
627
+ },
628
+ {
629
+ "epoch": 1.63,
630
+ "learning_rate": 2.968559552458043e-05,
631
+ "loss": 2.3935,
632
+ "step": 52000
633
+ },
634
+ {
635
+ "epoch": 1.64,
636
+ "learning_rate": 2.9490264712316778e-05,
637
+ "loss": 2.379,
638
+ "step": 52500
639
+ },
640
+ {
641
+ "epoch": 1.66,
642
+ "learning_rate": 2.9294933900053127e-05,
643
+ "loss": 2.284,
644
+ "step": 53000
645
+ },
646
+ {
647
+ "epoch": 1.67,
648
+ "learning_rate": 2.9099603087789483e-05,
649
+ "loss": 2.317,
650
+ "step": 53500
651
+ },
652
+ {
653
+ "epoch": 1.69,
654
+ "learning_rate": 2.8904272275525835e-05,
655
+ "loss": 2.3177,
656
+ "step": 54000
657
+ },
658
+ {
659
+ "epoch": 1.7,
660
+ "learning_rate": 2.8708941463262185e-05,
661
+ "loss": 2.3619,
662
+ "step": 54500
663
+ },
664
+ {
665
+ "epoch": 1.72,
666
+ "learning_rate": 2.8513610650998534e-05,
667
+ "loss": 2.3052,
668
+ "step": 55000
669
+ },
670
+ {
671
+ "epoch": 1.73,
672
+ "learning_rate": 2.8318279838734886e-05,
673
+ "loss": 2.3617,
674
+ "step": 55500
675
+ },
676
+ {
677
+ "epoch": 1.75,
678
+ "learning_rate": 2.8122949026471235e-05,
679
+ "loss": 2.3074,
680
+ "step": 56000
681
+ },
682
+ {
683
+ "epoch": 1.77,
684
+ "learning_rate": 2.7927618214207584e-05,
685
+ "loss": 2.3894,
686
+ "step": 56500
687
+ },
688
+ {
689
+ "epoch": 1.78,
690
+ "learning_rate": 2.7732287401943933e-05,
691
+ "loss": 2.3208,
692
+ "step": 57000
693
+ },
694
+ {
695
+ "epoch": 1.8,
696
+ "learning_rate": 2.7536956589680286e-05,
697
+ "loss": 2.3569,
698
+ "step": 57500
699
+ },
700
+ {
701
+ "epoch": 1.81,
702
+ "learning_rate": 2.7341625777416635e-05,
703
+ "loss": 2.4341,
704
+ "step": 58000
705
+ },
706
+ {
707
+ "epoch": 1.83,
708
+ "learning_rate": 2.7146294965152984e-05,
709
+ "loss": 2.246,
710
+ "step": 58500
711
+ },
712
+ {
713
+ "epoch": 1.84,
714
+ "learning_rate": 2.6950964152889337e-05,
715
+ "loss": 2.2938,
716
+ "step": 59000
717
+ },
718
+ {
719
+ "epoch": 1.86,
720
+ "learning_rate": 2.6755633340625686e-05,
721
+ "loss": 2.3378,
722
+ "step": 59500
723
+ },
724
+ {
725
+ "epoch": 1.88,
726
+ "learning_rate": 2.6560302528362035e-05,
727
+ "loss": 2.297,
728
+ "step": 60000
729
+ },
730
+ {
731
+ "epoch": 1.89,
732
+ "learning_rate": 2.6364971716098384e-05,
733
+ "loss": 2.3859,
734
+ "step": 60500
735
+ },
736
+ {
737
+ "epoch": 1.91,
738
+ "learning_rate": 2.6169640903834737e-05,
739
+ "loss": 2.4035,
740
+ "step": 61000
741
+ },
742
+ {
743
+ "epoch": 1.92,
744
+ "learning_rate": 2.5974310091571086e-05,
745
+ "loss": 2.3027,
746
+ "step": 61500
747
+ },
748
