artelabsuper commited on
Commit
e10840f
1 Parent(s): d8f39ef
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_ilpost
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_ilpost
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: 2.3640
22
+ - Rouge1: 38.9101
23
+ - Rouge2: 21.384
24
+ - Rougel: 32.0517
25
+ - Rougelsum: 35.0743
26
+ - Gen Len: 39.8843
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": 39.8843,
4
+ "eval_loss": 2.3640177249908447,
5
+ "eval_rouge1": 38.9101,
6
+ "eval_rouge2": 21.384,
7
+ "eval_rougeL": 32.0517,
8
+ "eval_rougeLsum": 35.0743,
9
+ "eval_runtime": 3305.1745,
10
+ "eval_samples": 4400,
11
+ "eval_samples_per_second": 1.331,
12
+ "eval_steps_per_second": 1.331,
13
+ "predict_gen_len": 39.8843,
14
+ "predict_loss": 2.3640177249908447,
15
+ "predict_rouge1": 38.9101,
16
+ "predict_rouge2": 21.384,
17
+ "predict_rougeL": 32.0517,
18
+ "predict_rougeLsum": 35.0743,
19
+ "predict_runtime": 3307.2928,
20
+ "predict_samples": 4400,
21
+ "predict_samples_per_second": 1.33,
22
+ "predict_steps_per_second": 1.33,
23
+ "train_loss": 1.5602563560180023,
24
+ "train_runtime": 41389.8114,
25
+ "train_samples": 35201,
26
+ "train_samples_per_second": 3.402,
27
+ "train_steps_per_second": 3.402
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": 39.8843,
4
+ "eval_loss": 2.3640177249908447,
5
+ "eval_rouge1": 38.9101,
6
+ "eval_rouge2": 21.384,
7
+ "eval_rougeL": 32.0517,
8
+ "eval_rougeLsum": 35.0743,
9
+ "eval_runtime": 3305.1745,
10
+ "eval_samples": 4400,
11
+ "eval_samples_per_second": 1.331,
12
+ "eval_steps_per_second": 1.331
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": 39.8843,
3
+ "predict_loss": 2.3640177249908447,
4
+ "predict_rouge1": 38.9101,
5
+ "predict_rouge2": 21.384,
6
+ "predict_rougeL": 32.0517,
7
+ "predict_rougeLsum": 35.0743,
8
+ "predict_runtime": 3307.2928,
9
+ "predict_samples": 4400,
10
+ "predict_samples_per_second": 1.33,
11
+ "predict_steps_per_second": 1.33
12
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5012676683febe9eaa832c566d4528ba148f0b31239ba8ad72b6b3c9b43f595f
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": 1.5602563560180023,
4
+ "train_runtime": 41389.8114,
5
+ "train_samples": 35201,
6
+ "train_samples_per_second": 3.402,
7
+ "train_steps_per_second": 3.402
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1711 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.0,
5
+ "global_step": 140804,
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": 4.9822448225902675e-05,
13
+ "loss": 3.154,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.03,
18
+ "learning_rate": 4.9644896451805353e-05,
19
+ "loss": 2.9047,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.04,
24
+ "learning_rate": 4.946734467770802e-05,
25
+ "loss": 2.8071,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.06,
30
+ "learning_rate": 4.928979290361069e-05,
31
+ "loss": 2.9013,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.07,
36
+ "learning_rate": 4.911224112951337e-05,
37
+ "loss": 2.7409,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.09,
42
+ "learning_rate": 4.893468935541604e-05,
43
+ "loss": 7.1257,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.1,
48
+ "learning_rate": 4.8757137581318714e-05,
49
+ "loss": 2.904,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.11,
54
+ "learning_rate": 4.8579585807221386e-05,
55
+ "loss": 2.823,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.13,
60
+ "learning_rate": 4.840203403312406e-05,
61
+ "loss": 2.7295,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.14,
66
+ "learning_rate": 4.822448225902673e-05,
67
+ "loss": 2.7083,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.16,
72
+ "learning_rate": 4.804693048492941e-05,
73
+ "loss": 2.6503,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.17,
78
+ "learning_rate": 4.786937871083208e-05,
79
+ "loss": 2.6333,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.18,
84
+ "learning_rate": 4.7691826936734754e-05,
85
+ "loss": 2.6817,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.2,
90
+ "learning_rate": 4.7514275162637426e-05,
91
+ "loss": 2.5641,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.21,
96
+ "learning_rate": 4.73367233885401e-05,
97
+ "loss": 2.5646,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.23,
102
+ "learning_rate": 4.715917161444278e-05,
103
+ "loss": 2.5781,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.24,
108
+ "learning_rate": 4.698161984034545e-05,
109
+ "loss": 2.5461,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.26,
114
+ "learning_rate": 4.680406806624812e-05,
115
+ "loss": 2.5121,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.27,
120
+ "learning_rate": 4.662651629215079e-05,
121
+ "loss": 2.5666,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 0.28,
126
+ "learning_rate": 4.6448964518053465e-05,
127
