Hugging Face's logo

yesj1234
/
jako_mbartLarge_100p_run1

Text2Text Generation
Transformers
Safetensors
Japanese
Korean
mbart
Generated from Trainer
Inference Endpoints
Model card Files Community
yesj1234 commited on
Commit
c8c376e
•
1 Parent(s): de1c0fb

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. .gitattributes +5 -0
  2. README.md +74 -0
  3. all_results.json +22 -0
  4. checkpoint-174018/config.json +59 -0
  5. checkpoint-174018/generation_config.json +12 -0
  6. checkpoint-174018/model.safetensors +3 -0
  7. checkpoint-174018/optimizer.pt +3 -0
  8. checkpoint-174018/rng_state_0.pth +3 -0
  9. checkpoint-174018/rng_state_1.pth +3 -0
  10. checkpoint-174018/rng_state_2.pth +3 -0
  11. checkpoint-174018/rng_state_3.pth +3 -0
  12. checkpoint-174018/scheduler.pt +3 -0
  13. checkpoint-174018/sentencepiece.bpe.model +3 -0
  14. checkpoint-174018/special_tokens_map.json +69 -0
  15. checkpoint-174018/tokenizer.json +3 -0
  16. checkpoint-174018/tokenizer_config.json +528 -0
  17. checkpoint-174018/trainer_state.json +2147 -0
  18. checkpoint-174018/training_args.bin +3 -0
  19. checkpoint-217522/config.json +59 -0
  20. checkpoint-217522/generation_config.json +12 -0
  21. checkpoint-217522/model.safetensors +3 -0
  22. checkpoint-217522/optimizer.pt +3 -0
  23. checkpoint-217522/rng_state_0.pth +3 -0
  24. checkpoint-217522/rng_state_1.pth +3 -0
  25. checkpoint-217522/rng_state_2.pth +3 -0
  26. checkpoint-217522/rng_state_3.pth +3 -0
  27. checkpoint-217522/scheduler.pt +3 -0
  28. checkpoint-217522/sentencepiece.bpe.model +3 -0
  29. checkpoint-217522/special_tokens_map.json +69 -0
  30. checkpoint-217522/tokenizer.json +3 -0
  31. checkpoint-217522/tokenizer_config.json +528 -0
  32. checkpoint-217522/trainer_state.json +2679 -0
  33. checkpoint-217522/training_args.bin +3 -0
  34. checkpoint-87009/config.json +59 -0
  35. checkpoint-87009/generation_config.json +12 -0
  36. checkpoint-87009/model.safetensors +3 -0
  37. checkpoint-87009/optimizer.pt +3 -0
  38. checkpoint-87009/rng_state_0.pth +3 -0
  39. checkpoint-87009/rng_state_1.pth +3 -0
  40. checkpoint-87009/rng_state_2.pth +3 -0
  41. checkpoint-87009/rng_state_3.pth +3 -0
  42. checkpoint-87009/scheduler.pt +3 -0
  43. checkpoint-87009/sentencepiece.bpe.model +3 -0
  44. checkpoint-87009/special_tokens_map.json +69 -0
  45. checkpoint-87009/tokenizer.json +3 -0
  46. checkpoint-87009/tokenizer_config.json +528 -0
  47. checkpoint-87009/trainer_state.json +1083 -0
  48. checkpoint-87009/training_args.bin +3 -0
  49. config.json +59 -0
  50. eval_results.json +10 -0
.gitattributes CHANGED
@@ -33,3 +33,8 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ checkpoint-174018/tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
+ checkpoint-217522/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
+ checkpoint-87009/tokenizer.json filter=lfs diff=lfs merge=lfs -text
39
+ generated_predictions.txt filter=lfs diff=lfs merge=lfs -text
40
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
README.md ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - ja
4
+ - ko
5
+ base_model: facebook/mbart-large-50-many-to-many-mmt
6
+ tags:
7
+ - generated_from_trainer
8
+ metrics:
9
+ - bleu
10
+ model-index:
11
+ - name: jako_mbartLarge_100p_run1
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # jako_mbartLarge_100p_run1
19
+
20
+ This model is a fine-tuned version of [facebook/mbart-large-50-many-to-many-mmt](https://huggingface.co/facebook/mbart-large-50-many-to-many-mmt) on an unknown dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 0.6815
23
+ - Bleu: 57.9768
24
+ - Gen Len: 17.2796
25
+
26
+ ## Model description
27
+
28
+ More information needed
29
+
30
+ ## Intended uses & limitations
31
+
32
+ More information needed
33
+
34
+ ## Training and evaluation data
35
+
36
+ More information needed
37
+
38
+ ## Training procedure
39
+
40
+ ### Training hyperparameters
41
+
42
+ The following hyperparameters were used during training:
43
+ - learning_rate: 5e-05
44
+ - train_batch_size: 2
45
+ - eval_batch_size: 2
46
+ - seed: 42
47
+ - distributed_type: multi-GPU
48
+ - num_devices: 4
49
+ - gradient_accumulation_steps: 2
50
+ - total_train_batch_size: 16
51
+ - total_eval_batch_size: 8
52
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
53
+ - lr_scheduler_type: linear
54
+ - lr_scheduler_warmup_steps: 500
55
+ - num_epochs: 15
56
+ - mixed_precision_training: Native AMP
57
+
58
+ ### Training results
59
+
60
+ | Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len |
61
+ |:-------------:|:-----:|:------:|:---------------:|:-------:|:-------:|
62
+ | 0.7241 | 1.0 | 43504 | 0.7178 | 56.112 | 17.3165 |
63
+ | 0.5523 | 2.0 | 87009 | 0.6815 | 57.9768 | 17.2796 |
64
+ | 0.4356 | 3.0 | 130513 | 0.7014 | 58.5493 | 17.1569 |
65
+ | 0.3407 | 4.0 | 174018 | 0.7513 | 58.3826 | 17.1623 |
66
+ | 0.2605 | 5.0 | 217522 | 0.8122 | 58.7238 | 17.0888 |
67
+
68
+
69
+ ### Framework versions
70
+
71
+ - Transformers 4.35.2
72
+ - Pytorch 2.1.1+cu121
73
+ - Datasets 2.15.0
74
+ - Tokenizers 0.15.0
all_results.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_bleu": 57.9768,
4
+ "eval_gen_len": 17.2796,
5
+ "eval_loss": 0.6814723610877991,
6
+ "eval_runtime": 7593.4193,
7
+ "eval_samples": 87008,
8
+ "eval_samples_per_second": 11.458,
9
+ "eval_steps_per_second": 1.432,
10
+ "predict_bleu": 58.0783,
11
+ "predict_gen_len": 17.2409,
12
+ "predict_loss": 0.6792132258415222,
13
+ "predict_runtime": 7605.1607,
14
+ "predict_samples": 87008,
15
+ "predict_samples_per_second": 11.441,
16
+ "predict_steps_per_second": 1.43,
17
+ "train_loss": 0.5236641572812728,
18
+ "train_runtime": 206817.4399,
19
+ "train_samples": 696070,
20
+ "train_samples_per_second": 50.484,
21
+ "train_steps_per_second": 3.155
22
+ }
checkpoint-174018/config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
3
+ "_num_labels": 3,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "relu",
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": 2,
21
+ "dropout": 0.1,
22
+ "early_stopping": true,
23
+ "encoder_attention_heads": 16,
24
+ "encoder_ffn_dim": 4096,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 12,
27
+ "eos_token_id": 2,
28
+ "forced_bos_token_id": 250014,
29
+ "forced_eos_token_id": 2,
30
+ "gradient_checkpointing": false,
31
+ "id2label": {
32
+ "0": "LABEL_0",
33
+ "1": "LABEL_1",
34
+ "2": "LABEL_2"
35
+ },
36
+ "init_std": 0.02,
37
+ "is_encoder_decoder": true,
38
+ "label2id": {
39
+ "LABEL_0": 0,
40
+ "LABEL_1": 1,
41
+ "LABEL_2": 2
42
+ },
43
+ "max_length": 200,
44
+ "max_position_embeddings": 1024,
45
+ "model_type": "mbart",
46
+ "normalize_before": true,
47
+ "normalize_embedding": true,
48
+ "num_beams": 5,
49
+ "num_hidden_layers": 12,
50
+ "output_past": true,
51
+ "pad_token_id": 1,
52
+ "scale_embedding": true,
53
+ "static_position_embeddings": false,
54
+ "tokenizer_class": "MBart50Tokenizer",
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.35.2",
57
+ "use_cache": true,
58
+ "vocab_size": 250054
59
+ }
checkpoint-174018/generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 2,
4
+ "early_stopping": true,
5
+ "eos_token_id": 2,
6
+ "forced_bos_token_id": 250014,
7
+ "forced_eos_token_id": 2,
8
+ "max_length": 200,
9
+ "num_beams": 5,
10
+ "pad_token_id": 1,
11
+ "transformers_version": "4.35.2"
12
+ }
checkpoint-174018/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce0daaf266acfb58de0e85179a61d51aa567bce7c68bcdb934b0d3a498282e60
3
+ size 2444578688
checkpoint-174018/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dd63b45e4e4b733a906c37454b12b8fb9b9a3244287d5a8e43a457c96501ccca
3
+ size 4887473903
checkpoint-174018/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a78b891684fb7bc91fc7dbbbfd7f39ac988c343fef4b94e549d1492766531d3
3
+ size 15024
checkpoint-174018/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96c03af03e7d52ac4e1a2c50634f1e30eef025648cef26578eabf28b2761e79e
3
+ size 15024