+ {
749
+ "epoch": 1.94,
750
+ "learning_rate": 2.5778979279307435e-05,
751
+ "loss": 2.2821,
752
+ "step": 62000
753
+ },
754
+ {
755
+ "epoch": 1.95,
756
+ "learning_rate": 2.5583648467043787e-05,
757
+ "loss": 2.3267,
758
+ "step": 62500
759
+ },
760
+ {
761
+ "epoch": 1.97,
762
+ "learning_rate": 2.5388317654780136e-05,
763
+ "loss": 2.3205,
764
+ "step": 63000
765
+ },
766
+ {
767
+ "epoch": 1.98,
768
+ "learning_rate": 2.5192986842516485e-05,
769
+ "loss": 2.3695,
770
+ "step": 63500
771
+ },
772
+ {
773
+ "epoch": 2.0,
774
+ "learning_rate": 2.4997656030252835e-05,
775
+ "loss": 2.3184,
776
+ "step": 64000
777
+ },
778
+ {
779
+ "epoch": 2.02,
780
+ "learning_rate": 2.4802325217989187e-05,
781
+ "loss": 1.6629,
782
+ "step": 64500
783
+ },
784
+ {
785
+ "epoch": 2.03,
786
+ "learning_rate": 2.460699440572554e-05,
787
+ "loss": 1.6359,
788
+ "step": 65000
789
+ },
790
+ {
791
+ "epoch": 2.05,
792
+ "learning_rate": 2.441166359346189e-05,
793
+ "loss": 1.5733,
794
+ "step": 65500
795
+ },
796
+ {
797
+ "epoch": 2.06,
798
+ "learning_rate": 2.4216332781198238e-05,
799
+ "loss": 1.5338,
800
+ "step": 66000
801
+ },
802
+ {
803
+ "epoch": 2.08,
804
+ "learning_rate": 2.402100196893459e-05,
805
+ "loss": 1.5802,
806
+ "step": 66500
807
+ },
808
+ {
809
+ "epoch": 2.09,
810
+ "learning_rate": 2.382567115667094e-05,
811
+ "loss": 1.5904,
812
+ "step": 67000
813
+ },
814
+ {
815
+ "epoch": 2.11,
816
+ "learning_rate": 2.363034034440729e-05,
817
+ "loss": 1.6188,
818
+ "step": 67500
819
+ },
820
+ {
821
+ "epoch": 2.13,
822
+ "learning_rate": 2.343500953214364e-05,
823
+ "loss": 1.6242,
824
+ "step": 68000
825
+ },
826
+ {
827
+ "epoch": 2.14,
828
+ "learning_rate": 2.323967871987999e-05,
829
+ "loss": 1.6195,
830
+ "step": 68500
831
+ },
832
+ {
833
+ "epoch": 2.16,
834
+ "learning_rate": 2.304434790761634e-05,
835
+ "loss": 1.5721,
836
+ "step": 69000
837
+ },
838
+ {
839
+ "epoch": 2.17,
840
+ "learning_rate": 2.284901709535269e-05,
841
+ "loss": 1.6324,
842
+ "step": 69500
843
+ },
844
+ {
845
+ "epoch": 2.19,
846
+ "learning_rate": 2.265368628308904e-05,
847
+ "loss": 1.5727,
848
+ "step": 70000
849
+ },
850
+ {
851
+ "epoch": 2.2,
852
+ "learning_rate": 2.245835547082539e-05,
853
+ "loss": 1.5782,
854
+ "step": 70500
855
+ },
856
+ {
857
+ "epoch": 2.22,
858
+ "learning_rate": 2.226302465856174e-05,
859
+ "loss": 1.5516,
860
+ "step": 71000
861
+ },
862
+ {
863
+ "epoch": 2.23,
864
+ "learning_rate": 2.206769384629809e-05,
865
+ "loss": 1.6206,
866
+ "step": 71500
867
+ },
868
+ {
869
+ "epoch": 2.25,