+ "loss": 2.5357,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 0.3,
132
+ "learning_rate": 4.627141274395614e-05,
133
+ "loss": 2.4651,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 0.31,
138
+ "learning_rate": 4.6093860969858816e-05,
139
+ "loss": 2.5033,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 0.33,
144
+ "learning_rate": 4.591630919576149e-05,
145
+ "loss": 2.5227,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 0.34,
150
+ "learning_rate": 4.573875742166416e-05,
151
+ "loss": 2.4618,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 0.36,
156
+ "learning_rate": 4.556120564756683e-05,
157
+ "loss": 2.4844,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 0.37,
162
+ "learning_rate": 4.5383653873469505e-05,
163
+ "loss": 2.5515,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 0.38,
168
+ "learning_rate": 4.520610209937218e-05,
169
+ "loss": 2.5156,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 0.4,
174
+ "learning_rate": 4.5028550325274856e-05,
175
+ "loss": 2.4882,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 0.41,
180
+ "learning_rate": 4.485099855117753e-05,
181
+ "loss": 2.4225,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 0.43,
186
+ "learning_rate": 4.46734467770802e-05,
187
+ "loss": 2.4459,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 0.44,
192
+ "learning_rate": 4.449589500298287e-05,
193
+ "loss": 2.5628,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 0.45,
198
+ "learning_rate": 4.4318343228885544e-05,
199
+ "loss": 2.4405,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 0.47,
204
+ "learning_rate": 4.4140791454788216e-05,
205
+ "loss": 2.5211,
206
+ "step": 16500
207
+ },
208
+ {
209
+ "epoch": 0.48,
210
+ "learning_rate": 4.3963239680690895e-05,
211
+ "loss": 2.5868,
212
+ "step": 17000
213
+ },
214
+ {
215
+ "epoch": 0.5,
216
+ "learning_rate": 4.378568790659357e-05,
217
+ "loss": 2.4548,
218
+ "step": 17500
219
+ },
220
+ {
221
+ "epoch": 0.51,
222
+ "learning_rate": 4.360813613249623e-05,
223
+ "loss": 2.4125,
224
+ "step": 18000
225
+ },
226
+ {
227
+ "epoch": 0.53,
228
+ "learning_rate": 4.343058435839891e-05,
229
+ "loss": 2.3719,
230
+ "step": 18500
231
+ },
232
+ {
233
+ "epoch": 0.54,
234
+ "learning_rate": 4.3253032584301584e-05,
235
+ "loss": 2.4151,
236
+ "step": 19000
237
+ },
238
+ {
239
+ "epoch": 0.55,
240
+ "learning_rate": 4.3075480810204256e-05,
241
+ "loss": 2.4405,
242
+ "step": 19500
243
+ },
244
+ {
245
+ "epoch": 0.57,
246
+ "learning_rate": 4.2897929036106935e-05,
247
+ "loss": 2.4156,
248
+ "step": 20000
249
+ },
250
+ {
251
+ "epoch": 0.58,
252
+ "learning_rate": 4.27203772620096e-05,
253
+ "loss": 2.4121,
254
+ "step": 20500
255
+ },
256
+ {
257
+ "epoch": 0.6,
258
+ "learning_rate": 4.254282548791227e-05,
259
+ "loss": 2.3612,
260
+ "step": 21000
261
+ },
262
+ {
263
+ "epoch": 0.61,
264
+ "learning_rate": 4.236527371381495e-05,
265
+ "loss": 2.3231,
266
+ "step": 21500
267
+ },
268
+ {
269
+ "epoch": 0.62,
270
+ "learning_rate": 4.2187721939717623e-05,
271
+ "loss": 2.3604,
272
+ "step": 22000
273
+ },
274
+ {
275
+ "epoch": 0.64,
276
+ "learning_rate": 4.20101701656203e-05,
277
+ "loss": 2.3752,
278
+ "step": 22500
279
+ },
280
+ {
281
+ "epoch": 0.65,
282
+ "learning_rate": 4.183261839152297e-05,
283
+ "loss": 2.3747,
284
+ "step": 23000
285
+ },
286
+ {
287
+ "epoch": 0.67,
288
+ "learning_rate": 4.165506661742564e-05,
289
+ "loss": 2.3668,
290
+ "step": 23500
291
+ },
292
+ {
293
+ "epoch": 0.68,
294
+ "learning_rate": 4.147751484332832e-05,
295
+ "loss": 2.4354,
296
+ "step": 24000
297
+ },
298
+ {
299
+ "epoch": 0.7,
300
+ "learning_rate": 4.129996306923099e-05,
301
+ "loss": 2.3519,
302
+ "step": 24500
303
+ },
304
+ {
305
+ "epoch": 0.71,
306
+ "learning_rate": 4.112241129513366e-05,
307
+ "loss": 2.3312,
308
+ "step": 25000
309
+ },
310
+ {
311
+ "epoch": 0.72,
312
+ "learning_rate": 4.0944859521036335e-05,
313
+ "loss": 2.3092,
314
+ "step": 25500
315
+ },
316
+ {
317
+ "epoch": 0.74,
318
+ "learning_rate": 4.076730774693901e-05,
319
+ "loss": 2.378,
320
+ "step": 26000
321
+ },
322
+ {
323
+ "epoch": 0.75,
324
+ "learning_rate": 4.058975597284168e-05,
325
+ "loss": 2.371,
326
+ "step": 26500
327
+ },
328
+ {
329
+ "epoch": 0.77,
330
+ "learning_rate": 4.041220419874436e-05,
331
+ "loss": 2.2957,
332
+ "step": 27000
333
+ },
334
+ {
335
+ "epoch": 0.78,
336
+ "learning_rate": 4.023465242464703e-05,
337
+ "loss": 2.3008,
338
+ "step": 27500
339
+ },
340
+ {
341
+ "epoch": 0.8,
342
+ "learning_rate": 4.00571006505497e-05,
343
+ "loss": 2.3263,
344
+ "step": 28000
345
+ },
346
+ {
347
+ "epoch": 0.81,
348
+ "learning_rate": 3.9879548876452375e-05,
349
+ "loss": 2.2668,
350
+ "step": 28500
351
+ },
352
+ {
353
+ "epoch": 0.82,
354
+ "learning_rate": 3.970199710235505e-05,
355
+ "loss": 2.3133,
356
+ "step": 29000
357
+ },
358
+ {
359
+ "epoch": 0.84,
360
+ "learning_rate": 3.952444532825772e-05,
361
+ "loss": 2.2935,