checkpoint-174018/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:81197b1f61f83f0d3897cf98950a95171022cacd5c648bdc859f77040cb0a396
3
+ size 15024
checkpoint-174018/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a114c2303bea63782da5da0b278dd19c752ea443e98f041d04e7a01432897af
3
+ size 15024
checkpoint-174018/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d08813b32a91856678827a4a2af2b3cb7ecd26ffd837c5ca98ef97f1c231a466
3
+ size 1064
checkpoint-174018/sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
checkpoint-174018/special_tokens_map.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "ar_AR",
4
+ "cs_CZ",
5
+ "de_DE",
6
+ "en_XX",
7
+ "es_XX",
8
+ "et_EE",
9
+ "fi_FI",
10
+ "fr_XX",
11
+ "gu_IN",
12
+ "hi_IN",
13
+ "it_IT",
14
+ "ja_XX",
15
+ "kk_KZ",
16
+ "ko_KR",
17
+ "lt_LT",
18
+ "lv_LV",
19
+ "my_MM",
20
+ "ne_NP",
21
+ "nl_XX",
22
+ "ro_RO",
23
+ "ru_RU",
24
+ "si_LK",
25
+ "tr_TR",
26
+ "vi_VN",
27
+ "zh_CN",
28
+ "af_ZA",
29
+ "az_AZ",
30
+ "bn_IN",
31
+ "fa_IR",
32
+ "he_IL",
33
+ "hr_HR",
34
+ "id_ID",
35
+ "ka_GE",
36
+ "km_KH",
37
+ "mk_MK",
38
+ "ml_IN",
39
+ "mn_MN",
40
+ "mr_IN",
41
+ "pl_PL",
42
+ "ps_AF",
43
+ "pt_XX",
44
+ "sv_SE",
45
+ "sw_KE",
46
+ "ta_IN",
47
+ "te_IN",
48
+ "th_TH",
49
+ "tl_XX",
50
+ "uk_UA",
51
+ "ur_PK",
52
+ "xh_ZA",
53
+ "gl_ES",
54
+ "sl_SI"
55
+ ],
56
+ "bos_token": "<s>",
57
+ "cls_token": "<s>",
58
+ "eos_token": "</s>",
59
+ "mask_token": {
60
+ "content": "<mask>",
61
+ "lstrip": true,
62
+ "normalized": true,
63
+ "rstrip": false,
64
+ "single_word": false
65
+ },
66
+ "pad_token": "<pad>",
67
+ "sep_token": "</s>",
68
+ "unk_token": "<unk>"
69
+ }
checkpoint-174018/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:faf1fb0f45fdd5ae9fc44fafee492652a423fc749833ef57c2fa801358549d92
3
+ size 17110107
checkpoint-174018/tokenizer_config.json ADDED
@@ -0,0 +1,528 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "250001": {
36
+ "content": "ar_AR",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "250002": {
44
+ "content": "cs_CZ",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "250003": {
52
+ "content": "de_DE",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "250004": {
60
+ "content": "en_XX",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": true
66
+ },
67
+ "250005": {
68
+ "content": "es_XX",
69
+ "lstrip": false,
70
+ "normalized": false,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": true
74
+ },
75
+ "250006": {
76
+ "content": "et_EE",
77
+ "lstrip": false,
78
+ "normalized": false,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": true
82
+ },
83
+ "250007": {
84
+ "content": "fi_FI",
85
+ "lstrip": false,
86
+ "normalized": false,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": true
90
+ },
91
+ "250008": {
92
+ "content": "fr_XX",
93
+ "lstrip": false,
94
+ "normalized": false,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": true
98
+ },
99
+ "250009": {
100
+ "content": "gu_IN",
101
+ "lstrip": false,
102
+ "normalized": false,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": true
106
+ },
107
+ "250010": {
108
+ "content": "hi_IN",
109
+ "lstrip": false,
110
+ "normalized": false,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": true
114
+ },
115
+ "250011": {
116
+ "content": "it_IT",
117
+ "lstrip": false,
118
+ "normalized": false,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": true
122
+ },
123
+ "250012": {
124
+ "content": "ja_XX",
125
+ "lstrip": false,
126
+ "normalized": false,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": true
130
+ },
131
+ "250013": {
132
+ "content": "kk_KZ",
133
+ "lstrip": false,
134
+ "normalized": false,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": true
138
+ },
139
+ "250014": {
140
+ "content": "ko_KR",
141
+ "lstrip": false,
142
+ "normalized": false,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": true
146
+ },
147
+ "250015": {
148
+ "content": "lt_LT",
149
+ "lstrip": false,
150
+ "normalized": false,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": true
154
+ },
155
+ "250016": {
156
+ "content": "lv_LV",
157
+ "lstrip": false,
158
+ "normalized": false,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": true
162
+ },
163
+ "250017": {
164
+ "content": "my_MM",
165
+ "lstrip": false,
166
+ "normalized": false,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": true
170
+ },
171
+ "250018": {
172
+ "content": "ne_NP",
173
+ "lstrip": false,
174
+ "normalized": false,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": true
178
+ },
179
+ "250019": {
180
+ "content": "nl_XX",
181
+ "lstrip": false,
182
+ "normalized": false,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": true
186
+ },
187
+ "250020": {
188
+ "content": "ro_RO",
189
+ "lstrip": false,
190
+ "normalized": false,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": true
194
+ },
195
+ "250021": {
196
+ "content": "ru_RU",
197
+ "lstrip": false,
198
+ "normalized": false,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": true
202
+ },
203
+ "250022": {
204
+ "content": "si_LK",
205
+ "lstrip": false,
206
+ "normalized": false,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": true
210
+ },
211
+ "250023": {
212
+ "content": "tr_TR",
213
+ "lstrip": false,
214
+ "normalized": false,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": true
218
+ },
219
+ "250024": {
220
+ "content": "vi_VN",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "250025": {
228
+ "content": "zh_CN",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "250026": {
236
+ "content": "af_ZA",
237
+ "lstrip": false,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "250027": {
244
+ "content": "az_AZ",
245
+ "lstrip": false,
246
+ "normalized": false,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "250028": {
252
+ "content": "bn_IN",
253
+ "lstrip": false,
254
+ "normalized": false,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "250029": {
260
+ "content": "fa_IR",
261
+ "lstrip": false,
262
+ "normalized": false,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "250030": {
268
+ "content": "he_IL",
269
+ "lstrip": false,
270
+ "normalized": false,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": true
274
+ },
275
+ "250031": {
276
+ "content": "hr_HR",
277
+ "lstrip": false,
278
+ "normalized": false,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": true
282
+ },
283
+ "250032": {
284
+ "content": "id_ID",
285
+ "lstrip": false,
286
+ "normalized": false,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": true
290
+ },
291
+ "250033": {
292
+ "content": "ka_GE",
293
+ "lstrip": false,
294
+ "normalized": false,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": true
298
+ },
299
+ "250034": {
300
+ "content": "km_KH",
301
+ "lstrip": false,
302
+ "normalized": false,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": true
306
+ },
307
+ "250035": {
308
+ "content": "mk_MK",
309
+ "lstrip": false,
310
+ "normalized": false,
311
+ "rstrip": false,
312
+ "single_word": false,
313
+ "special": true
314
+ },
315
+ "250036": {
316
+ "content": "ml_IN",
317
+ "lstrip": false,
318
+ "normalized": false,
319
+ "rstrip": false,
320
+ "single_word": false,
321
+ "special": true
322
+ },
323
+ "250037": {
324
+ "content": "mn_MN",
325
+ "lstrip": false,
326
+ "normalized": false,
327
+ "rstrip": false,
328
+ "single_word": false,
329
+ "special": true
330
+ },
331
+ "250038": {
332
+ "content": "mr_IN",
333
+ "lstrip": false,
334
+ "normalized": false,
335
+ "rstrip": false,
336
+ "single_word": false,
337
+ "special": true
338
+ },
339
+ "250039": {
340
+ "content": "pl_PL",
341
+ "lstrip": false,
342
+ "normalized": false,
343
+ "rstrip": false,
344
+ "single_word": false,
345
+ "special": true
346
+ },
347
+ "250040": {
348
+ "content": "ps_AF",
349
+ "lstrip": false,
350
+ "normalized": false,
351
+ "rstrip": false,
352
+ "single_word": false,
353
+ "special": true
354
+ },
355
+ "250041": {
356
+ "content": "pt_XX",
357
+ "lstrip": false,
358
+ "normalized": false,
359