870
+ "learning_rate": 2.187236303403444e-05,
871
+ "loss": 1.6315,
872
+ "step": 72000
873
+ },
874
+ {
875
+ "epoch": 2.27,
876
+ "learning_rate": 2.1677032221770793e-05,
877
+ "loss": 1.6529,
878
+ "step": 72500
879
+ },
880
+ {
881
+ "epoch": 2.28,
882
+ "learning_rate": 2.1481701409507142e-05,
883
+ "loss": 1.5855,
884
+ "step": 73000
885
+ },
886
+ {
887
+ "epoch": 2.3,
888
+ "learning_rate": 2.1286370597243495e-05,
889
+ "loss": 1.565,
890
+ "step": 73500
891
+ },
892
+ {
893
+ "epoch": 2.31,
894
+ "learning_rate": 2.1091039784979844e-05,
895
+ "loss": 1.621,
896
+ "step": 74000
897
+ },
898
+ {
899
+ "epoch": 2.33,
900
+ "learning_rate": 2.0895708972716193e-05,
901
+ "loss": 1.6038,
902
+ "step": 74500
903
+ },
904
+ {
905
+ "epoch": 2.34,
906
+ "learning_rate": 2.0700378160452542e-05,
907
+ "loss": 1.6686,
908
+ "step": 75000
909
+ },
910
+ {
911
+ "epoch": 2.36,
912
+ "learning_rate": 2.0505047348188895e-05,
913
+ "loss": 1.6477,
914
+ "step": 75500
915
+ },
916
+ {
917
+ "epoch": 2.38,
918
+ "learning_rate": 2.0309716535925244e-05,
919
+ "loss": 1.6113,
920
+ "step": 76000
921
+ },
922
+ {
923
+ "epoch": 2.39,
924
+ "learning_rate": 2.0114385723661593e-05,
925
+ "loss": 1.6444,
926
+ "step": 76500
927
+ },
928
+ {
929
+ "epoch": 2.41,
930
+ "learning_rate": 1.9919054911397945e-05,
931
+ "loss": 1.5774,
932
+ "step": 77000
933
+ },
934
+ {
935
+ "epoch": 2.42,
936
+ "learning_rate": 1.9723724099134295e-05,
937
+ "loss": 1.6326,
938
+ "step": 77500
939
+ },
940
+ {
941
+ "epoch": 2.44,
942
+ "learning_rate": 1.9528393286870644e-05,
943
+ "loss": 1.689,
944
+ "step": 78000
945
+ },
946
+ {
947
+ "epoch": 2.45,
948
+ "learning_rate": 1.9333062474606993e-05,
949
+ "loss": 1.5713,
950
+ "step": 78500
951
+ },
952
+ {
953
+ "epoch": 2.47,
954
+ "learning_rate": 1.9137731662343345e-05,
955
+ "loss": 1.6681,
956
+ "step": 79000
957
+ },
958
+ {
959
+ "epoch": 2.48,
960
+ "learning_rate": 1.8942400850079694e-05,
961
+ "loss": 1.6577,
962
+ "step": 79500
963
+ },
964
+ {
965
+ "epoch": 2.5,
966
+ "learning_rate": 1.8747070037816047e-05,
967
+ "loss": 1.6331,
968
+ "step": 80000
969
+ },
970
+ {
971
+ "epoch": 2.52,
972
+ "learning_rate": 1.8551739225552396e-05,
973
+ "loss": 1.6913,
974
+ "step": 80500
975
+ },
976
+ {
977
+ "epoch": 2.53,
978
+ "learning_rate": 1.835640841328875e-05,
979
+ "loss": 1.5789,
980
+ "step": 81000
981
+ },
982
+ {
983
+ "epoch": 2.55,
984
+ "learning_rate": 1.8161077601025098e-05,
985
+ "loss": 1.6053,
986
+ "step": 81500
987
+ },
988
+ {
989
+ "epoch": 2.56,
990