362
+ "step": 29500
363
+ },
364
+ {
365
+ "epoch": 0.85,
366
+ "learning_rate": 3.93468935541604e-05,
367
+ "loss": 2.2905,
368
+ "step": 30000
369
+ },
370
+ {
371
+ "epoch": 0.87,
372
+ "learning_rate": 3.916934178006307e-05,
373
+ "loss": 2.2517,
374
+ "step": 30500
375
+ },
376
+ {
377
+ "epoch": 0.88,
378
+ "learning_rate": 3.899179000596574e-05,
379
+ "loss": 2.1286,
380
+ "step": 31000
381
+ },
382
+ {
383
+ "epoch": 0.89,
384
+ "learning_rate": 3.8814238231868414e-05,
385
+ "loss": 2.2542,
386
+ "step": 31500
387
+ },
388
+ {
389
+ "epoch": 0.91,
390
+ "learning_rate": 3.8636686457771086e-05,
391
+ "loss": 2.272,
392
+ "step": 32000
393
+ },
394
+ {
395
+ "epoch": 0.92,
396
+ "learning_rate": 3.845913468367376e-05,
397
+ "loss": 2.3143,
398
+ "step": 32500
399
+ },
400
+ {
401
+ "epoch": 0.94,
402
+ "learning_rate": 3.828158290957644e-05,
403
+ "loss": 2.2873,
404
+ "step": 33000
405
+ },
406
+ {
407
+ "epoch": 0.95,
408
+ "learning_rate": 3.810403113547911e-05,
409
+ "loss": 2.2756,
410
+ "step": 33500
411
+ },
412
+ {
413
+ "epoch": 0.97,
414
+ "learning_rate": 3.7926479361381775e-05,
415
+ "loss": 2.3062,
416
+ "step": 34000
417
+ },
418
+ {
419
+ "epoch": 0.98,
420
+ "learning_rate": 3.7748927587284454e-05,
421
+ "loss": 2.27,
422
+ "step": 34500
423
+ },
424
+ {
425
+ "epoch": 0.99,
426
+ "learning_rate": 3.7571375813187126e-05,
427
+ "loss": 2.1815,
428
+ "step": 35000
429
+ },
430
+ {
431
+ "epoch": 1.01,
432
+ "learning_rate": 3.7393824039089805e-05,
433
+ "loss": 2.001,
434
+ "step": 35500
435
+ },
436
+ {
437
+ "epoch": 1.02,
438
+ "learning_rate": 3.721627226499248e-05,
439
+ "loss": 1.7794,
440
+ "step": 36000
441
+ },
442
+ {
443
+ "epoch": 1.04,
444
+ "learning_rate": 3.703872049089515e-05,
445
+ "loss": 1.782,
446
+ "step": 36500
447
+ },
448
+ {
449
+ "epoch": 1.05,
450
+ "learning_rate": 3.686116871679782e-05,
451
+ "loss": 1.7742,
452
+ "step": 37000
453
+ },
454
+ {
455
+ "epoch": 1.07,
456
+ "learning_rate": 3.668361694270049e-05,
457
+ "loss": 1.8578,
458
+ "step": 37500
459
+ },
460
+ {
461
+ "epoch": 1.08,
462
+ "learning_rate": 3.6506065168603165e-05,
463
+ "loss": 1.7768,
464
+ "step": 38000
465
+ },
466
+ {
467
+ "epoch": 1.09,
468
+ "learning_rate": 3.6328513394505844e-05,
469
+ "loss": 1.8162,
470
+ "step": 38500
471
+ },
472
+ {
473
+ "epoch": 1.11,
474
+ "learning_rate": 3.6150961620408516e-05,
475
+ "loss": 1.7432,
476
+ "step": 39000
477
+ },
478
+ {
479
+ "epoch": 1.12,
480
+ "learning_rate": 3.597340984631118e-05,
481
+ "loss": 1.7879,
482
+ "step": 39500
483
+ },
484
+ {
485
+ "epoch": 1.14,
486
+ "learning_rate": 3.579585807221386e-05,
487
+ "loss": 1.8667,
488
+ "step": 40000
489
+ },
490
+ {
491
+ "epoch": 1.15,
492
+ "learning_rate": 3.561830629811653e-05,
493
+ "loss": 1.8724,
494
+ "step": 40500
495
+ },
496
+ {
497
+ "epoch": 1.16,
498
+ "learning_rate": 3.5440754524019205e-05,
499
+ "loss": 1.8405,
500
+ "step": 41000
501
+ },
502
+ {
503
+ "epoch": 1.18,
504
+ "learning_rate": 3.5263202749921884e-05,
505
+ "loss": 1.801,
506
+ "step": 41500
507
+ },
508
+ {
509
+ "epoch": 1.19,
510
+ "learning_rate": 3.508565097582455e-05,
511
+ "loss": 1.8085,
512
+ "step": 42000
513
+ },
514
+ {
515
+ "epoch": 1.21,
516
+ "learning_rate": 3.490809920172722e-05,
517
+ "loss": 1.8483,
518
+ "step": 42500
519
+ },
520
+ {
521
+ "epoch": 1.22,
522
+ "learning_rate": 3.47305474276299e-05,
523
+ "loss": 1.7794,
524
+ "step": 43000
525
+ },
526
+ {
527
+ "epoch": 1.24,
528
+ "learning_rate": 3.455299565353257e-05,
529
+ "loss": 1.8514,
530
+ "step": 43500
531
+ },
532
+ {
533
+ "epoch": 1.25,
534
+ "learning_rate": 3.4375443879435245e-05,
535
+ "loss": 1.8237,
536
+ "step": 44000
537
+ },
538
+ {
539
+ "epoch": 1.26,
540
+ "learning_rate": 3.419789210533792e-05,
541
+ "loss": 1.7449,
542
+ "step": 44500
543
+ },
544
+ {
545
+ "epoch": 1.28,
546
+ "learning_rate": 3.402034033124059e-05,
547
+ "loss": 1.8584,
548
+ "step": 45000
549
+ },
550
+ {
551
+ "epoch": 1.29,
552
+ "learning_rate": 3.384278855714326e-05,
553
+ "loss": 1.7689,
554
+ "step": 45500
555
+ },
556
+ {
557
+ "epoch": 1.31,
558
+ "learning_rate": 3.366523678304594e-05,
559
+ "loss": 1.743,
560
+ "step": 46000
561
+ },
562
+ {
563
+ "epoch": 1.32,
564
+ "learning_rate": 3.348768500894861e-05,
565
+ "loss": 1.805,
566
+ "step": 46500
567
+ },
568
+ {
569
+ "epoch": 1.34,
570
+ "learning_rate": 3.3310133234851284e-05,
571
+ "loss": 1.7769,
572
+ "step": 47000
573
+ },
574
+ {
575
+ "epoch": 1.35,
576
+ "learning_rate": 3.3132581460753956e-05,
577
+ "loss": 1.8575,
578
+ "step": 47500
579
+ },
580
+ {
581
+ "epoch": 1.36,
582
+ "learning_rate": 3.295502968665663e-05,
583
+ "loss": 1.8076,
584
+ "step": 48000
585
+ },
586
+ {
587
+ "epoch": 1.38,
588
+ "learning_rate": 3.27774779125593e-05,
589
+ "loss": 1.7576,
590
+ "step": 48500
591
+ },
592
+ {
593
+ "epoch": 1.39,
594
+ "learning_rate": 3.259992613846198e-05,