+ "rstrip": false,
360
+ "single_word": false,
361
+ "special": true
362
+ },
363
+ "250042": {
364
+ "content": "sv_SE",
365
+ "lstrip": false,
366
+ "normalized": false,
367
+ "rstrip": false,
368
+ "single_word": false,
369
+ "special": true
370
+ },
371
+ "250043": {
372
+ "content": "sw_KE",
373
+ "lstrip": false,
374
+ "normalized": false,
375
+ "rstrip": false,
376
+ "single_word": false,
377
+ "special": true
378
+ },
379
+ "250044": {
380
+ "content": "ta_IN",
381
+ "lstrip": false,
382
+ "normalized": false,
383
+ "rstrip": false,
384
+ "single_word": false,
385
+ "special": true
386
+ },
387
+ "250045": {
388
+ "content": "te_IN",
389
+ "lstrip": false,
390
+ "normalized": false,
391
+ "rstrip": false,
392
+ "single_word": false,
393
+ "special": true
394
+ },
395
+ "250046": {
396
+ "content": "th_TH",
397
+ "lstrip": false,
398
+ "normalized": false,
399
+ "rstrip": false,
400
+ "single_word": false,
401
+ "special": true
402
+ },
403
+ "250047": {
404
+ "content": "tl_XX",
405
+ "lstrip": false,
406
+ "normalized": false,
407
+ "rstrip": false,
408
+ "single_word": false,
409
+ "special": true
410
+ },
411
+ "250048": {
412
+ "content": "uk_UA",
413
+ "lstrip": false,
414
+ "normalized": false,
415
+ "rstrip": false,
416
+ "single_word": false,
417
+ "special": true
418
+ },
419
+ "250049": {
420
+ "content": "ur_PK",
421
+ "lstrip": false,
422
+ "normalized": false,
423
+ "rstrip": false,
424
+ "single_word": false,
425
+ "special": true
426
+ },
427
+ "250050": {
428
+ "content": "xh_ZA",
429
+ "lstrip": false,
430
+ "normalized": false,
431
+ "rstrip": false,
432
+ "single_word": false,
433
+ "special": true
434
+ },
435
+ "250051": {
436
+ "content": "gl_ES",
437
+ "lstrip": false,
438
+ "normalized": false,
439
+ "rstrip": false,
440
+ "single_word": false,
441
+ "special": true
442
+ },
443
+ "250052": {
444
+ "content": "sl_SI",
445
+ "lstrip": false,
446
+ "normalized": false,
447
+ "rstrip": false,
448
+ "single_word": false,
449
+ "special": true
450
+ },
451
+ "250053": {
452
+ "content": "<mask>",
453
+ "lstrip": true,
454
+ "normalized": true,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": true
458
+ }
459
+ },
460
+ "additional_special_tokens": [
461
+ "ar_AR",
462
+ "cs_CZ",
463
+ "de_DE",
464
+ "en_XX",
465
+ "es_XX",
466
+ "et_EE",
467
+ "fi_FI",
468
+ "fr_XX",
469
+ "gu_IN",
470
+ "hi_IN",
471
+ "it_IT",
472
+ "ja_XX",
473
+ "kk_KZ",
474
+ "ko_KR",
475
+ "lt_LT",
476
+ "lv_LV",
477
+ "my_MM",
478
+ "ne_NP",
479
+ "nl_XX",
480
+ "ro_RO",
481
+ "ru_RU",
482
+ "si_LK",
483
+ "tr_TR",
484
+ "vi_VN",
485
+ "zh_CN",
486
+ "af_ZA",
487
+ "az_AZ",
488
+ "bn_IN",
489
+ "fa_IR",
490
+ "he_IL",
491
+ "hr_HR",
492
+ "id_ID",
493
+ "ka_GE",
494
+ "km_KH",
495
+ "mk_MK",
496
+ "ml_IN",
497
+ "mn_MN",
498
+ "mr_IN",
499
+ "pl_PL",
500
+ "ps_AF",
501
+ "pt_XX",
502
+ "sv_SE",
503
+ "sw_KE",
504
+ "ta_IN",
505
+ "te_IN",
506
+ "th_TH",
507
+ "tl_XX",
508
+ "uk_UA",
509
+ "ur_PK",
510
+ "xh_ZA",
511
+ "gl_ES",
512
+ "sl_SI"
513
+ ],
514
+ "bos_token": "<s>",
515
+ "clean_up_tokenization_spaces": true,
516
+ "cls_token": "<s>",
517
+ "eos_token": "</s>",
518
+ "language_codes": "ML50",
519
+ "mask_token": "<mask>",
520
+ "model_max_length": 1000000000000000019884624838656,
521
+ "pad_token": "<pad>",
522
+ "sep_token": "</s>",
523
+ "sp_model_kwargs": {},
524
+ "src_lang": "ja_XX",
525
+ "tgt_lang": "ko_KR",
526
+ "tokenizer_class": "MBart50Tokenizer",
527
+ "unk_token": "<unk>"
528
+ }
checkpoint-174018/trainer_state.json ADDED
@@ -0,0 +1,2147 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6814723610877991,
3
+ "best_model_checkpoint": "./jako_mbartLarge_100p_run1/checkpoint-87009",
4
+ "epoch": 4.0,
5
+ "eval_steps": 500,
6
+ "global_step": 174018,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 4.97e-05,
14
+ "loss": 1.9891,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.02,
19
+ "learning_rate": 4.996196669018189e-05,
20
+ "loss": 1.5354,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.03,
25
+ "learning_rate": 4.9923626660123305e-05,
26
+ "loss": 1.3833,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.05,
31
+ "learning_rate": 4.988528663006472e-05,
32
+ "loss": 1.2912,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.06,
37
+ "learning_rate": 4.984694660000614e-05,
38
+ "loss": 1.2355,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.07,
43
+ "learning_rate": 4.980860656994755e-05,
44
+ "loss": 1.1927,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.08,
49
+ "learning_rate": 4.977026653988897e-05,
50
+ "loss": 1.166,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.09,
55
+ "learning_rate": 4.9731926509830386e-05,
56
+ "loss": 1.1238,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.1,
61
+ "learning_rate": 4.9693586479771804e-05,
62
+ "loss": 1.0972,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.11,
67
+ "learning_rate": 4.9655246449713215e-05,
68
+ "loss": 1.0964,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.13,
73
+ "learning_rate": 4.961690641965463e-05,
74
+ "loss": 1.0611,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.14,
79
+ "learning_rate": 4.957871974971629e-05,
80
+ "loss": 1.0487,
81
+ "step": 6000
82
+ },
83
+ {
84
+ "epoch": 0.15,
85
+ "learning_rate": 4.9540379719657705e-05,
86
+ "loss": 1.0166,
87
+ "step": 6500
88
+ },
89
+ {
90
+ "epoch": 0.16,
91
+ "learning_rate": 4.9502039689599116e-05,
92
+ "loss": 1.0162,
93
+ "step": 7000
94
+ },
95
+ {
96
+ "epoch": 0.17,
97
+ "learning_rate": 4.946369965954053e-05,
98
+ "loss": 0.9952,
99
+ "step": 7500
100
+ },
101
+ {
102
+ "epoch": 0.18,
103
+ "learning_rate": 4.942535962948195e-05,
104
+ "loss": 0.9853,
105
+ "step": 8000
106
+ },
107
+ {
108
+ "epoch": 0.2,
109
+ "learning_rate": 4.9387096279483484e-05,
110
+ "loss": 0.9835,
111
+ "step": 8500
112
+ },
113
+ {
114
+ "epoch": 0.21,
115
+ "learning_rate": 4.9348832929485017e-05,
116
+ "loss": 0.9634,
117
+ "step": 9000
118
+ },
119
+ {
120
+ "epoch": 0.22,
121
+ "learning_rate": 4.9310492899426434e-05,
122
+ "loss": 0.9532,
123
+ "step": 9500
124
+ },
125
+ {
126
+ "epoch": 0.23,
127
+ "learning_rate": 4.927215286936785e-05,
128
+ "loss": 0.955,
129
+ "step": 10000
130
+ },
131
+ {
132
+ "epoch": 0.24,
133
+ "learning_rate": 4.923381283930927e-05,
134
+ "loss": 0.9392,
135
+ "step": 10500
136
+ },
137
+ {
138
+ "epoch": 0.25,
139
+ "learning_rate": 4.919547280925069e-05,
140
+ "loss": 0.9403,
141
+ "step": 11000
142
+ },
143
+ {
144
+ "epoch": 0.26,
145
+ "learning_rate": 4.91571327791921e-05,
146
+ "loss": 0.9177,
147
+ "step": 11500
148
+ },
149
+ {
150
+ "epoch": 0.28,
151
+ "learning_rate": 4.9118792749133516e-05,
152
+ "loss": 0.919,
153
+ "step": 12000
154
+ },
155
+ {
156
+ "epoch": 0.29,
157
+ "learning_rate": 4.908045271907493e-05,
158
+ "loss": 0.9036,
159
+ "step": 12500
160
+ },
161
+ {
162
+ "epoch": 0.3,
163
+ "learning_rate": 4.9042189369076466e-05,
164
+ "loss": 0.9019,
165
+ "step": 13000
166
+ },
167
+ {
168
+ "epoch": 0.31,
169
+ "learning_rate": 4.9003849339017884e-05,
170
+ "loss": 0.8969,
171
+ "step": 13500
172
+ },
173
+ {
174
+ "epoch": 0.32,
175
+ "learning_rate": 4.89655093089593e-05,
176
+ "loss": 0.891,
177
+ "step": 14000
178
+ },
179
+ {
180
+ "epoch": 0.33,
181
+ "learning_rate": 4.892716927890072e-05,
182
+ "loss": 0.8904,
183
+ "step": 14500
184
+ },
185
+ {
186
+ "epoch": 0.34,
187
+ "learning_rate": 4.888882924884213e-05,
188
+ "loss": 0.8833,
189
+ "step": 15000
190
+ },
191
+ {
192
+ "epoch": 0.36,
193
+ "learning_rate": 4.8850489218783554e-05,
194
+ "loss": 0.8668,
195
+ "step": 15500
196
+ },
197
+ {
198
+ "epoch": 0.37,
199
+ "learning_rate": 4.8812149188724965e-05,
200
+ "loss": 0.8692,
201
+ "step": 16000
202
+ },
203
+ {
204
+ "epoch": 0.38,
205
+ "learning_rate": 4.877380915866638e-05,
206
+ "loss": 0.8617,
207
+ "step": 16500
208
+ },
209
+ {
210
+ "epoch": 0.39,
211
+ "learning_rate": 4.8735469128607794e-05,
212
+ "loss": 0.866,
213
+ "step": 17000
214
+ },
215
+ {
216
+ "epoch": 0.4,
217