+ "learning_rate": 1.7965746788761447e-05,
991
+ "loss": 1.6313,
992
+ "step": 82000
993
+ },
994
+ {
995
+ "epoch": 2.58,
996
+ "learning_rate": 1.77704159764978e-05,
997
+ "loss": 1.6604,
998
+ "step": 82500
999
+ },
1000
+ {
1001
+ "epoch": 2.59,
1002
+ "learning_rate": 1.757508516423415e-05,
1003
+ "loss": 1.6661,
1004
+ "step": 83000
1005
+ },
1006
+ {
1007
+ "epoch": 2.61,
1008
+ "learning_rate": 1.7379754351970497e-05,
1009
+ "loss": 1.7325,
1010
+ "step": 83500
1011
+ },
1012
+ {
1013
+ "epoch": 2.63,
1014
+ "learning_rate": 1.7184423539706847e-05,
1015
+ "loss": 1.6432,
1016
+ "step": 84000
1017
+ },
1018
+ {
1019
+ "epoch": 2.64,
1020
+ "learning_rate": 1.69890927274432e-05,
1021
+ "loss": 1.6206,
1022
+ "step": 84500
1023
+ },
1024
+ {
1025
+ "epoch": 2.66,
1026
+ "learning_rate": 1.6793761915179548e-05,
1027
+ "loss": 1.5495,
1028
+ "step": 85000
1029
+ },
1030
+ {
1031
+ "epoch": 2.67,
1032
+ "learning_rate": 1.6598431102915897e-05,
1033
+ "loss": 1.6021,
1034
+ "step": 85500
1035
+ },
1036
+ {
1037
+ "epoch": 2.69,
1038
+ "learning_rate": 1.640310029065225e-05,
1039
+ "loss": 1.6682,
1040
+ "step": 86000
1041
+ },
1042
+ {
1043
+ "epoch": 2.7,
1044
+ "learning_rate": 1.62077694783886e-05,
1045
+ "loss": 1.6304,
1046
+ "step": 86500
1047
+ },
1048
+ {
1049
+ "epoch": 2.72,
1050
+ "learning_rate": 1.6012438666124948e-05,
1051
+ "loss": 1.6224,
1052
+ "step": 87000
1053
+ },
1054
+ {
1055
+ "epoch": 2.73,
1056
+ "learning_rate": 1.58171078538613e-05,
1057
+ "loss": 1.5918,
1058
+ "step": 87500
1059
+ },
1060
+ {
1061
+ "epoch": 2.75,
1062
+ "learning_rate": 1.5621777041597653e-05,
1063
+ "loss": 1.6452,
1064
+ "step": 88000
1065
+ },
1066
+ {
1067
+ "epoch": 2.77,
1068
+ "learning_rate": 1.5426446229334002e-05,
1069
+ "loss": 1.5782,
1070
+ "step": 88500
1071
+ },
1072
+ {
1073
+ "epoch": 2.78,
1074
+ "learning_rate": 1.5231115417070351e-05,
1075
+ "loss": 1.5876,
1076
+ "step": 89000
1077
+ },
1078
+ {
1079
+ "epoch": 2.8,
1080
+ "learning_rate": 1.5035784604806702e-05,
1081
+ "loss": 1.6174,
1082
+ "step": 89500
1083
+ },
1084
+ {
1085
+ "epoch": 2.81,
1086
+ "learning_rate": 1.4840453792543051e-05,
1087
+ "loss": 1.6148,
1088
+ "step": 90000
1089
+ },
1090
+ {
1091
+ "epoch": 2.83,
1092
+ "learning_rate": 1.4645122980279402e-05,
1093
+ "loss": 1.6081,
1094
+ "step": 90500
1095
+ },
1096
+ {
1097
+ "epoch": 2.84,
1098
+ "learning_rate": 1.4449792168015753e-05,
1099
+ "loss": 1.5744,
1100
+ "step": 91000
1101
+ },
1102
+ {
1103
+ "epoch": 2.86,
1104
+ "learning_rate": 1.4254461355752102e-05,
1105