595
+ "loss": 1.7733,
596
+ "step": 49000
597
+ },
598
+ {
599
+ "epoch": 1.41,
600
+ "learning_rate": 3.242237436436465e-05,
601
+ "loss": 1.8068,
602
+ "step": 49500
603
+ },
604
+ {
605
+ "epoch": 1.42,
606
+ "learning_rate": 3.2244822590267324e-05,
607
+ "loss": 1.771,
608
+ "step": 50000
609
+ },
610
+ {
611
+ "epoch": 1.43,
612
+ "learning_rate": 3.2067270816169996e-05,
613
+ "loss": 1.7543,
614
+ "step": 50500
615
+ },
616
+ {
617
+ "epoch": 1.45,
618
+ "learning_rate": 3.188971904207267e-05,
619
+ "loss": 1.757,
620
+ "step": 51000
621
+ },
622
+ {
623
+ "epoch": 1.46,
624
+ "learning_rate": 3.171216726797535e-05,
625
+ "loss": 1.8191,
626
+ "step": 51500
627
+ },
628
+ {
629
+ "epoch": 1.48,
630
+ "learning_rate": 3.153461549387802e-05,
631
+ "loss": 1.7823,
632
+ "step": 52000
633
+ },
634
+ {
635
+ "epoch": 1.49,
636
+ "learning_rate": 3.135706371978069e-05,
637
+ "loss": 1.7887,
638
+ "step": 52500
639
+ },
640
+ {
641
+ "epoch": 1.51,
642
+ "learning_rate": 3.117951194568336e-05,
643
+ "loss": 1.8028,
644
+ "step": 53000
645
+ },
646
+ {
647
+ "epoch": 1.52,
648
+ "learning_rate": 3.1001960171586035e-05,
649
+ "loss": 1.7986,
650
+ "step": 53500
651
+ },
652
+ {
653
+ "epoch": 1.53,
654
+ "learning_rate": 3.082440839748871e-05,
655
+ "loss": 1.8395,
656
+ "step": 54000
657
+ },
658
+ {
659
+ "epoch": 1.55,
660
+ "learning_rate": 3.0646856623391386e-05,
661
+ "loss": 1.7667,
662
+ "step": 54500
663
+ },
664
+ {
665
+ "epoch": 1.56,
666
+ "learning_rate": 3.0469304849294055e-05,
667
+ "loss": 1.7733,
668
+ "step": 55000
669
+ },
670
+ {
671
+ "epoch": 1.58,
672
+ "learning_rate": 3.0291753075196727e-05,
673
+ "loss": 1.8444,
674
+ "step": 55500
675
+ },
676
+ {
677
+ "epoch": 1.59,
678
+ "learning_rate": 3.0114201301099403e-05,
679
+ "loss": 1.7362,
680
+ "step": 56000
681
+ },
682
+ {
683
+ "epoch": 1.61,
684
+ "learning_rate": 2.9936649527002075e-05,
685
+ "loss": 1.783,
686
+ "step": 56500
687
+ },
688
+ {
689
+ "epoch": 1.62,
690
+ "learning_rate": 2.9759097752904747e-05,
691
+ "loss": 1.7838,
692
+ "step": 57000
693
+ },
694
+ {
695
+ "epoch": 1.63,
696
+ "learning_rate": 2.9581545978807422e-05,
697
+ "loss": 1.8201,
698
+ "step": 57500
699
+ },
700
+ {
701
+ "epoch": 1.65,
702
+ "learning_rate": 2.9403994204710095e-05,
703
+ "loss": 1.7784,
704
+ "step": 58000
705
+ },
706
+ {
707
+ "epoch": 1.66,
708
+ "learning_rate": 2.9226442430612767e-05,
709
+ "loss": 1.7708,
710
+ "step": 58500
711
+ },
712
+ {
713
+ "epoch": 1.68,
714
+ "learning_rate": 2.9048890656515442e-05,
715
+ "loss": 1.7986,
716
+ "step": 59000
717
+ },
718
+ {
719
+ "epoch": 1.69,
720
+ "learning_rate": 2.8871338882418114e-05,
721
+ "loss": 1.7504,
722
+ "step": 59500
723
+ },
724
+ {
725
+ "epoch": 1.7,
726
+ "learning_rate": 2.8693787108320786e-05,
727
+ "loss": 1.7938,
728
+ "step": 60000
729
+ },
730
+ {
731
+ "epoch": 1.72,
732
+ "learning_rate": 2.8516235334223462e-05,
733
+ "loss": 1.8189,
734
+ "step": 60500
735
+ },
736
+ {
737
+ "epoch": 1.73,
738
+ "learning_rate": 2.8338683560126134e-05,
739
+ "loss": 1.8234,
740
+ "step": 61000
741
+ },
742
+ {
743
+ "epoch": 1.75,
744
+ "learning_rate": 2.8161131786028806e-05,
745
+ "loss": 1.7313,
746
+ "step": 61500
747
+ },
748
+ {
749
+ "epoch": 1.76,
750
+ "learning_rate": 2.7983580011931482e-05,
751
+ "loss": 1.7483,
752
+ "step": 62000
753
+ },
754
+ {
755
+ "epoch": 1.78,
756
+ "learning_rate": 2.7806028237834154e-05,
757
+ "loss": 1.7735,
758
+ "step": 62500
759
+ },
760
+ {
761
+ "epoch": 1.79,
762
+ "learning_rate": 2.7628476463736823e-05,
763
+ "loss": 1.7133,
764
+ "step": 63000
765
+ },
766
+ {
767
+ "epoch": 1.8,
768
+ "learning_rate": 2.74509246896395e-05,
769
+ "loss": 1.7454,
770
+ "step": 63500
771
+ },
772
+ {
773
+ "epoch": 1.82,
774
+ "learning_rate": 2.7273372915542174e-05,
775
+ "loss": 1.7284,
776
+ "step": 64000
777
+ },
778
+ {
779
+ "epoch": 1.83,
780
+ "learning_rate": 2.709582114144485e-05,
781
+ "loss": 1.7353,
782
+ "step": 64500
783
+ },
784
+ {
785
+ "epoch": 1.85,
786
+ "learning_rate": 2.691826936734752e-05,
787
+ "loss": 1.751,
788
+ "step": 65000
789
+ },
790
+ {
791
+ "epoch": 1.86,
792
+ "learning_rate": 2.6740717593250193e-05,
793
+ "loss": 1.751,
794
+ "step": 65500
795
+ },
796
+ {
797
+ "epoch": 1.87,
798
+ "learning_rate": 2.656316581915287e-05,
799
+ "loss": 1.7628,
800
+ "step": 66000
801
+ },
802
+ {
803
+ "epoch": 1.89,
804
+ "learning_rate": 2.638561404505554e-05,
805
+ "loss": 1.732,
806
+ "step": 66500
807
+ },
808
+ {
809
+ "epoch": 1.9,
810
+ "learning_rate": 2.620806227095821e-05,
811
+ "loss": 1.7149,
812
+ "step": 67000
813
+ },
814
+ {
815
+ "epoch": 1.92,
816
+ "learning_rate": 2.603051049686089e-05,
817
+ "loss": 1.7349,
818
+ "step": 67500
819
+ },
820
+ {
821
+ "epoch": 1.93,
822
+ "learning_rate": 2.585295872276356e-05,
823
+ "loss": 1.7029,
824
+ "step": 68000
825
+ },
826
+ {
827
+ "epoch": 1.95,
828