+ "learning_rate": 4.869720577860933e-05,
218
+ "loss": 0.8592,
219
+ "step": 17500
220
+ },
221
+ {
222
+ "epoch": 0.41,
223
+ "learning_rate": 4.8658865748550744e-05,
224
+ "loss": 0.8586,
225
+ "step": 18000
226
+ },
227
+ {
228
+ "epoch": 0.43,
229
+ "learning_rate": 4.862052571849217e-05,
230
+ "loss": 0.84,
231
+ "step": 18500
232
+ },
233
+ {
234
+ "epoch": 0.44,
235
+ "learning_rate": 4.8582262368493695e-05,
236
+ "loss": 0.84,
237
+ "step": 19000
238
+ },
239
+ {
240
+ "epoch": 0.45,
241
+ "learning_rate": 4.854392233843512e-05,
242
+ "loss": 0.8397,
243
+ "step": 19500
244
+ },
245
+ {
246
+ "epoch": 0.46,
247
+ "learning_rate": 4.850581234855688e-05,
248
+ "loss": 0.8338,
249
+ "step": 20000
250
+ },
251
+ {
252
+ "epoch": 0.47,
253
+ "learning_rate": 4.84674723184983e-05,
254
+ "loss": 0.838,
255
+ "step": 20500
256
+ },
257
+ {
258
+ "epoch": 0.48,
259
+ "learning_rate": 4.842913228843972e-05,
260
+ "loss": 0.8391,
261
+ "step": 21000
262
+ },
263
+ {
264
+ "epoch": 0.49,
265
+ "learning_rate": 4.8390792258381135e-05,
266
+ "loss": 0.8276,
267
+ "step": 21500
268
+ },
269
+ {
270
+ "epoch": 0.51,
271
+ "learning_rate": 4.8352452228322546e-05,
272
+ "loss": 0.8198,
273
+ "step": 22000
274
+ },
275
+ {
276
+ "epoch": 0.52,
277
+ "learning_rate": 4.8314112198263963e-05,
278
+ "loss": 0.8127,
279
+ "step": 22500
280
+ },
281
+ {
282
+ "epoch": 0.53,
283
+ "learning_rate": 4.827577216820538e-05,
284
+ "loss": 0.8184,
285
+ "step": 23000
286
+ },
287
+ {
288
+ "epoch": 0.54,
289
+ "learning_rate": 4.82374321381468e-05,
290
+ "loss": 0.817,
291
+ "step": 23500
292
+ },
293
+ {
294
+ "epoch": 0.55,
295
+ "learning_rate": 4.8199092108088216e-05,
296
+ "loss": 0.8196,
297
+ "step": 24000
298
+ },
299
+ {
300
+ "epoch": 0.56,
301
+ "learning_rate": 4.816075207802963e-05,
302
+ "loss": 0.8024,
303
+ "step": 24500
304
+ },
305
+ {
306
+ "epoch": 0.57,
307
+ "learning_rate": 4.812241204797105e-05,
308
+ "loss": 0.8133,
309
+ "step": 25000
310
+ },
311
+ {
312
+ "epoch": 0.59,
313
+ "learning_rate": 4.808407201791246e-05,
314
+ "loss": 0.8056,
315
+ "step": 25500
316
+ },
317
+ {
318
+ "epoch": 0.6,
319
+ "learning_rate": 4.804573198785388e-05,
320
+ "loss": 0.7972,
321
+ "step": 26000
322
+ },
323
+ {
324
+ "epoch": 0.61,
325
+ "learning_rate": 4.80073919577953e-05,
326
+ "loss": 0.8026,
327
+ "step": 26500
328
+ },
329
+ {
330
+ "epoch": 0.62,
331
+ "learning_rate": 4.7969051927736716e-05,
332
+ "loss": 0.7983,
333
+ "step": 27000
334
+ },
335
+ {
336
+ "epoch": 0.63,
337
+ "learning_rate": 4.793078857773824e-05,
338
+ "loss": 0.795,
339
+ "step": 27500
340
+ },
341
+ {
342
+ "epoch": 0.64,
343
+ "learning_rate": 4.7892448547679666e-05,
344
+ "loss": 0.803,
345
+ "step": 28000
346
+ },
347
+ {
348
+ "epoch": 0.66,
349
+ "learning_rate": 4.7854108517621084e-05,
350
+ "loss": 0.7911,
351
+ "step": 28500
352
+ },
353
+ {
354
+ "epoch": 0.67,
355
+ "learning_rate": 4.7815768487562494e-05,
356
+ "loss": 0.7757,
357
+ "step": 29000
358
+ },
359
+ {
360
+ "epoch": 0.68,
361
+ "learning_rate": 4.777742845750391e-05,
362
+ "loss": 0.777,
363
+ "step": 29500
364
+ },
365
+ {
366
+ "epoch": 0.69,
367
+ "learning_rate": 4.773908842744533e-05,
368
+ "loss": 0.7848,
369
+ "step": 30000
370
+ },
371
+ {
372
+ "epoch": 0.7,
373
+ "learning_rate": 4.770074839738675e-05,
374
+ "loss": 0.7834,
375
+ "step": 30500
376
+ },
377
+ {
378
+ "epoch": 0.71,
379
+ "learning_rate": 4.766240836732816e-05,
380
+ "loss": 0.7753,
381
+ "step": 31000
382
+ },
383
+ {
384
+ "epoch": 0.72,
385
+ "learning_rate": 4.7624068337269576e-05,
386
+ "loss": 0.7715,
387
+ "step": 31500
388
+ },
389
+ {
390
+ "epoch": 0.74,
391
+ "learning_rate": 4.7585728307211e-05,
392
+ "loss": 0.7714,
393
+ "step": 32000
394
+ },
395
+ {
396
+ "epoch": 0.75,
397
+ "learning_rate": 4.754738827715241e-05,
398
+ "loss": 0.7736,
399
+ "step": 32500
400
+ },
401
+ {
402
+ "epoch": 0.76,
403
+ "learning_rate": 4.750904824709383e-05,
404
+ "loss": 0.7767,
405
+ "step": 33000
406
+ },
407
+ {
408
+ "epoch": 0.77,
409
+ "learning_rate": 4.747070821703524e-05,
410
+ "loss": 0.7661,
411
+ "step": 33500
412
+ },
413
+ {
414
+ "epoch": 0.78,
415
+ "learning_rate": 4.7432368186976664e-05,
416
+ "loss": 0.7565,
417
+ "step": 34000
418
+ },
419
+ {
420
+ "epoch": 0.79,
421
+ "learning_rate": 4.739410483697819e-05,
422
+ "loss": 0.7515,
423
+ "step": 34500
424
+ },
425
+ {
426
+ "epoch": 0.8,
427
+ "learning_rate": 4.735584148697973e-05,
428
+ "loss": 0.7558,
429
+ "step": 35000
430
+ },
431
+ {
432
+ "epoch": 0.82,
433
+ "learning_rate": 4.731750145692114e-05,
434
+ "loss": 0.7642,
435
+ "step": 35500
436
+ },
437
+ {
438
+ "epoch": 0.83,
439
+ "learning_rate": 4.727916142686256e-05,
440
+ "loss": 0.7626,
441
+ "step": 36000
442
+ },
443
+ {
444
+ "epoch": 0.84,
445
+ "learning_rate": 4.7240821396803976e-05,
446
+ "loss": 0.7627,
447
+ "step": 36500
448
+ },
449
+ {
450
+ "epoch": 0.85,
451
+ "learning_rate": 4.7202481366745394e-05,
452
+ "loss": 0.7495,
453
+ "step": 37000
454
+ },
455
+ {
456
+ "epoch": 0.86,
457
+ "learning_rate": 4.716414133668681e-05,
458
+ "loss": 0.7582,
459
+ "step": 37500
460
+ },
461
+ {
462
+ "epoch": 0.87,
463
+ "learning_rate": 4.7125877986688344e-05,
464
+ "loss": 0.7481,
465
+ "step": 38000
466
+ },
467
+ {
468
+ "epoch": 0.88,
469
+ "learning_rate": 4.708753795662976e-05,
470
+ "loss": 0.7557,
471
+ "step": 38500
472
+ },
473
+ {
474
+ "epoch": 0.9,
475
+ "learning_rate": 4.704919792657117e-05,
476
+ "loss": 0.7384,
477
+ "step": 39000
478
+ },
479
+ {
480
+ "epoch": 0.91,
481
+ "learning_rate": 4.70108578965126e-05,
482
+ "loss": 0.742,
483
+ "step": 39500
484
+ },
485
+ {
486
+ "epoch": 0.92,
487
+ "learning_rate": 4.697251786645401e-05,
488
+ "loss": 0.7388,
489
+ "step": 40000
490
+ },
491
+ {
492
+ "epoch": 0.93,
493
+ "learning_rate": 4.6934177836395425e-05,
494
+ "loss": 0.7503,
495
+ "step": 40500
496
+ },
497
+ {
498
+ "epoch": 0.94,
499
+ "learning_rate": 4.689583780633684e-05,
500
+ "loss": 0.745,
501
+ "step": 41000
502
+ },
503
+ {
504
+ "epoch": 0.95,
505
+ "learning_rate": 4.685749777627826e-05,
506
+ "loss": 0.7228,
507
+ "step": 41500
508
+ },
509
+ {
510
+ "epoch": 0.97,
511
+ "learning_rate": 4.681915774621968e-05,
512
+ "loss": 0.7333,
513
+ "step": 42000
514
+ },
515
+ {
516
+ "epoch": 0.98,
517
+ "learning_rate": 4.678081771616109e-05,
518
+ "loss": 0.7371,
519
+ "step": 42500
520
+ },
521
+ {
522
+ "epoch": 0.99,
523
+ "learning_rate": 4.674247768610251e-05,
524
+ "loss": 0.733,
525
+ "step": 43000
526
+ },
527
+ {
528
+ "epoch": 1.0,
529
+ "learning_rate": 4.6704137656043925e-05,
530
+ "loss": 0.7241,
531
+ "step": 43500
532
+ },
533
+ {
534
+ "epoch": 1.0,
535
+ "eval_bleu": 56.112,
536
+ "eval_gen_len": 17.3165,
537
+ "eval_loss": 0.7177675366401672,
538
+ "eval_runtime": 7671.1849,
539
+ "eval_samples_per_second": 11.342,
540
+ "eval_steps_per_second": 1.418,
541
+ "step": 43504
542
+ },
543
+ {
544
+ "epoch": 1.01,
545
+ "learning_rate": 4.666587430604546e-05,
546
+ "loss": 0.7322,
547
+ "step": 44000
548
+ },
549
+ {
550
+ "epoch": 1.02,
551
+ "learning_rate": 4.6627534275986875e-05,
552
+ "loss": 0.7051,
553
+ "step": 44500
554
+ },
555
+ {
556
+ "epoch": 1.03,
557
+ "learning_rate": 4.658919424592829e-05,
558
+ "loss": 0.6935,
559
+ "step": 45000
560
+ },
561
+ {
562
+ "epoch": 1.05,
563
+ "learning_rate": 4.6550854215869704e-05,
564
+ "loss": 0.6831,
565
+ "step": 45500
566
+ },
567
+ {
568
+ "epoch": 1.06,
569
+ "learning_rate": 4.651259086587124e-05,
570
+ "loss": 0.6787,
571
+ "step": 46000
572
+ },
573
+ {
574
+ "epoch": 1.07,
575
+ "learning_rate": 4.6474327515872776e-05,
576
+ "loss": 0.6813,
577
+ "step": 46500
578
+ },
579
+ {
580
+ "epoch": 1.08,
581
+ "learning_rate": 4.6436140845934424e-05,
582
+ "loss": 0.6841,
583
+ "step": 47000
584
+ },
585
+ {
586
+ "epoch": 1.09,
587
+ "learning_rate": 4.639780081587584e-05,
588
+ "loss": 0.6697,
589
+ "step": 47500
590
+ },
591
+ {
592
+ "epoch": 1.1,
593
+ "learning_rate": 4.635946078581726e-05,
594