+ "loss": 1.6439,
1106
+ "step": 91500
1107
+ },
1108
+ {
1109
+ "epoch": 2.88,
1110
+ "learning_rate": 1.4059130543488453e-05,
1111
+ "loss": 1.665,
1112
+ "step": 92000
1113
+ },
1114
+ {
1115
+ "epoch": 2.89,
1116
+ "learning_rate": 1.3863799731224802e-05,
1117
+ "loss": 1.5745,
1118
+ "step": 92500
1119
+ },
1120
+ {
1121
+ "epoch": 2.91,
1122
+ "learning_rate": 1.3668468918961153e-05,
1123
+ "loss": 1.6089,
1124
+ "step": 93000
1125
+ },
1126
+ {
1127
+ "epoch": 2.92,
1128
+ "learning_rate": 1.3473138106697503e-05,
1129
+ "loss": 1.586,
1130
+ "step": 93500
1131
+ },
1132
+ {
1133
+ "epoch": 2.94,
1134
+ "learning_rate": 1.3277807294433853e-05,
1135
+ "loss": 1.6731,
1136
+ "step": 94000
1137
+ },
1138
+ {
1139
+ "epoch": 2.95,
1140
+ "learning_rate": 1.3082476482170203e-05,
1141
+ "loss": 1.5726,
1142
+ "step": 94500
1143
+ },
1144
+ {
1145
+ "epoch": 2.97,
1146
+ "learning_rate": 1.2887145669906556e-05,
1147
+ "loss": 1.5644,
1148
+ "step": 95000
1149
+ },
1150
+ {
1151
+ "epoch": 2.98,
1152
+ "learning_rate": 1.2691814857642905e-05,
1153
+ "loss": 1.5762,
1154
+ "step": 95500
1155
+ },
1156
+ {
1157
+ "epoch": 3.0,
1158
+ "learning_rate": 1.2496484045379254e-05,
1159
+ "loss": 1.5657,
1160
+ "step": 96000
1161
+ },
1162
+ {
1163
+ "epoch": 3.02,
1164
+ "learning_rate": 1.2301153233115605e-05,
1165
+ "loss": 0.9433,
1166
+ "step": 96500
1167
+ },
1168
+ {
1169
+ "epoch": 3.03,
1170
+ "learning_rate": 1.2105822420851956e-05,
1171
+ "loss": 0.946,
1172
+ "step": 97000
1173
+ },
1174
+ {
1175
+ "epoch": 3.05,
1176
+ "learning_rate": 1.1910491608588307e-05,
1177
+ "loss": 0.9428,
1178
+ "step": 97500
1179
+ },
1180
+ {
1181
+ "epoch": 3.06,
1182
+ "learning_rate": 1.1715160796324656e-05,
1183
+ "loss": 0.8946,
1184
+ "step": 98000
1185
+ },
1186
+ {
1187
+ "epoch": 3.08,
1188
+ "learning_rate": 1.1519829984061006e-05,
1189
+ "loss": 0.9151,
1190
+ "step": 98500
1191
+ },
1192
+ {
1193
+ "epoch": 3.09,
1194
+ "learning_rate": 1.1324499171797357e-05,
1195
+ "loss": 0.9333,
1196
+ "step": 99000
1197
+ },
1198
+ {
1199
+ "epoch": 3.11,
1200
+ "learning_rate": 1.1129168359533706e-05,
1201
+ "loss": 0.9164,
1202
+ "step": 99500
1203
+ },
1204
+ {
1205
+ "epoch": 3.13,
1206
+ "learning_rate": 1.0933837547270057e-05,
1207
+ "loss": 0.9953,
1208
+ "step": 100000
1209
+ },
1210
+ {
1211
+ "epoch": 3.14,
1212
+ "learning_rate": 1.0738506735006406e-05,
1213
+ "loss": 0.9017,
1214
+ "step": 100500
1215
+ },
1216
+ {
1217
+ "epoch": 3.16,
1218
+ "learning_rate": 1.0543175922742759e-05,
1219
+ "loss": 0.9446,
1220