+ "learning_rate": 2.567540694866623e-05,
829
+ "loss": 1.7073,
830
+ "step": 68500
831
+ },
832
+ {
833
+ "epoch": 1.96,
834
+ "learning_rate": 2.549785517456891e-05,
835
+ "loss": 1.7113,
836
+ "step": 69000
837
+ },
838
+ {
839
+ "epoch": 1.97,
840
+ "learning_rate": 2.5320303400471577e-05,
841
+ "loss": 1.7561,
842
+ "step": 69500
843
+ },
844
+ {
845
+ "epoch": 1.99,
846
+ "learning_rate": 2.514275162637425e-05,
847
+ "loss": 1.707,
848
+ "step": 70000
849
+ },
850
+ {
851
+ "epoch": 2.0,
852
+ "learning_rate": 2.4965199852276925e-05,
853
+ "loss": 1.6343,
854
+ "step": 70500
855
+ },
856
+ {
857
+ "epoch": 2.02,
858
+ "learning_rate": 2.4787648078179597e-05,
859
+ "loss": 1.2461,
860
+ "step": 71000
861
+ },
862
+ {
863
+ "epoch": 2.03,
864
+ "learning_rate": 2.4610096304082273e-05,
865
+ "loss": 1.1708,
866
+ "step": 71500
867
+ },
868
+ {
869
+ "epoch": 2.05,
870
+ "learning_rate": 2.4432544529984945e-05,
871
+ "loss": 1.2014,
872
+ "step": 72000
873
+ },
874
+ {
875
+ "epoch": 2.06,
876
+ "learning_rate": 2.4254992755887617e-05,
877
+ "loss": 1.223,
878
+ "step": 72500
879
+ },
880
+ {
881
+ "epoch": 2.07,
882
+ "learning_rate": 2.4077440981790292e-05,
883
+ "loss": 1.2299,
884
+ "step": 73000
885
+ },
886
+ {
887
+ "epoch": 2.09,
888
+ "learning_rate": 2.3899889207692964e-05,
889
+ "loss": 1.1634,
890
+ "step": 73500
891
+ },
892
+ {
893
+ "epoch": 2.1,
894
+ "learning_rate": 2.3722337433595637e-05,
895
+ "loss": 1.1944,
896
+ "step": 74000
897
+ },
898
+ {
899
+ "epoch": 2.12,
900
+ "learning_rate": 2.3544785659498312e-05,
901
+ "loss": 1.2264,
902
+ "step": 74500
903
+ },
904
+ {
905
+ "epoch": 2.13,
906
+ "learning_rate": 2.3367233885400984e-05,
907
+ "loss": 1.2196,
908
+ "step": 75000
909
+ },
910
+ {
911
+ "epoch": 2.14,
912
+ "learning_rate": 2.3189682111303656e-05,
913
+ "loss": 1.209,
914
+ "step": 75500
915
+ },
916
+ {
917
+ "epoch": 2.16,
918
+ "learning_rate": 2.3012130337206332e-05,
919
+ "loss": 1.1837,
920
+ "step": 76000
921
+ },
922
+ {
923
+ "epoch": 2.17,
924
+ "learning_rate": 2.2834578563109004e-05,
925
+ "loss": 1.1855,
926
+ "step": 76500
927
+ },
928
+ {
929
+ "epoch": 2.19,
930
+ "learning_rate": 2.2657026789011676e-05,
931
+ "loss": 1.236,
932
+ "step": 77000
933
+ },
934
+ {
935
+ "epoch": 2.2,
936
+ "learning_rate": 2.247947501491435e-05,
937
+ "loss": 1.2629,
938
+ "step": 77500
939
+ },
940
+ {
941
+ "epoch": 2.22,
942
+ "learning_rate": 2.2301923240817024e-05,
943
+ "loss": 1.2625,
944
+ "step": 78000
945
+ },
946
+ {
947
+ "epoch": 2.23,
948
+ "learning_rate": 2.2124371466719696e-05,
949
+ "loss": 1.2181,
950
+ "step": 78500
951
+ },
952
+ {
953
+ "epoch": 2.24,
954
+ "learning_rate": 2.1946819692622368e-05,
955
+ "loss": 1.2551,
956
+ "step": 79000
957
+ },
958
+ {
959
+ "epoch": 2.26,
960
+ "learning_rate": 2.1769267918525044e-05,
961
+ "loss": 1.2755,
962
+ "step": 79500
963
+ },
964
+ {
965
+ "epoch": 2.27,
966
+ "learning_rate": 2.1591716144427716e-05,
967
+ "loss": 1.2594,
968
+ "step": 80000
969
+ },
970
+ {
971
+ "epoch": 2.29,
972
+ "learning_rate": 2.1414164370330388e-05,
973
+ "loss": 1.2469,
974
+ "step": 80500
975
+ },
976
+ {
977
+ "epoch": 2.3,
978
+ "learning_rate": 2.1236612596233063e-05,
979
+ "loss": 1.1807,
980
+ "step": 81000
981
+ },
982
+ {
983
+ "epoch": 2.32,
984
+ "learning_rate": 2.105906082213574e-05,
985
+ "loss": 1.2618,
986
+ "step": 81500
987
+ },
988
+ {
989
+ "epoch": 2.33,
990
+ "learning_rate": 2.0881509048038408e-05,
991
+ "loss": 1.2172,
992
+ "step": 82000
993
+ },
994
+ {
995
+ "epoch": 2.34,
996
+ "learning_rate": 2.0703957273941083e-05,
997
+ "loss": 1.2428,
998
+ "step": 82500
999
+ },
1000
+ {
1001
+ "epoch": 2.36,
1002
+ "learning_rate": 2.0526405499843755e-05,
1003
+ "loss": 1.2209,
1004
+ "step": 83000
1005
+ },
1006
+ {
1007
+ "epoch": 2.37,
1008
+ "learning_rate": 2.0348853725746427e-05,
1009
+ "loss": 1.2078,
1010
+ "step": 83500
1011
+ },
1012
+ {
1013
+ "epoch": 2.39,
1014
+ "learning_rate": 2.0171301951649103e-05,
1015
+ "loss": 1.23,
1016
+ "step": 84000
1017
+ },
1018
+ {
1019
+ "epoch": 2.4,
1020
+ "learning_rate": 1.9993750177551775e-05,
1021
+ "loss": 1.2381,
1022
+ "step": 84500
1023
+ },
1024
+ {
1025
+ "epoch": 2.41,
1026
+ "learning_rate": 1.9816198403454447e-05,
1027
+ "loss": 1.224,
1028
+ "step": 85000
1029
+ },
1030
+ {
1031
+ "epoch": 2.43,
1032
+ "learning_rate": 1.9638646629357123e-05,
1033
+ "loss": 1.2287,
1034
+ "step": 85500
1035
+ },
1036
+ {
1037
+ "epoch": 2.44,
1038
+ "learning_rate": 1.9461094855259795e-05,
1039
+ "loss": 1.1941,
1040
+ "step": 86000
1041
+ },
1042
+ {
1043
+ "epoch": 2.46,
1044
+ "learning_rate": 1.9283543081162467e-05,
1045
+ "loss": 1.222,
1046
+ "step": 86500
1047
+ },
1048
+ {
1049
+ "epoch": 2.47,
1050
+ "learning_rate": 1.9105991307065142e-05,
1051
+ "loss": 1.2172,
1052
+ "step": 87000
1053
+ },
1054
+ {
1055
+ "epoch": 2.49,
1056
+ "learning_rate": 1.8928439532967814e-05,
1057