+ "loss": 0.6605,
595
+ "step": 48000
596
+ },
597
+ {
598
+ "epoch": 1.11,
599
+ "learning_rate": 4.632112075575868e-05,
600
+ "loss": 0.675,
601
+ "step": 48500
602
+ },
603
+ {
604
+ "epoch": 1.13,
605
+ "learning_rate": 4.6282780725700094e-05,
606
+ "loss": 0.6623,
607
+ "step": 49000
608
+ },
609
+ {
610
+ "epoch": 1.14,
611
+ "learning_rate": 4.6244440695641505e-05,
612
+ "loss": 0.6589,
613
+ "step": 49500
614
+ },
615
+ {
616
+ "epoch": 1.15,
617
+ "learning_rate": 4.620610066558292e-05,
618
+ "loss": 0.6476,
619
+ "step": 50000
620
+ },
621
+ {
622
+ "epoch": 1.16,
623
+ "learning_rate": 4.616776063552434e-05,
624
+ "loss": 0.6551,
625
+ "step": 50500
626
+ },
627
+ {
628
+ "epoch": 1.17,
629
+ "learning_rate": 4.612942060546576e-05,
630
+ "loss": 0.6502,
631
+ "step": 51000
632
+ },
633
+ {
634
+ "epoch": 1.18,
635
+ "learning_rate": 4.6091080575407176e-05,
636
+ "loss": 0.6478,
637
+ "step": 51500
638
+ },
639
+ {
640
+ "epoch": 1.2,
641
+ "learning_rate": 4.605274054534859e-05,
642
+ "loss": 0.6485,
643
+ "step": 52000
644
+ },
645
+ {
646
+ "epoch": 1.21,
647
+ "learning_rate": 4.6014400515290004e-05,
648
+ "loss": 0.6412,
649
+ "step": 52500
650
+ },
651
+ {
652
+ "epoch": 1.22,
653
+ "learning_rate": 4.597606048523142e-05,
654
+ "loss": 0.6376,
655
+ "step": 53000
656
+ },
657
+ {
658
+ "epoch": 1.23,
659
+ "learning_rate": 4.593772045517284e-05,
660
+ "loss": 0.6385,
661
+ "step": 53500
662
+ },
663
+ {
664
+ "epoch": 1.24,
665
+ "learning_rate": 4.589945710517437e-05,
666
+ "loss": 0.6366,
667
+ "step": 54000
668
+ },
669
+ {
670
+ "epoch": 1.25,
671
+ "learning_rate": 4.586111707511579e-05,
672
+ "loss": 0.6444,
673
+ "step": 54500
674
+ },
675
+ {
676
+ "epoch": 1.26,
677
+ "learning_rate": 4.582277704505721e-05,
678
+ "loss": 0.6311,
679
+ "step": 55000
680
+ },
681
+ {
682
+ "epoch": 1.28,
683
+ "learning_rate": 4.578451369505874e-05,
684
+ "loss": 0.6306,
685
+ "step": 55500
686
+ },
687
+ {
688
+ "epoch": 1.29,
689
+ "learning_rate": 4.574617366500016e-05,
690
+ "loss": 0.6199,
691
+ "step": 56000
692
+ },
693
+ {
694
+ "epoch": 1.3,
695
+ "learning_rate": 4.570783363494157e-05,
696
+ "loss": 0.6268,
697
+ "step": 56500
698
+ },
699
+ {
700
+ "epoch": 1.31,
701
+ "learning_rate": 4.566949360488299e-05,
702
+ "loss": 0.623,
703
+ "step": 57000
704
+ },
705
+ {
706
+ "epoch": 1.32,
707
+ "learning_rate": 4.5631153574824404e-05,
708
+ "loss": 0.6207,
709
+ "step": 57500
710
+ },
711
+ {
712
+ "epoch": 1.33,
713
+ "learning_rate": 4.559281354476582e-05,
714
+ "loss": 0.6204,
715
+ "step": 58000
716
+ },
717
+ {
718
+ "epoch": 1.34,
719
+ "learning_rate": 4.555447351470723e-05,
720
+ "loss": 0.6215,
721
+ "step": 58500
722
+ },
723
+ {
724
+ "epoch": 1.36,
725
+ "learning_rate": 4.551613348464866e-05,
726
+ "loss": 0.6114,
727
+ "step": 59000
728
+ },
729
+ {
730
+ "epoch": 1.37,
731
+ "learning_rate": 4.5477793454590075e-05,
732
+ "loss": 0.6134,
733
+ "step": 59500
734
+ },
735
+ {
736
+ "epoch": 1.38,
737
+ "learning_rate": 4.543953010459161e-05,
738
+ "loss": 0.6142,
739
+ "step": 60000
740
+ },
741
+ {
742
+ "epoch": 1.39,
743
+ "learning_rate": 4.5401190074533025e-05,
744
+ "loss": 0.6149,
745
+ "step": 60500
746
+ },
747
+ {
748
+ "epoch": 1.4,
749
+ "learning_rate": 4.5362850044474436e-05,
750
+ "loss": 0.6123,
751
+ "step": 61000
752
+ },
753
+ {
754
+ "epoch": 1.41,
755
+ "learning_rate": 4.5324510014415854e-05,
756
+ "loss": 0.6159,
757
+ "step": 61500
758
+ },
759
+ {
760
+ "epoch": 1.43,
761
+ "learning_rate": 4.5286246664417387e-05,
762
+ "loss": 0.6001,
763
+ "step": 62000
764
+ },
765
+ {
766
+ "epoch": 1.44,
767
+ "learning_rate": 4.5247906634358804e-05,
768
+ "loss": 0.6034,
769
+ "step": 62500
770
+ },
771
+ {
772
+ "epoch": 1.45,
773
+ "learning_rate": 4.520956660430022e-05,
774
+ "loss": 0.5999,
775
+ "step": 63000
776
+ },
777
+ {
778
+ "epoch": 1.46,
779
+ "learning_rate": 4.517122657424164e-05,
780
+ "loss": 0.6011,
781
+ "step": 63500
782
+ },
783
+ {
784
+ "epoch": 1.47,
785
+ "learning_rate": 4.513288654418305e-05,
786
+ "loss": 0.6051,
787
+ "step": 64000
788
+ },
789
+ {
790
+ "epoch": 1.48,
791
+ "learning_rate": 4.509454651412447e-05,
792
+ "loss": 0.6061,
793
+ "step": 64500
794
+ },
795
+ {
796
+ "epoch": 1.49,
797
+ "learning_rate": 4.5056206484065886e-05,
798
+ "loss": 0.598,
799
+ "step": 65000
800
+ },
801
+ {
802
+ "epoch": 1.51,
803
+ "learning_rate": 4.50178664540073e-05,
804
+ "loss": 0.5939,
805
+ "step": 65500
806
+ },
807
+ {
808
+ "epoch": 1.52,
809
+ "learning_rate": 4.497952642394872e-05,
810
+ "loss": 0.5905,
811
+ "step": 66000
812
+ },
813
+ {
814
+ "epoch": 1.53,
815
+ "learning_rate": 4.494118639389013e-05,
816
+ "loss": 0.5968,
817
+ "step": 66500
818
+ },
819
+ {
820
+ "epoch": 1.54,
821
+ "learning_rate": 4.490284636383155e-05,
822
+ "loss": 0.5943,
823
+ "step": 67000
824
+ },
825
+ {
826
+ "epoch": 1.55,
827
+ "learning_rate": 4.486450633377297e-05,
828
+ "loss": 0.6025,
829
+ "step": 67500
830
+ },
831
+ {
832
+ "epoch": 1.56,
833
+ "learning_rate": 4.4826166303714385e-05,
834
+ "loss": 0.5918,
835
+ "step": 68000
836
+ },
837
+ {
838
+ "epoch": 1.57,
839
+ "learning_rate": 4.47878262736558e-05,
840
+ "loss": 0.5983,
841
+ "step": 68500
842
+ },
843
+ {
844
+ "epoch": 1.59,
845
+ "learning_rate": 4.474948624359721e-05,
846
+ "loss": 0.5907,
847
+ "step": 69000
848
+ },
849
+ {
850
+ "epoch": 1.6,
851
+ "learning_rate": 4.471114621353864e-05,
852
+ "loss": 0.5852,
853
+ "step": 69500
854
+ },
855
+ {
856
+ "epoch": 1.61,
857
+ "learning_rate": 4.4672882863540164e-05,
858
+ "loss": 0.5886,
859
+ "step": 70000
860
+ },
861
+ {
862
+ "epoch": 1.62,
863
+ "learning_rate": 4.463454283348159e-05,
864
+ "loss": 0.5895,
865
+ "step": 70500
866
+ },
867
+ {
868
+ "epoch": 1.63,
869
+ "learning_rate": 4.4596202803423e-05,
870
+ "loss": 0.5887,
871
+ "step": 71000
872
+ },
873
+ {
874
+ "epoch": 1.64,
875
+ "learning_rate": 4.455786277336442e-05,
876
+ "loss": 0.5953,
877
+ "step": 71500
878
+ },
879
+ {
880
+ "epoch": 1.66,
881
+ "learning_rate": 4.451952274330583e-05,
882
+ "loss": 0.5888,
883
+ "step": 72000
884
+ },
885
+ {
886
+ "epoch": 1.67,
887
+ "learning_rate": 4.448118271324725e-05,
888
+ "loss": 0.571,
889
+ "step": 72500
890
+ },
891
+ {
892
+ "epoch": 1.68,
893
+ "learning_rate": 4.444291936324878e-05,
894
+ "loss": 0.576,
895
+ "step": 73000
896
+ },
897
+ {
898
+ "epoch": 1.69,
899
+ "learning_rate": 4.44045793331902e-05,
900
+ "loss": 0.5828,
901
+ "step": 73500
902
+ },
903
+ {
904
+ "epoch": 1.7,
905
+ "learning_rate": 4.436631598319173e-05,
906
+ "loss": 0.5875,
907
+ "step": 74000
908
+ },
909
+ {
910
+ "epoch": 1.71,
911
+ "learning_rate": 4.432797595313315e-05,
912
+ "loss": 0.5773,
913
+ "step": 74500
914
+ },
915
+ {
916
+ "epoch": 1.72,
917
+ "learning_rate": 4.428963592307457e-05,
918
+ "loss": 0.5741,
919
+ "step": 75000
920
+ },
921
+ {
922
+ "epoch": 1.74,
923
+ "learning_rate": 4.425129589301598e-05,
924
+ "loss": 0.5721,
925
+ "step": 75500
926
+ },
927
+ {
928
+ "epoch": 1.75,
929
+ "learning_rate": 4.42129558629574e-05,
930
+ "loss": 0.5797,
931
+ "step": 76000
932
+ },
933
+ {
934
+ "epoch": 1.76,
935
+ "learning_rate": 4.417461583289882e-05,
936
+ "loss": 0.5787,
937
+ "step": 76500
938
+ },
939
+ {
940
+ "epoch": 1.77,
941
+ "learning_rate": 4.4136275802840234e-05,
942
+ "loss": 0.5731,
943
+ "step": 77000
944
+ },
945
+ {
946
+ "epoch": 1.78,
947
+ "learning_rate": 4.409801245284177e-05,
948
+ "loss": 0.5642,
949
+ "step": 77500
950
+ },
951
+ {
952
+ "epoch": 1.79,
953
+ "learning_rate": 4.4059672422783185e-05,
954
+ "loss": 0.5653,
955
+ "step": 78000
956
+ },
957
+ {
958
+ "epoch": 1.8,
959
+ "learning_rate": 4.4021332392724596e-05,
960
+ "loss": 0.5663,
961
+ "step": 78500
962
+ },
963
+ {
964
+ "epoch": 1.82,
965
+ "learning_rate": 4.398299236266601e-05,
966
+ "loss": 0.5762,
967
+ "step": 79000
968
+ },
969
+ {
970
+ "epoch": 1.83,
971
+ "learning_rate": 4.394465233260743e-05,
972
+ "loss": 0.5742,
973
+ "step": 79500
974
+ },
975
+ {
976