+ "step": 101000
1221
+ },
1222
+ {
1223
+ "epoch": 3.17,
1224
+ "learning_rate": 1.0347845110479108e-05,
1225
+ "loss": 0.9605,
1226
+ "step": 101500
1227
+ },
1228
+ {
1229
+ "epoch": 3.19,
1230
+ "learning_rate": 1.0152514298215459e-05,
1231
+ "loss": 0.9269,
1232
+ "step": 102000
1233
+ },
1234
+ {
1235
+ "epoch": 3.2,
1236
+ "learning_rate": 9.957183485951808e-06,
1237
+ "loss": 0.9677,
1238
+ "step": 102500
1239
+ },
1240
+ {
1241
+ "epoch": 3.22,
1242
+ "learning_rate": 9.761852673688159e-06,
1243
+ "loss": 0.9406,
1244
+ "step": 103000
1245
+ },
1246
+ {
1247
+ "epoch": 3.23,
1248
+ "learning_rate": 9.56652186142451e-06,
1249
+ "loss": 0.9382,
1250
+ "step": 103500
1251
+ },
1252
+ {
1253
+ "epoch": 3.25,
1254
+ "learning_rate": 9.371191049160859e-06,
1255
+ "loss": 0.9711,
1256
+ "step": 104000
1257
+ },
1258
+ {
1259
+ "epoch": 3.27,
1260
+ "learning_rate": 9.175860236897211e-06,
1261
+ "loss": 0.9505,
1262
+ "step": 104500
1263
+ },
1264
+ {
1265
+ "epoch": 3.28,
1266
+ "learning_rate": 8.98052942463356e-06,
1267
+ "loss": 0.9686,
1268
+ "step": 105000
1269
+ },
1270
+ {
1271
+ "epoch": 3.3,
1272
+ "learning_rate": 8.785198612369911e-06,
1273
+ "loss": 0.9394,
1274
+ "step": 105500
1275
+ },
1276
+ {
1277
+ "epoch": 3.31,
1278
+ "learning_rate": 8.58986780010626e-06,
1279
+ "loss": 0.9612,
1280
+ "step": 106000
1281
+ },
1282
+ {
1283
+ "epoch": 3.33,
1284
+ "learning_rate": 8.394536987842611e-06,
1285
+ "loss": 0.968,
1286
+ "step": 106500
1287
+ },
1288
+ {
1289
+ "epoch": 3.34,
1290
+ "learning_rate": 8.19920617557896e-06,
1291
+ "loss": 0.9907,
1292
+ "step": 107000
1293
+ },
1294
+ {
1295
+ "epoch": 3.36,
1296
+ "learning_rate": 8.00387536331531e-06,
1297
+ "loss": 0.964,
1298
+ "step": 107500
1299
+ },
1300
+ {
1301
+ "epoch": 3.38,
1302
+ "learning_rate": 7.808544551051662e-06,
1303
+ "loss": 0.9707,
1304
+ "step": 108000
1305
+ },
1306
+ {
1307
+ "epoch": 3.39,
1308
+ "learning_rate": 7.613213738788012e-06,
1309
+ "loss": 0.9111,
1310
+ "step": 108500
1311
+ },
1312
+ {
1313
+ "epoch": 3.41,
1314
+ "learning_rate": 7.417882926524362e-06,
1315
+ "loss": 0.9294,
1316
+ "step": 109000
1317
+ },
1318
+ {
1319
+ "epoch": 3.42,
1320
+ "learning_rate": 7.222552114260712e-06,
1321
+ "loss": 0.9288,
1322
+ "step": 109500
1323
+ },
1324
+ {
1325
+ "epoch": 3.44,
1326
+ "learning_rate": 7.027221301997062e-06,
1327
+ "loss": 0.9456,
1328
+ "step": 110000
1329
+ },
1330
+ {
1331
+ "epoch": 3.45,
1332
+ "learning_rate": 6.831890489733412e-06,
1333
+ "loss": 0.9388,
1334
+ "step": 110500
1335
+ },