+ "loss": 1.2138,
1058
+ "step": 87500
1059
+ },
1060
+ {
1061
+ "epoch": 2.5,
1062
+ "learning_rate": 1.8750887758870487e-05,
1063
+ "loss": 1.229,
1064
+ "step": 88000
1065
+ },
1066
+ {
1067
+ "epoch": 2.51,
1068
+ "learning_rate": 1.857333598477316e-05,
1069
+ "loss": 1.2155,
1070
+ "step": 88500
1071
+ },
1072
+ {
1073
+ "epoch": 2.53,
1074
+ "learning_rate": 1.8395784210675834e-05,
1075
+ "loss": 1.1969,
1076
+ "step": 89000
1077
+ },
1078
+ {
1079
+ "epoch": 2.54,
1080
+ "learning_rate": 1.821823243657851e-05,
1081
+ "loss": 1.1926,
1082
+ "step": 89500
1083
+ },
1084
+ {
1085
+ "epoch": 2.56,
1086
+ "learning_rate": 1.804068066248118e-05,
1087
+ "loss": 1.2056,
1088
+ "step": 90000
1089
+ },
1090
+ {
1091
+ "epoch": 2.57,
1092
+ "learning_rate": 1.7863128888383854e-05,
1093
+ "loss": 1.1989,
1094
+ "step": 90500
1095
+ },
1096
+ {
1097
+ "epoch": 2.59,
1098
+ "learning_rate": 1.7685577114286526e-05,
1099
+ "loss": 1.2367,
1100
+ "step": 91000
1101
+ },
1102
+ {
1103
+ "epoch": 2.6,
1104
+ "learning_rate": 1.7508025340189198e-05,
1105
+ "loss": 1.2222,
1106
+ "step": 91500
1107
+ },
1108
+ {
1109
+ "epoch": 2.61,
1110
+ "learning_rate": 1.7330473566091874e-05,
1111
+ "loss": 1.161,
1112
+ "step": 92000
1113
+ },
1114
+ {
1115
+ "epoch": 2.63,
1116
+ "learning_rate": 1.7152921791994546e-05,
1117
+ "loss": 1.2405,
1118
+ "step": 92500
1119
+ },
1120
+ {
1121
+ "epoch": 2.64,
1122
+ "learning_rate": 1.6975370017897218e-05,
1123
+ "loss": 1.2184,
1124
+ "step": 93000
1125
+ },
1126
+ {
1127
+ "epoch": 2.66,
1128
+ "learning_rate": 1.6797818243799894e-05,
1129
+ "loss": 1.2103,
1130
+ "step": 93500
1131
+ },
1132
+ {
1133
+ "epoch": 2.67,
1134
+ "learning_rate": 1.6620266469702566e-05,
1135
+ "loss": 1.2416,
1136
+ "step": 94000
1137
+ },
1138
+ {
1139
+ "epoch": 2.68,
1140
+ "learning_rate": 1.6442714695605238e-05,
1141
+ "loss": 1.2149,
1142
+ "step": 94500
1143
+ },
1144
+ {
1145
+ "epoch": 2.7,
1146
+ "learning_rate": 1.6265162921507913e-05,
1147
+ "loss": 1.1707,
1148
+ "step": 95000
1149
+ },
1150
+ {
1151
+ "epoch": 2.71,
1152
+ "learning_rate": 1.6087611147410585e-05,
1153
+ "loss": 1.2351,
1154
+ "step": 95500
1155
+ },
1156
+ {
1157
+ "epoch": 2.73,
1158
+ "learning_rate": 1.591005937331326e-05,
1159
+ "loss": 1.2179,
1160
+ "step": 96000
1161
+ },
1162
+ {
1163
+ "epoch": 2.74,
1164
+ "learning_rate": 1.5732507599215933e-05,
1165
+ "loss": 1.201,
1166
+ "step": 96500
1167
+ },
1168
+ {
1169
+ "epoch": 2.76,
1170
+ "learning_rate": 1.5554955825118605e-05,
1171
+ "loss": 1.2169,
1172
+ "step": 97000
1173
+ },
1174
+ {
1175
+ "epoch": 2.77,
1176
+ "learning_rate": 1.537740405102128e-05,
1177
+ "loss": 1.1485,
1178
+ "step": 97500
1179
+ },
1180
+ {
1181
+ "epoch": 2.78,
1182
+ "learning_rate": 1.5199852276923951e-05,
1183
+ "loss": 1.166,
1184
+ "step": 98000
1185
+ },
1186
+ {
1187
+ "epoch": 2.8,
1188
+ "learning_rate": 1.5022300502826625e-05,
1189
+ "loss": 1.1665,
1190
+ "step": 98500
1191
+ },
1192
+ {
1193
+ "epoch": 2.81,
1194
+ "learning_rate": 1.4844748728729299e-05,
1195
+ "loss": 1.184,
1196
+ "step": 99000
1197
+ },
1198
+ {
1199
+ "epoch": 2.83,
1200
+ "learning_rate": 1.4667196954631971e-05,
1201
+ "loss": 1.2234,
1202
+ "step": 99500
1203
+ },
1204
+ {
1205
+ "epoch": 2.84,
1206
+ "learning_rate": 1.4489645180534645e-05,
1207
+ "loss": 1.1631,
1208
+ "step": 100000
1209
+ },
1210
+ {
1211
+ "epoch": 2.86,
1212
+ "learning_rate": 1.4312093406437319e-05,
1213
+ "loss": 1.1708,
1214
+ "step": 100500
1215
+ },
1216
+ {
1217
+ "epoch": 2.87,
1218
+ "learning_rate": 1.4134541632339989e-05,
1219
+ "loss": 1.1783,
1220
+ "step": 101000
1221
+ },
1222
+ {
1223
+ "epoch": 2.88,
1224
+ "learning_rate": 1.3956989858242665e-05,
1225
+ "loss": 1.2218,
1226
+ "step": 101500
1227
+ },
1228
+ {
1229
+ "epoch": 2.9,
1230
+ "learning_rate": 1.3779438084145338e-05,
1231
+ "loss": 1.1166,
1232
+ "step": 102000
1233
+ },
1234
+ {
1235
+ "epoch": 2.91,
1236
+ "learning_rate": 1.3601886310048009e-05,
1237
+ "loss": 1.1316,
1238
+ "step": 102500
1239
+ },
1240
+ {
1241
+ "epoch": 2.93,
1242
+ "learning_rate": 1.3424334535950683e-05,
1243
+ "loss": 1.1974,
1244
+ "step": 103000
1245
+ },
1246
+ {
1247
+ "epoch": 2.94,
1248
+ "learning_rate": 1.3246782761853358e-05,
1249
+ "loss": 1.2132,
1250
+ "step": 103500
1251
+ },
1252
+ {
1253
+ "epoch": 2.95,
1254
+ "learning_rate": 1.3069230987756032e-05,
1255
+ "loss": 1.1827,
1256
+ "step": 104000
1257
+ },
1258
+ {
1259
+ "epoch": 2.97,
1260
+ "learning_rate": 1.2891679213658702e-05,
1261
+ "loss": 1.221,
1262
+ "step": 104500
1263
+ },
1264
+ {
1265
+ "epoch": 2.98,
1266
+ "learning_rate": 1.2714127439561376e-05,
1267
+ "loss": 1.1951,
1268
+ "step": 105000
1269
+ },
1270
+ {
1271
+ "epoch": 3.0,
1272
+ "learning_rate": 1.2536575665464052e-05,
1273
+ "loss": 1.1593,
1274
+ "step": 105500
1275
+ },
1276
+ {
1277
+ "epoch": 3.01,
1278
+ "learning_rate": 1.2359023891366724e-05,