+ "epoch": 1.84,
977
+ "learning_rate": 4.390631230254885e-05,
978
+ "loss": 0.5739,
979
+ "step": 80000
980
+ },
981
+ {
982
+ "epoch": 1.85,
983
+ "learning_rate": 4.3867972272490266e-05,
984
+ "loss": 0.5686,
985
+ "step": 80500
986
+ },
987
+ {
988
+ "epoch": 1.86,
989
+ "learning_rate": 4.382963224243168e-05,
990
+ "loss": 0.5757,
991
+ "step": 81000
992
+ },
993
+ {
994
+ "epoch": 1.87,
995
+ "learning_rate": 4.3791292212373095e-05,
996
+ "loss": 0.5654,
997
+ "step": 81500
998
+ },
999
+ {
1000
+ "epoch": 1.88,
1001
+ "learning_rate": 4.375295218231451e-05,
1002
+ "loss": 0.5697,
1003
+ "step": 82000
1004
+ },
1005
+ {
1006
+ "epoch": 1.9,
1007
+ "learning_rate": 4.371461215225593e-05,
1008
+ "loss": 0.5582,
1009
+ "step": 82500
1010
+ },
1011
+ {
1012
+ "epoch": 1.91,
1013
+ "learning_rate": 4.367627212219735e-05,
1014
+ "loss": 0.5603,
1015
+ "step": 83000
1016
+ },
1017
+ {
1018
+ "epoch": 1.92,
1019
+ "learning_rate": 4.363793209213876e-05,
1020
+ "loss": 0.5564,
1021
+ "step": 83500
1022
+ },
1023
+ {
1024
+ "epoch": 1.93,
1025
+ "learning_rate": 4.359959206208018e-05,
1026
+ "loss": 0.5682,
1027
+ "step": 84000
1028
+ },
1029
+ {
1030
+ "epoch": 1.94,
1031
+ "learning_rate": 4.356132871208171e-05,
1032
+ "loss": 0.566,
1033
+ "step": 84500
1034
+ },
1035
+ {
1036
+ "epoch": 1.95,
1037
+ "learning_rate": 4.3522988682023133e-05,
1038
+ "loss": 0.5485,
1039
+ "step": 85000
1040
+ },
1041
+ {
1042
+ "epoch": 1.97,
1043
+ "learning_rate": 4.3484648651964544e-05,
1044
+ "loss": 0.5558,
1045
+ "step": 85500
1046
+ },
1047
+ {
1048
+ "epoch": 1.98,
1049
+ "learning_rate": 4.344630862190596e-05,
1050
+ "loss": 0.5592,
1051
+ "step": 86000
1052
+ },
1053
+ {
1054
+ "epoch": 1.99,
1055
+ "learning_rate": 4.340796859184737e-05,
1056
+ "loss": 0.5611,
1057
+ "step": 86500
1058
+ },
1059
+ {
1060
+ "epoch": 2.0,
1061
+ "learning_rate": 4.337001196208938e-05,
1062
+ "loss": 0.5523,
1063
+ "step": 87000
1064
+ },
1065
+ {
1066
+ "epoch": 2.0,
1067
+ "eval_bleu": 57.9768,
1068
+ "eval_gen_len": 17.2796,
1069
+ "eval_loss": 0.6814723610877991,
1070
+ "eval_runtime": 7592.1301,
1071
+ "eval_samples_per_second": 11.46,
1072
+ "eval_steps_per_second": 1.433,
1073
+ "step": 87009
1074
+ },
1075
+ {
1076
+ "epoch": 2.01,
1077
+ "learning_rate": 4.33316719320308e-05,
1078
+ "loss": 0.5617,
1079
+ "step": 87500
1080
+ },
1081
+ {
1082
+ "epoch": 2.02,
1083
+ "learning_rate": 4.3293331901972215e-05,
1084
+ "loss": 0.5392,
1085
+ "step": 88000
1086
+ },
1087
+ {
1088
+ "epoch": 2.03,
1089
+ "learning_rate": 4.325499187191363e-05,
1090
+ "loss": 0.5313,
1091
+ "step": 88500
1092
+ },
1093
+ {
1094
+ "epoch": 2.05,
1095
+ "learning_rate": 4.321665184185504e-05,
1096
+ "loss": 0.5247,
1097
+ "step": 89000
1098
+ },
1099
+ {
1100
+ "epoch": 2.06,
1101
+ "learning_rate": 4.317831181179646e-05,
1102
+ "loss": 0.5183,
1103
+ "step": 89500
1104
+ },
1105
+ {
1106
+ "epoch": 2.07,
1107
+ "learning_rate": 4.313997178173788e-05,
1108
+ "loss": 0.5266,
1109
+ "step": 90000
1110
+ },
1111
+ {
1112
+ "epoch": 2.08,
1113
+ "learning_rate": 4.3101631751679296e-05,
1114
+ "loss": 0.5256,
1115
+ "step": 90500
1116
+ },
1117
+ {
1118
+ "epoch": 2.09,
1119
+ "learning_rate": 4.3063291721620714e-05,
1120
+ "loss": 0.5148,
1121
+ "step": 91000
1122
+ },
1123
+ {
1124
+ "epoch": 2.1,
1125
+ "learning_rate": 4.3024951691562125e-05,
1126
+ "loss": 0.5097,
1127
+ "step": 91500
1128
+ },
1129
+ {
1130
+ "epoch": 2.11,
1131
+ "learning_rate": 4.298661166150354e-05,
1132
+ "loss": 0.5192,
1133
+ "step": 92000
1134
+ },
1135
+ {
1136
+ "epoch": 2.13,
1137
+ "learning_rate": 4.294827163144496e-05,
1138
+ "loss": 0.5125,
1139
+ "step": 92500
1140
+ },
1141
+ {
1142
+ "epoch": 2.14,
1143
+ "learning_rate": 4.290993160138638e-05,
1144
+ "loss": 0.5049,
1145
+ "step": 93000
1146
+ },
1147
+ {
1148
+ "epoch": 2.15,
1149
+ "learning_rate": 4.2871591571327795e-05,
1150
+ "loss": 0.4974,
1151
+ "step": 93500
1152
+ },
1153
+ {
1154
+ "epoch": 2.16,
1155
+ "learning_rate": 4.2833251541269206e-05,
1156
+ "loss": 0.5044,
1157
+ "step": 94000
1158
+ },
1159
+ {
1160
+ "epoch": 2.17,
1161
+ "learning_rate": 4.279491151121063e-05,
1162
+ "loss": 0.5026,
1163
+ "step": 94500
1164
+ },
1165
+ {
1166
+ "epoch": 2.18,
1167
+ "learning_rate": 4.275657148115204e-05,
1168
+ "loss": 0.5028,
1169
+ "step": 95000
1170
+ },
1171
+ {
1172
+ "epoch": 2.2,
1173
+ "learning_rate": 4.271823145109346e-05,
1174
+ "loss": 0.4992,
1175
+ "step": 95500
1176
+ },
1177
+ {
1178
+ "epoch": 2.21,
1179
+ "learning_rate": 4.267989142103488e-05,
1180
+ "loss": 0.5,
1181
+ "step": 96000
1182
+ },
1183
+ {
1184
+ "epoch": 2.22,
1185
+ "learning_rate": 4.264162807103641e-05,
1186
+ "loss": 0.4942,
1187
+ "step": 96500
1188
+ },
1189
+ {
1190
+ "epoch": 2.23,
1191
+ "learning_rate": 4.260328804097782e-05,
1192
+ "loss": 0.4942,
1193
+ "step": 97000
1194
+ },
1195
+ {
1196
+ "epoch": 2.24,
1197
+ "learning_rate": 4.2564948010919245e-05,
1198
+ "loss": 0.4952,
1199
+ "step": 97500
1200
+ },
1201
+ {
1202
+ "epoch": 2.25,
1203
+ "learning_rate": 4.252660798086066e-05,
1204
+ "loss": 0.4977,
1205
+ "step": 98000
1206
+ },
1207
+ {
1208
+ "epoch": 2.26,
1209
+ "learning_rate": 4.2488344630862195e-05,
1210
+ "loss": 0.4894,
1211
+ "step": 98500
1212
+ },
1213
+ {
1214
+ "epoch": 2.28,
1215
+ "learning_rate": 4.245008128086372e-05,
1216
+ "loss": 0.4887,
1217
+ "step": 99000
1218
+ },
1219
+ {
1220
+ "epoch": 2.29,
1221
+ "learning_rate": 4.2411741250805146e-05,
1222
+ "loss": 0.4813,
1223
+ "step": 99500
1224
+ },
1225
+ {
1226
+ "epoch": 2.3,
1227
+ "learning_rate": 4.2373401220746563e-05,
1228
+ "loss": 0.4851,
1229
+ "step": 100000
1230
+ },
1231
+ {
1232
+ "epoch": 2.31,
1233
+ "learning_rate": 4.2335061190687974e-05,
1234
+ "loss": 0.4865,
1235
+ "step": 100500
1236
+ },
1237
+ {
1238
+ "epoch": 2.32,
1239
+ "learning_rate": 4.229672116062939e-05,
1240
+ "loss": 0.4823,
1241
+ "step": 101000
1242
+ },
1243
+ {
1244
+ "epoch": 2.33,
1245
+ "learning_rate": 4.225838113057081e-05,
1246
+ "loss": 0.4806,
1247
+ "step": 101500
1248
+ },
1249
+ {
1250
+ "epoch": 2.34,
1251
+ "learning_rate": 4.222011778057234e-05,
1252
+ "loss": 0.4813,
1253
+ "step": 102000
1254
+ },
1255
+ {
1256
+ "epoch": 2.36,
1257
+ "learning_rate": 4.218177775051376e-05,
1258
+ "loss": 0.4734,
1259
+ "step": 102500
1260
+ },
1261
+ {
1262
+ "epoch": 2.37,
1263
+ "learning_rate": 4.214343772045518e-05,
1264
+ "loss": 0.479,
1265
+ "step": 103000
1266
+ },
1267
+ {
1268
+ "epoch": 2.38,
1269
+ "learning_rate": 4.210509769039659e-05,
1270
+ "loss": 0.4778,
1271
+ "step": 103500
1272
+ },
1273
+ {
1274
+ "epoch": 2.39,
1275
+ "learning_rate": 4.2066757660338006e-05,
1276
+ "loss": 0.4797,
1277
+ "step": 104000
1278
+ },
1279
+ {
1280
+ "epoch": 2.4,
1281
+ "learning_rate": 4.2028417630279424e-05,
1282
+ "loss": 0.474,
1283
+ "step": 104500
1284
+ },
1285
+ {
1286
+ "epoch": 2.41,
1287
+ "learning_rate": 4.199007760022084e-05,
1288
+ "loss": 0.4811,
1289
+ "step": 105000
1290
+ },
1291
+ {
1292
+ "epoch": 2.43,
1293
+ "learning_rate": 4.1951814250222374e-05,
1294
+ "loss": 0.464,
1295
+ "step": 105500
1296
+ },
1297
+ {
1298
+ "epoch": 2.44,
1299
+ "learning_rate": 4.191347422016379e-05,
1300
+ "loss": 0.4699,
1301
+ "step": 106000
1302
+ },
1303
+ {
1304
+ "epoch": 2.45,
1305
+ "learning_rate": 4.187513419010521e-05,
1306
+ "loss": 0.4655,
1307
+ "step": 106500
1308
+ },
1309
+ {
1310
+ "epoch": 2.46,
1311
+ "learning_rate": 4.183679416004662e-05,
1312
+ "loss": 0.4694,
1313
+ "step": 107000
1314
+ },
1315
+ {
1316
+ "epoch": 2.47,
1317
+ "learning_rate": 4.179845412998804e-05,
1318
+ "loss": 0.4698,
1319
+ "step": 107500
1320
+ },
1321
+ {
1322
+ "epoch": 2.48,
1323
+ "learning_rate": 4.1760114099929456e-05,
1324
+ "loss": 0.4714,
1325
+ "step": 108000
1326
+ },
1327
+ {
1328
+ "epoch": 2.49,
1329
+ "learning_rate": 4.1721774069870873e-05,
1330
+ "loss": 0.4627,
1331
+ "step": 108500
1332
+ },
1333
+ {
1334
+ "epoch": 2.51,
1335
+ "learning_rate": 4.168343403981229e-05,
1336