1336
+ {
1337
+ "epoch": 3.47,
1338
+ "learning_rate": 6.636559677469763e-06,
1339
+ "loss": 0.9385,
1340
+ "step": 111000
1341
+ },
1342
+ {
1343
+ "epoch": 3.48,
1344
+ "learning_rate": 6.441228865206113e-06,
1345
+ "loss": 0.9457,
1346
+ "step": 111500
1347
+ },
1348
+ {
1349
+ "epoch": 3.5,
1350
+ "learning_rate": 6.245898052942464e-06,
1351
+ "loss": 0.9371,
1352
+ "step": 112000
1353
+ },
1354
+ {
1355
+ "epoch": 3.52,
1356
+ "learning_rate": 6.050567240678814e-06,
1357
+ "loss": 0.892,
1358
+ "step": 112500
1359
+ },
1360
+ {
1361
+ "epoch": 3.53,
1362
+ "learning_rate": 5.855236428415165e-06,
1363
+ "loss": 0.9104,
1364
+ "step": 113000
1365
+ },
1366
+ {
1367
+ "epoch": 3.55,
1368
+ "learning_rate": 5.6599056161515146e-06,
1369
+ "loss": 0.9317,
1370
+ "step": 113500
1371
+ },
1372
+ {
1373
+ "epoch": 3.56,
1374
+ "learning_rate": 5.4645748038878645e-06,
1375
+ "loss": 0.8673,
1376
+ "step": 114000
1377
+ },
1378
+ {
1379
+ "epoch": 3.58,
1380
+ "learning_rate": 5.2692439916242145e-06,
1381
+ "loss": 0.9329,
1382
+ "step": 114500
1383
+ },
1384
+ {
1385
+ "epoch": 3.59,
1386
+ "learning_rate": 5.073913179360565e-06,
1387
+ "loss": 0.9171,
1388
+ "step": 115000
1389
+ },
1390
+ {
1391
+ "epoch": 3.61,
1392
+ "learning_rate": 4.878582367096915e-06,
1393
+ "loss": 0.8971,
1394
+ "step": 115500
1395
+ },
1396
+ {
1397
+ "epoch": 3.63,
1398
+ "learning_rate": 4.683251554833266e-06,
1399
+ "loss": 0.9205,
1400
+ "step": 116000
1401
+ },
1402
+ {
1403
+ "epoch": 3.64,
1404
+ "learning_rate": 4.487920742569616e-06,
1405
+ "loss": 0.9046,
1406
+ "step": 116500
1407
+ },
1408
+ {
1409
+ "epoch": 3.66,
1410
+ "learning_rate": 4.292589930305967e-06,
1411
+ "loss": 0.9163,
1412
+ "step": 117000
1413
+ },
1414
+ {
1415
+ "epoch": 3.67,
1416
+ "learning_rate": 4.097259118042317e-06,
1417
+ "loss": 0.9241,
1418
+ "step": 117500
1419
+ },
1420
+ {
1421
+ "epoch": 3.69,
1422
+ "learning_rate": 3.901928305778667e-06,
1423
+ "loss": 0.8993,
1424
+ "step": 118000
1425
+ },
1426
+ {
1427
+ "epoch": 3.7,
1428
+ "learning_rate": 3.7065974935150167e-06,
1429
+ "loss": 0.8975,
1430
+ "step": 118500
1431
+ },
1432
+ {
1433
+ "epoch": 3.72,
1434
+ "learning_rate": 3.5112666812513675e-06,
1435
+ "loss": 0.9508,
1436
+ "step": 119000
1437
+ },
1438
+ {
1439
+ "epoch": 3.73,
1440
+ "learning_rate": 3.315935868987718e-06,
1441
+ "loss": 0.8833,
1442
+ "step": 119500
1443
+ },
1444
+ {
1445
+ "epoch": 3.75,
1446
+ "learning_rate": 3.120605056724068e-06,
1447
+ "loss": 0.941,
1448
+ "step": 120000
1449
+ },
1450
+ {
1451