1279
+ "loss": 0.8358,
1280
+ "step": 106000
1281
+ },
1282
+ {
1283
+ "epoch": 3.03,
1284
+ "learning_rate": 1.2181472117269396e-05,
1285
+ "loss": 0.7614,
1286
+ "step": 106500
1287
+ },
1288
+ {
1289
+ "epoch": 3.04,
1290
+ "learning_rate": 1.2003920343172068e-05,
1291
+ "loss": 0.7304,
1292
+ "step": 107000
1293
+ },
1294
+ {
1295
+ "epoch": 3.05,
1296
+ "learning_rate": 1.1826368569074744e-05,
1297
+ "loss": 0.7472,
1298
+ "step": 107500
1299
+ },
1300
+ {
1301
+ "epoch": 3.07,
1302
+ "learning_rate": 1.1648816794977416e-05,
1303
+ "loss": 0.7296,
1304
+ "step": 108000
1305
+ },
1306
+ {
1307
+ "epoch": 3.08,
1308
+ "learning_rate": 1.147126502088009e-05,
1309
+ "loss": 0.7666,
1310
+ "step": 108500
1311
+ },
1312
+ {
1313
+ "epoch": 3.1,
1314
+ "learning_rate": 1.1293713246782762e-05,
1315
+ "loss": 0.7406,
1316
+ "step": 109000
1317
+ },
1318
+ {
1319
+ "epoch": 3.11,
1320
+ "learning_rate": 1.1116161472685436e-05,
1321
+ "loss": 0.7183,
1322
+ "step": 109500
1323
+ },
1324
+ {
1325
+ "epoch": 3.12,
1326
+ "learning_rate": 1.093860969858811e-05,
1327
+ "loss": 0.7044,
1328
+ "step": 110000
1329
+ },
1330
+ {
1331
+ "epoch": 3.14,
1332
+ "learning_rate": 1.0761057924490782e-05,
1333
+ "loss": 0.7215,
1334
+ "step": 110500
1335
+ },
1336
+ {
1337
+ "epoch": 3.15,
1338
+ "learning_rate": 1.0583506150393455e-05,
1339
+ "loss": 0.6939,
1340
+ "step": 111000
1341
+ },
1342
+ {
1343
+ "epoch": 3.17,
1344
+ "learning_rate": 1.0405954376296129e-05,
1345
+ "loss": 0.7412,
1346
+ "step": 111500
1347
+ },
1348
+ {
1349
+ "epoch": 3.18,
1350
+ "learning_rate": 1.0228402602198801e-05,
1351
+ "loss": 0.7525,
1352
+ "step": 112000
1353
+ },
1354
+ {
1355
+ "epoch": 3.2,
1356
+ "learning_rate": 1.0050850828101475e-05,
1357
+ "loss": 0.7109,
1358
+ "step": 112500
1359
+ },
1360
+ {
1361
+ "epoch": 3.21,
1362
+ "learning_rate": 9.873299054004149e-06,
1363
+ "loss": 0.7163,
1364
+ "step": 113000
1365
+ },
1366
+ {
1367
+ "epoch": 3.22,
1368
+ "learning_rate": 9.695747279906821e-06,
1369
+ "loss": 0.7131,
1370
+ "step": 113500
1371
+ },
1372
+ {
1373
+ "epoch": 3.24,
1374
+ "learning_rate": 9.518195505809495e-06,
1375
+ "loss": 0.7461,
1376
+ "step": 114000
1377
+ },
1378
+ {
1379
+ "epoch": 3.25,
1380
+ "learning_rate": 9.340643731712167e-06,
1381
+ "loss": 0.6805,
1382
+ "step": 114500
1383
+ },
1384
+ {
1385
+ "epoch": 3.27,
1386
+ "learning_rate": 9.16309195761484e-06,
1387
+ "loss": 0.7688,
1388
+ "step": 115000
1389
+ },
1390
+ {
1391
+ "epoch": 3.28,
1392
+ "learning_rate": 8.985540183517515e-06,
1393
+ "loss": 0.7359,
1394
+ "step": 115500
1395
+ },
1396
+ {
1397
+ "epoch": 3.3,
1398
+ "learning_rate": 8.807988409420187e-06,
1399
+ "loss": 0.7284,
1400
+ "step": 116000
1401
+ },
1402
+ {
1403
+ "epoch": 3.31,
1404
+ "learning_rate": 8.63043663532286e-06,
1405
+ "loss": 0.6892,
1406
+ "step": 116500
1407
+ },
1408
+ {
1409
+ "epoch": 3.32,
1410
+ "learning_rate": 8.452884861225534e-06,
1411
+ "loss": 0.6987,
1412
+ "step": 117000
1413
+ },
1414
+ {
1415
+ "epoch": 3.34,
1416
+ "learning_rate": 8.275333087128207e-06,
1417
+ "loss": 0.7161,
1418
+ "step": 117500
1419
+ },
1420
+ {
1421
+ "epoch": 3.35,
1422
+ "learning_rate": 8.09778131303088e-06,
1423
+ "loss": 0.7185,
1424
+ "step": 118000
1425
+ },
1426
+ {
1427
+ "epoch": 3.37,
1428
+ "learning_rate": 7.920229538933552e-06,
1429
+ "loss": 0.7227,
1430
+ "step": 118500
1431
+ },
1432
+ {
1433
+ "epoch": 3.38,
1434
+ "learning_rate": 7.742677764836228e-06,
1435
+ "loss": 0.6908,
1436
+ "step": 119000
1437
+ },
1438
+ {
1439
+ "epoch": 3.39,
1440
+ "learning_rate": 7.5651259907389e-06,
1441
+ "loss": 0.7278,
1442
+ "step": 119500
1443
+ },
1444
+ {
1445
+ "epoch": 3.41,
1446
+ "learning_rate": 7.387574216641572e-06,
1447
+ "loss": 0.7548,
1448
+ "step": 120000
1449
+ },
1450
+ {
1451
+ "epoch": 3.42,
1452
+ "learning_rate": 7.210022442544247e-06,
1453
+ "loss": 0.7345,
1454
+ "step": 120500
1455
+ },
1456
+ {
1457
+ "epoch": 3.44,
1458
+ "learning_rate": 7.032470668446919e-06,
1459
+ "loss": 0.6865,
1460
+ "step": 121000
1461
+ },
1462
+ {
1463
+ "epoch": 3.45,
1464
+ "learning_rate": 6.854918894349592e-06,
1465
+ "loss": 0.7485,
1466
+ "step": 121500
1467
+ },
1468
+ {
1469
+ "epoch": 3.47,
1470
+ "learning_rate": 6.677367120252266e-06,
1471
+ "loss": 0.7195,
1472
+ "step": 122000
1473
+ },
1474
+ {
1475
+ "epoch": 3.48,
1476
+ "learning_rate": 6.499815346154939e-06,
1477
+ "loss": 0.7444,
1478
+ "step": 122500
1479
+ },
1480
+ {
1481
+ "epoch": 3.49,
1482
+ "learning_rate": 6.322263572057613e-06,
1483
+ "loss": 0.736,
1484
+ "step": 123000
1485
+ },
1486
+ {
1487
+ "epoch": 3.51,
1488
+ "learning_rate": 6.144711797960286e-06,
1489
+ "loss": 0.7284,
1490
+ "step": 123500
1491
+ },
1492
+ {
1493
+ "epoch": 3.52,
1494
+ "learning_rate": 5.967160023862959e-06,
1495
+ "loss": 0.7029,
1496
+ "step": 124000
1497
+ },
1498
+ {
1499
+ "epoch": 3.54,
1500
+ "learning_rate": 5.7896082497656316e-06,