+ "loss": 0.4628,
1337
+ "step": 109000
1338
+ },
1339
+ {
1340
+ "epoch": 2.52,
1341
+ "learning_rate": 4.16450940097537e-05,
1342
+ "loss": 0.4565,
1343
+ "step": 109500
1344
+ },
1345
+ {
1346
+ "epoch": 2.53,
1347
+ "learning_rate": 4.1606753979695126e-05,
1348
+ "loss": 0.466,
1349
+ "step": 110000
1350
+ },
1351
+ {
1352
+ "epoch": 2.54,
1353
+ "learning_rate": 4.156841394963654e-05,
1354
+ "loss": 0.4635,
1355
+ "step": 110500
1356
+ },
1357
+ {
1358
+ "epoch": 2.55,
1359
+ "learning_rate": 4.1530073919577955e-05,
1360
+ "loss": 0.4694,
1361
+ "step": 111000
1362
+ },
1363
+ {
1364
+ "epoch": 2.56,
1365
+ "learning_rate": 4.149181056957949e-05,
1366
+ "loss": 0.4638,
1367
+ "step": 111500
1368
+ },
1369
+ {
1370
+ "epoch": 2.57,
1371
+ "learning_rate": 4.1453470539520905e-05,
1372
+ "loss": 0.4626,
1373
+ "step": 112000
1374
+ },
1375
+ {
1376
+ "epoch": 2.59,
1377
+ "learning_rate": 4.1415130509462316e-05,
1378
+ "loss": 0.4632,
1379
+ "step": 112500
1380
+ },
1381
+ {
1382
+ "epoch": 2.6,
1383
+ "learning_rate": 4.137679047940374e-05,
1384
+ "loss": 0.455,
1385
+ "step": 113000
1386
+ },
1387
+ {
1388
+ "epoch": 2.61,
1389
+ "learning_rate": 4.133845044934516e-05,
1390
+ "loss": 0.4553,
1391
+ "step": 113500
1392
+ },
1393
+ {
1394
+ "epoch": 2.62,
1395
+ "learning_rate": 4.130018709934669e-05,
1396
+ "loss": 0.4583,
1397
+ "step": 114000
1398
+ },
1399
+ {
1400
+ "epoch": 2.63,
1401
+ "learning_rate": 4.126192374934822e-05,
1402
+ "loss": 0.4596,
1403
+ "step": 114500
1404
+ },
1405
+ {
1406
+ "epoch": 2.64,
1407
+ "learning_rate": 4.1223583719289635e-05,
1408
+ "loss": 0.4634,
1409
+ "step": 115000
1410
+ },
1411
+ {
1412
+ "epoch": 2.65,
1413
+ "learning_rate": 4.118524368923106e-05,
1414
+ "loss": 0.4595,
1415
+ "step": 115500
1416
+ },
1417
+ {
1418
+ "epoch": 2.67,
1419
+ "learning_rate": 4.1146980339232585e-05,
1420
+ "loss": 0.4441,
1421
+ "step": 116000
1422
+ },
1423
+ {
1424
+ "epoch": 2.68,
1425
+ "learning_rate": 4.110864030917401e-05,
1426
+ "loss": 0.4517,
1427
+ "step": 116500
1428
+ },
1429
+ {
1430
+ "epoch": 2.69,
1431
+ "learning_rate": 4.107030027911542e-05,
1432
+ "loss": 0.4531,
1433
+ "step": 117000
1434
+ },
1435
+ {
1436
+ "epoch": 2.7,
1437
+ "learning_rate": 4.103196024905684e-05,
1438
+ "loss": 0.4556,
1439
+ "step": 117500
1440
+ },
1441
+ {
1442
+ "epoch": 2.71,
1443
+ "learning_rate": 4.099362021899825e-05,
1444
+ "loss": 0.4481,
1445
+ "step": 118000
1446
+ },
1447
+ {
1448
+ "epoch": 2.72,
1449
+ "learning_rate": 4.095528018893967e-05,
1450
+ "loss": 0.4483,
1451
+ "step": 118500
1452
+ },
1453
+ {
1454
+ "epoch": 2.74,
1455
+ "learning_rate": 4.0916940158881084e-05,
1456
+ "loss": 0.4457,
1457
+ "step": 119000
1458
+ },
1459
+ {
1460
+ "epoch": 2.75,
1461
+ "learning_rate": 4.08786001288225e-05,
1462
+ "loss": 0.452,
1463
+ "step": 119500
1464
+ },
1465
+ {
1466
+ "epoch": 2.76,
1467
+ "learning_rate": 4.084026009876392e-05,
1468
+ "loss": 0.4511,
1469
+ "step": 120000
1470
+ },
1471
+ {
1472
+ "epoch": 2.77,
1473
+ "learning_rate": 4.080192006870534e-05,
1474
+ "loss": 0.4466,
1475
+ "step": 120500
1476
+ },
1477
+ {
1478
+ "epoch": 2.78,
1479
+ "learning_rate": 4.0763580038646755e-05,
1480
+ "loss": 0.4368,
1481
+ "step": 121000
1482
+ },
1483
+ {
1484
+ "epoch": 2.79,
1485
+ "learning_rate": 4.0725240008588166e-05,
1486
+ "loss": 0.443,
1487
+ "step": 121500
1488
+ },
1489
+ {
1490
+ "epoch": 2.8,
1491
+ "learning_rate": 4.0686976658589705e-05,
1492
+ "loss": 0.441,
1493
+ "step": 122000
1494
+ },
1495
+ {
1496
+ "epoch": 2.82,
1497
+ "learning_rate": 4.0648636628531116e-05,
1498
+ "loss": 0.4503,
1499
+ "step": 122500
1500
+ },
1501
+ {
1502
+ "epoch": 2.83,
1503
+ "learning_rate": 4.0610296598472534e-05,
1504
+ "loss": 0.4487,
1505
+ "step": 123000
1506
+ },
1507
+ {
1508
+ "epoch": 2.84,
1509
+ "learning_rate": 4.057195656841395e-05,
1510
+ "loss": 0.4482,
1511
+ "step": 123500
1512
+ },
1513
+ {
1514
+ "epoch": 2.85,
1515
+ "learning_rate": 4.053361653835537e-05,
1516
+ "loss": 0.4449,
1517
+ "step": 124000
1518
+ },
1519
+ {
1520
+ "epoch": 2.86,
1521
+ "learning_rate": 4.049527650829679e-05,
1522
+ "loss": 0.4514,
1523
+ "step": 124500
1524
+ },
1525
+ {
1526
+ "epoch": 2.87,
1527
+ "learning_rate": 4.04569364782382e-05,
1528
+ "loss": 0.4407,
1529
+ "step": 125000
1530
+ },
1531
+ {
1532
+ "epoch": 2.88,
1533
+ "learning_rate": 4.041867312823974e-05,
1534
+ "loss": 0.443,
1535
+ "step": 125500
1536
+ },
1537
+ {
1538
+ "epoch": 2.9,
1539
+ "learning_rate": 4.038033309818115e-05,
1540
+ "loss": 0.434,
1541
+ "step": 126000
1542
+ },
1543
+ {
1544
+ "epoch": 2.91,
1545
+ "learning_rate": 4.0341993068122566e-05,
1546
+ "loss": 0.4384,
1547
+ "step": 126500
1548
+ },
1549
+ {
1550
+ "epoch": 2.92,
1551
+ "learning_rate": 4.030365303806398e-05,
1552
+ "loss": 0.4334,
1553
+ "step": 127000
1554
+ },
1555
+ {
1556
+ "epoch": 2.93,
1557
+ "learning_rate": 4.0265389688065516e-05,
1558
+ "loss": 0.4419,
1559
+ "step": 127500
1560
+ },
1561
+ {
1562
+ "epoch": 2.94,
1563
+ "learning_rate": 4.0227049658006934e-05,
1564
+ "loss": 0.4421,
1565
+ "step": 128000
1566
+ },
1567
+ {
1568
+ "epoch": 2.95,
1569
+ "learning_rate": 4.018870962794835e-05,
1570
+ "loss": 0.4244,
1571
+ "step": 128500
1572
+ },
1573
+ {
1574
+ "epoch": 2.97,
1575
+ "learning_rate": 4.015036959788976e-05,
1576
+ "loss": 0.4345,
1577
+ "step": 129000
1578
+ },
1579
+ {
1580
+ "epoch": 2.98,
1581
+ "learning_rate": 4.01121062478913e-05,
1582
+ "loss": 0.4374,
1583
+ "step": 129500
1584
+ },
1585
+ {
1586
+ "epoch": 2.99,
1587
+ "learning_rate": 4.007376621783271e-05,
1588
+ "loss": 0.4368,
1589
+ "step": 130000
1590
+ },
1591
+ {
1592
+ "epoch": 3.0,
1593
+ "learning_rate": 4.003542618777413e-05,
1594
+ "loss": 0.4356,
1595
+ "step": 130500
1596
+ },
1597
+ {
1598
+ "epoch": 3.0,
1599
+ "eval_bleu": 58.5493,
1600
+ "eval_gen_len": 17.1569,
1601
+ "eval_loss": 0.7014300227165222,
1602
+ "eval_runtime": 7450.9373,
1603
+ "eval_samples_per_second": 11.677,
1604
+ "eval_steps_per_second": 1.46,
1605
+ "step": 130513
1606
+ },
1607
+ {
1608
+ "epoch": 3.01,
1609
+ "learning_rate": 3.9997086157715555e-05,
1610
+ "loss": 0.4373,
1611
+ "step": 131000
1612
+ },
1613
+ {
1614
+ "epoch": 3.02,
1615
+ "learning_rate": 3.9958746127656966e-05,
1616
+ "loss": 0.42,
1617
+ "step": 131500
1618
+ },
1619
+ {
1620
+ "epoch": 3.03,
1621
+ "learning_rate": 3.992040609759838e-05,
1622
+ "loss": 0.4179,
1623
+ "step": 132000
1624
+ },
1625
+ {
1626
+ "epoch": 3.05,
1627
+ "learning_rate": 3.9882066067539794e-05,
1628
+ "loss": 0.406,
1629
+ "step": 132500
1630
+ },
1631
+ {
1632
+ "epoch": 3.06,
1633
+ "learning_rate": 3.984372603748122e-05,
1634
+ "loss": 0.4034,
1635
+ "step": 133000
1636
+ },
1637
+ {
1638
+ "epoch": 3.07,
1639
+ "learning_rate": 3.980538600742263e-05,
1640
+ "loss": 0.4103,
1641
+ "step": 133500
1642
+ },
1643
+ {
1644
+ "epoch": 3.08,
1645
+ "learning_rate": 3.976704597736405e-05,
1646
+ "loss": 0.4084,
1647
+ "step": 134000
1648
+ },
1649
+ {
1650
+ "epoch": 3.09,
1651
+ "learning_rate": 3.9728705947305465e-05,
1652
+ "loss": 0.4012,
1653
+ "step": 134500
1654
+ },
1655
+ {
1656
+ "epoch": 3.1,
1657
+ "learning_rate": 3.969036591724688e-05,
1658
+ "loss": 0.3966,
1659
+ "step": 135000
1660
+ },
1661
+ {
1662
+ "epoch": 3.11,
1663
+ "learning_rate": 3.96520258871883e-05,
1664
+ "loss": 0.4026,
1665
+ "step": 135500
1666
+ },
1667
+ {
1668
+ "epoch": 3.13,
1669
+ "learning_rate": 3.961368585712971e-05,
1670
+ "loss": 0.3983,
1671
+ "step": 136000
1672
+ },
1673
+ {
1674
+ "epoch": 3.14,
1675
+ "learning_rate": 3.957542250713125e-05,
1676
+ "loss": 0.3947,
1677
+ "step": 136500
1678
+ },
1679
+ {
1680
+ "epoch": 3.15,
1681
+ "learning_rate": 3.953708247707266e-05,
1682
+ "loss": 0.3894,
1683
+ "step": 137000
1684
+ },
1685
+ {
1686
+ "epoch": 3.16,
1687
+ "learning_rate": 3.94988191270742e-05,
1688
+ "loss": 0.3927,
1689
+ "step": 137500
1690
+ },
1691
+ {
1692
+ "epoch": 3.17,
1693
+ "learning_rate": 3.946047909701561e-05,
1694