+ "epoch": 3.77,
1452
+ "learning_rate": 2.9252742444604182e-06,
1453
+ "loss": 0.9286,
1454
+ "step": 120500
1455
+ },
1456
+ {
1457
+ "epoch": 3.78,
1458
+ "learning_rate": 2.7299434321967686e-06,
1459
+ "loss": 0.9103,
1460
+ "step": 121000
1461
+ },
1462
+ {
1463
+ "epoch": 3.8,
1464
+ "learning_rate": 2.534612619933119e-06,
1465
+ "loss": 0.9002,
1466
+ "step": 121500
1467
+ },
1468
+ {
1469
+ "epoch": 3.81,
1470
+ "learning_rate": 2.3392818076694694e-06,
1471
+ "loss": 0.8605,
1472
+ "step": 122000
1473
+ },
1474
+ {
1475
+ "epoch": 3.83,
1476
+ "learning_rate": 2.1439509954058193e-06,
1477
+ "loss": 0.8555,
1478
+ "step": 122500
1479
+ },
1480
+ {
1481
+ "epoch": 3.84,
1482
+ "learning_rate": 1.9486201831421697e-06,
1483
+ "loss": 0.8961,
1484
+ "step": 123000
1485
+ },
1486
+ {
1487
+ "epoch": 3.86,
1488
+ "learning_rate": 1.75328937087852e-06,
1489
+ "loss": 0.9079,
1490
+ "step": 123500
1491
+ },
1492
+ {
1493
+ "epoch": 3.88,
1494
+ "learning_rate": 1.55795855861487e-06,
1495
+ "loss": 0.9456,
1496
+ "step": 124000
1497
+ },
1498
+ {
1499
+ "epoch": 3.89,
1500
+ "learning_rate": 1.3626277463512205e-06,
1501
+ "loss": 0.905,
1502
+ "step": 124500
1503
+ },
1504
+ {
1505
+ "epoch": 3.91,
1506
+ "learning_rate": 1.1672969340875708e-06,
1507
+ "loss": 0.8558,
1508
+ "step": 125000
1509
+ },
1510
+ {
1511
+ "epoch": 3.92,
1512
+ "learning_rate": 9.71966121823921e-07,
1513
+ "loss": 0.9055,
1514
+ "step": 125500
1515
+ },
1516
+ {
1517
+ "epoch": 3.94,
1518
+ "learning_rate": 7.766353095602713e-07,
1519
+ "loss": 0.9165,
1520
+ "step": 126000
1521
+ },
1522
+ {
1523
+ "epoch": 3.95,
1524
+ "learning_rate": 5.813044972966216e-07,
1525
+ "loss": 0.9449,
1526
+ "step": 126500
1527
+ },
1528
+ {
1529
+ "epoch": 3.97,
1530
+ "learning_rate": 3.8597368503297185e-07,
1531
+ "loss": 0.8955,
1532
+ "step": 127000
1533
+ },
1534
+ {
1535
+ "epoch": 3.98,
1536
+ "learning_rate": 1.9064287276932213e-07,
1537
+ "loss": 0.9181,
1538
+ "step": 127500
1539
+ },
1540
+ {
1541
+ "epoch": 4.0,
1542
+ "step": 127988,
1543
+ "total_flos": 7.906061868859392e+16,
1544
+ "train_loss": 2.0082999333540426,
1545
+ "train_runtime": 37501.9411,
1546
+ "train_samples_per_second": 3.413,
1547
+ "train_steps_per_second": 3.413
1548
+ }
1549
+ ],
1550
+ "max_steps": 127988,
1551
+ "num_train_epochs": 4,
1552
+ "total_flos": 7.906061868859392e+16,
1553
+ "trial_name": null,
1554
+ "trial_params": null
1555
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:220f886d0999bdd69b763fad992695f6444333bfadff0c614901535ccdc10495
3
+ size 3119