1501
+ "loss": 0.7262,
1502
+ "step": 124500
1503
+ },
1504
+ {
1505
+ "epoch": 3.55,
1506
+ "learning_rate": 5.612056475668305e-06,
1507
+ "loss": 0.7153,
1508
+ "step": 125000
1509
+ },
1510
+ {
1511
+ "epoch": 3.57,
1512
+ "learning_rate": 5.434504701570978e-06,
1513
+ "loss": 0.7528,
1514
+ "step": 125500
1515
+ },
1516
+ {
1517
+ "epoch": 3.58,
1518
+ "learning_rate": 5.256952927473651e-06,
1519
+ "loss": 0.7051,
1520
+ "step": 126000
1521
+ },
1522
+ {
1523
+ "epoch": 3.59,
1524
+ "learning_rate": 5.079401153376324e-06,
1525
+ "loss": 0.7278,
1526
+ "step": 126500
1527
+ },
1528
+ {
1529
+ "epoch": 3.61,
1530
+ "learning_rate": 4.901849379278998e-06,
1531
+ "loss": 0.7206,
1532
+ "step": 127000
1533
+ },
1534
+ {
1535
+ "epoch": 3.62,
1536
+ "learning_rate": 4.724297605181671e-06,
1537
+ "loss": 0.7086,
1538
+ "step": 127500
1539
+ },
1540
+ {
1541
+ "epoch": 3.64,
1542
+ "learning_rate": 4.546745831084345e-06,
1543
+ "loss": 0.7092,
1544
+ "step": 128000
1545
+ },
1546
+ {
1547
+ "epoch": 3.65,
1548
+ "learning_rate": 4.369194056987018e-06,
1549
+ "loss": 0.7109,
1550
+ "step": 128500
1551
+ },
1552
+ {
1553
+ "epoch": 3.66,
1554
+ "learning_rate": 4.191642282889691e-06,
1555
+ "loss": 0.7321,
1556
+ "step": 129000
1557
+ },
1558
+ {
1559
+ "epoch": 3.68,
1560
+ "learning_rate": 4.014090508792364e-06,
1561
+ "loss": 0.7008,
1562
+ "step": 129500
1563
+ },
1564
+ {
1565
+ "epoch": 3.69,
1566
+ "learning_rate": 3.836538734695038e-06,
1567
+ "loss": 0.6671,
1568
+ "step": 130000
1569
+ },
1570
+ {
1571
+ "epoch": 3.71,
1572
+ "learning_rate": 3.6589869605977107e-06,
1573
+ "loss": 0.7047,
1574
+ "step": 130500
1575
+ },
1576
+ {
1577
+ "epoch": 3.72,
1578
+ "learning_rate": 3.4814351865003836e-06,
1579
+ "loss": 0.7033,
1580
+ "step": 131000
1581
+ },
1582
+ {
1583
+ "epoch": 3.74,
1584
+ "learning_rate": 3.303883412403057e-06,
1585
+ "loss": 0.7384,
1586
+ "step": 131500
1587
+ },
1588
+ {
1589
+ "epoch": 3.75,
1590
+ "learning_rate": 3.12633163830573e-06,
1591
+ "loss": 0.7303,
1592
+ "step": 132000
1593
+ },
1594
+ {
1595
+ "epoch": 3.76,
1596
+ "learning_rate": 2.9487798642084034e-06,
1597
+ "loss": 0.7056,
1598
+ "step": 132500
1599
+ },
1600
+ {
1601
+ "epoch": 3.78,
1602
+ "learning_rate": 2.771228090111077e-06,
1603
+ "loss": 0.7186,
1604
+ "step": 133000
1605
+ },
1606
+ {
1607
+ "epoch": 3.79,
1608
+ "learning_rate": 2.5936763160137498e-06,
1609
+ "loss": 0.7086,
1610
+ "step": 133500
1611
+ },
1612
+ {
1613
+ "epoch": 3.81,
1614
+ "learning_rate": 2.416124541916423e-06,
1615
+ "loss": 0.7213,
1616
+ "step": 134000
1617
+ },
1618
+ {
1619
+ "epoch": 3.82,
1620
+ "learning_rate": 2.238572767819096e-06,
1621
+ "loss": 0.6589,
1622
+ "step": 134500
1623
+ },
1624
+ {
1625
+ "epoch": 3.84,
1626
+ "learning_rate": 2.0610209937217696e-06,
1627
+ "loss": 0.7237,
1628
+ "step": 135000
1629
+ },
1630
+ {
1631
+ "epoch": 3.85,
1632
+ "learning_rate": 1.8834692196244425e-06,
1633
+ "loss": 0.711,
1634
+ "step": 135500
1635
+ },
1636
+ {
1637
+ "epoch": 3.86,
1638
+ "learning_rate": 1.705917445527116e-06,
1639
+ "loss": 0.7029,
1640
+ "step": 136000
1641
+ },
1642
+ {
1643
+ "epoch": 3.88,
1644
+ "learning_rate": 1.5283656714297891e-06,
1645
+ "loss": 0.7168,
1646
+ "step": 136500
1647
+ },
1648
+ {
1649
+ "epoch": 3.89,
1650
+ "learning_rate": 1.3508138973324623e-06,
1651
+ "loss": 0.6644,
1652
+ "step": 137000
1653
+ },
1654
+ {
1655
+ "epoch": 3.91,
1656
+ "learning_rate": 1.1732621232351355e-06,
1657
+ "loss": 0.6715,
1658
+ "step": 137500
1659
+ },
1660
+ {
1661
+ "epoch": 3.92,
1662
+ "learning_rate": 9.957103491378087e-07,
1663
+ "loss": 0.6871,
1664
+ "step": 138000
1665
+ },
1666
+ {
1667
+ "epoch": 3.93,
1668
+ "learning_rate": 8.181585750404819e-07,
1669
+ "loss": 0.6982,
1670
+ "step": 138500
1671
+ },
1672
+ {
1673
+ "epoch": 3.95,
1674
+ "learning_rate": 6.40606800943155e-07,
1675
+ "loss": 0.6891,
1676
+ "step": 139000
1677
+ },
1678
+ {
1679
+ "epoch": 3.96,
1680
+ "learning_rate": 4.6305502684582823e-07,
1681
+ "loss": 0.6792,
1682
+ "step": 139500
1683
+ },
1684
+ {
1685
+ "epoch": 3.98,
1686
+ "learning_rate": 2.855032527485015e-07,
1687
+ "loss": 0.7221,
1688
+ "step": 140000
1689
+ },
1690
+ {
1691
+ "epoch": 3.99,
1692
+ "learning_rate": 1.0795147865117468e-07,
1693
+ "loss": 0.7181,
1694
+ "step": 140500
1695
+ },
1696
+ {
1697
+ "epoch": 4.0,
1698
+ "step": 140804,
1699
+ "total_flos": 8.66406121808855e+16,
1700
+ "train_loss": 1.5602563560180023,
1701
+ "train_runtime": 41389.8114,
1702
+ "train_samples_per_second": 3.402,
1703
+ "train_steps_per_second": 3.402
1704
+ }
1705
+ ],
1706
+ "max_steps": 140804,
1707
+ "num_train_epochs": 4,
1708
+ "total_flos": 8.66406121808855e+16,
1709
+ "trial_name": null,
1710
+ "trial_params": null
1711
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eadb00240214e0e29aee093b60627c1c1d971b3c932b3046b5425f7918785f13
3
+ size 3119