+ "loss": 0.3919,
1695
+ "step": 138000
1696
+ },
1697
+ {
1698
+ "epoch": 3.18,
1699
+ "learning_rate": 3.942213906695703e-05,
1700
+ "loss": 0.3909,
1701
+ "step": 138500
1702
+ },
1703
+ {
1704
+ "epoch": 3.2,
1705
+ "learning_rate": 3.938379903689845e-05,
1706
+ "loss": 0.3914,
1707
+ "step": 139000
1708
+ },
1709
+ {
1710
+ "epoch": 3.21,
1711
+ "learning_rate": 3.9345459006839865e-05,
1712
+ "loss": 0.3876,
1713
+ "step": 139500
1714
+ },
1715
+ {
1716
+ "epoch": 3.22,
1717
+ "learning_rate": 3.930727233690151e-05,
1718
+ "loss": 0.3844,
1719
+ "step": 140000
1720
+ },
1721
+ {
1722
+ "epoch": 3.23,
1723
+ "learning_rate": 3.926893230684293e-05,
1724
+ "loss": 0.3871,
1725
+ "step": 140500
1726
+ },
1727
+ {
1728
+ "epoch": 3.24,
1729
+ "learning_rate": 3.923059227678435e-05,
1730
+ "loss": 0.3835,
1731
+ "step": 141000
1732
+ },
1733
+ {
1734
+ "epoch": 3.25,
1735
+ "learning_rate": 3.9192252246725766e-05,
1736
+ "loss": 0.3871,
1737
+ "step": 141500
1738
+ },
1739
+ {
1740
+ "epoch": 3.26,
1741
+ "learning_rate": 3.915391221666718e-05,
1742
+ "loss": 0.3803,
1743
+ "step": 142000
1744
+ },
1745
+ {
1746
+ "epoch": 3.28,
1747
+ "learning_rate": 3.9115572186608594e-05,
1748
+ "loss": 0.3776,
1749
+ "step": 142500
1750
+ },
1751
+ {
1752
+ "epoch": 3.29,
1753
+ "learning_rate": 3.907723215655001e-05,
1754
+ "loss": 0.3742,
1755
+ "step": 143000
1756
+ },
1757
+ {
1758
+ "epoch": 3.3,
1759
+ "learning_rate": 3.903889212649143e-05,
1760
+ "loss": 0.3778,
1761
+ "step": 143500
1762
+ },
1763
+ {
1764
+ "epoch": 3.31,
1765
+ "learning_rate": 3.900055209643285e-05,
1766
+ "loss": 0.3759,
1767
+ "step": 144000
1768
+ },
1769
+ {
1770
+ "epoch": 3.32,
1771
+ "learning_rate": 3.896221206637426e-05,
1772
+ "loss": 0.3756,
1773
+ "step": 144500
1774
+ },
1775
+ {
1776
+ "epoch": 3.33,
1777
+ "learning_rate": 3.89239487163758e-05,
1778
+ "loss": 0.3737,
1779
+ "step": 145000
1780
+ },
1781
+ {
1782
+ "epoch": 3.34,
1783
+ "learning_rate": 3.888560868631721e-05,
1784
+ "loss": 0.3749,
1785
+ "step": 145500
1786
+ },
1787
+ {
1788
+ "epoch": 3.36,
1789
+ "learning_rate": 3.884734533631875e-05,
1790
+ "loss": 0.3701,
1791
+ "step": 146000
1792
+ },
1793
+ {
1794
+ "epoch": 3.37,
1795
+ "learning_rate": 3.880900530626016e-05,
1796
+ "loss": 0.3736,
1797
+ "step": 146500
1798
+ },
1799
+ {
1800
+ "epoch": 3.38,
1801
+ "learning_rate": 3.8770665276201576e-05,
1802
+ "loss": 0.3706,
1803
+ "step": 147000
1804
+ },
1805
+ {
1806
+ "epoch": 3.39,
1807
+ "learning_rate": 3.8732325246143e-05,
1808
+ "loss": 0.3743,
1809
+ "step": 147500
1810
+ },
1811
+ {
1812
+ "epoch": 3.4,
1813
+ "learning_rate": 3.869398521608441e-05,
1814
+ "loss": 0.3665,
1815
+ "step": 148000
1816
+ },
1817
+ {
1818
+ "epoch": 3.41,
1819
+ "learning_rate": 3.865564518602583e-05,
1820
+ "loss": 0.3781,
1821
+ "step": 148500
1822
+ },
1823
+ {
1824
+ "epoch": 3.42,
1825
+ "learning_rate": 3.861738183602736e-05,
1826
+ "loss": 0.3591,
1827
+ "step": 149000
1828
+ },
1829
+ {
1830
+ "epoch": 3.44,
1831
+ "learning_rate": 3.857904180596878e-05,
1832
+ "loss": 0.365,
1833
+ "step": 149500
1834
+ },
1835
+ {
1836
+ "epoch": 3.45,
1837
+ "learning_rate": 3.854070177591019e-05,
1838
+ "loss": 0.3648,
1839
+ "step": 150000
1840
+ },
1841
+ {
1842
+ "epoch": 3.46,
1843
+ "learning_rate": 3.8502361745851615e-05,
1844
+ "loss": 0.3635,
1845
+ "step": 150500
1846
+ },
1847
+ {
1848
+ "epoch": 3.47,
1849
+ "learning_rate": 3.8464021715793026e-05,
1850
+ "loss": 0.3658,
1851
+ "step": 151000
1852
+ },
1853
+ {
1854
+ "epoch": 3.48,
1855
+ "learning_rate": 3.8425681685734444e-05,
1856
+ "loss": 0.3666,
1857
+ "step": 151500
1858
+ },
1859
+ {
1860
+ "epoch": 3.49,
1861
+ "learning_rate": 3.838734165567586e-05,
1862
+ "loss": 0.3579,
1863
+ "step": 152000
1864
+ },
1865
+ {
1866
+ "epoch": 3.51,
1867
+ "learning_rate": 3.834900162561728e-05,
1868
+ "loss": 0.3582,
1869
+ "step": 152500
1870
+ },
1871
+ {
1872
+ "epoch": 3.52,
1873
+ "learning_rate": 3.8310661595558697e-05,
1874
+ "loss": 0.3554,
1875
+ "step": 153000
1876
+ },
1877
+ {
1878
+ "epoch": 3.53,
1879
+ "learning_rate": 3.8272474925620344e-05,
1880
+ "loss": 0.3645,
1881
+ "step": 153500
1882
+ },
1883
+ {
1884
+ "epoch": 3.54,
1885
+ "learning_rate": 3.823421157562188e-05,
1886
+ "loss": 0.3605,
1887
+ "step": 154000
1888
+ },
1889
+ {
1890
+ "epoch": 3.55,
1891
+ "learning_rate": 3.8195871545563295e-05,
1892
+ "loss": 0.3657,
1893
+ "step": 154500
1894
+ },
1895
+ {
1896
+ "epoch": 3.56,
1897
+ "learning_rate": 3.8157531515504706e-05,
1898
+ "loss": 0.362,
1899
+ "step": 155000
1900
+ },
1901
+ {
1902
+ "epoch": 3.57,
1903
+ "learning_rate": 3.811919148544612e-05,
1904
+ "loss": 0.3585,
1905
+ "step": 155500
1906
+ },
1907
+ {
1908
+ "epoch": 3.59,
1909
+ "learning_rate": 3.808085145538755e-05,
1910
+ "loss": 0.3606,
1911
+ "step": 156000
1912
+ },
1913
+ {
1914
+ "epoch": 3.6,
1915
+ "learning_rate": 3.8042664785449196e-05,
1916
+ "loss": 0.3539,
1917
+ "step": 156500
1918
+ },
1919
+ {
1920
+ "epoch": 3.61,
1921
+ "learning_rate": 3.8004324755390606e-05,
1922
+ "loss": 0.3543,
1923
+ "step": 157000
1924
+ },
1925
+ {
1926
+ "epoch": 3.62,
1927
+ "learning_rate": 3.7966061405392146e-05,
1928
+ "loss": 0.3599,
1929
+ "step": 157500
1930
+ },
1931
+ {
1932
+ "epoch": 3.63,
1933
+ "learning_rate": 3.792772137533356e-05,
1934
+ "loss": 0.3613,
1935
+ "step": 158000
1936
+ },
1937
+ {
1938
+ "epoch": 3.64,
1939
+ "learning_rate": 3.7889381345274975e-05,
1940
+ "loss": 0.3616,
1941
+ "step": 158500
1942
+ },
1943
+ {
1944
+ "epoch": 3.65,
1945
+ "learning_rate": 3.78510413152164e-05,
1946
+ "loss": 0.357,
1947
+ "step": 159000
1948
+ },
1949
+ {
1950
+ "epoch": 3.67,
1951
+ "learning_rate": 3.781270128515781e-05,
1952
+ "loss": 0.3444,
1953
+ "step": 159500
1954
+ },
1955
+ {
1956
+ "epoch": 3.68,
1957
+ "learning_rate": 3.777436125509923e-05,
1958
+ "loss": 0.35,
1959
+ "step": 160000
1960
+ },
1961
+ {
1962
+ "epoch": 3.69,
1963
+ "learning_rate": 3.773602122504064e-05,
1964
+ "loss": 0.3515,
1965
+ "step": 160500
1966
+ },
1967
+ {
1968
+ "epoch": 3.7,
1969
+ "learning_rate": 3.769768119498206e-05,
1970
+ "loss": 0.3571,
1971
+ "step": 161000
1972
+ },
1973
+ {
1974
+ "epoch": 3.71,
1975
+ "learning_rate": 3.7659341164923474e-05,
1976
+ "loss": 0.3484,
1977
+ "step": 161500
1978
+ },
1979
+ {
1980
+ "epoch": 3.72,
1981
+ "learning_rate": 3.762100113486489e-05,
1982
+ "loss": 0.3467,
1983
+ "step": 162000
1984
+ },
1985
+ {
1986
+ "epoch": 3.74,
1987
+ "learning_rate": 3.758266110480631e-05,
1988
+ "loss": 0.3489,
1989
+ "step": 162500
1990
+ },
1991
+ {
1992
+ "epoch": 3.75,
1993
+ "learning_rate": 3.754432107474773e-05,
1994
+ "loss": 0.3505,
1995
+ "step": 163000
1996
+ },
1997
+ {
1998
+ "epoch": 3.76,
1999
+ "learning_rate": 3.7505981044689144e-05,
2000
+ "loss": 0.351,
2001
+ "step": 163500
2002
+ },
2003
+ {
2004
+ "epoch": 3.77,
2005
+ "learning_rate": 3.7467641014630555e-05,
2006
+ "loss": 0.3486,
2007
+ "step": 164000
2008
+ },
2009
+ {
2010
+ "epoch": 3.78,
2011
+ "learning_rate": 3.7429377664632095e-05,
2012
+ "loss": 0.3394,
2013
+ "step": 164500
2014
+ },
2015
+ {
2016
+ "epoch": 3.79,
2017
+ "learning_rate": 3.7391037634573506e-05,
2018
+ "loss": 0.3445,
2019
+ "step": 165000
2020
+ },
2021
+ {
2022
+ "epoch": 3.8,
2023
+ "learning_rate": 3.735269760451492e-05,
2024
+ "loss": 0.3417,
2025
+ "step": 165500
2026
+ },
2027
+ {
2028
+ "epoch": 3.82,
2029
+ "learning_rate": 3.731435757445634e-05,
2030
+ "loss": 0.3492,
2031
+ "step": 166000
2032
+ },
2033
+ {
2034
+ "epoch": 3.83,
2035
+ "learning_rate": 3.727601754439776e-05,
2036
+ "loss": 0.3519,
2037
+ "step": 166500
2038
+ },
2039
+ {
2040
+ "epoch": 3.84,
2041
+ "learning_rate": 3.7237677514339176e-05,
2042
+ "loss": 0.3484,
2043
+ "step": 167000
2044
+ },
2045
+ {
2046
+ "epoch": 3.85,
2047
+ "learning_rate": 3.719933748428059e-05,
2048
+ "loss": 0.3446,
2049
+ "step": 167500
2050
+ },
2051
+ {
2052
+ "epoch": 3.86,
2053
+ "learning_rate": 3.7160997454222005e-05,
2054
+ "loss": 0.3499,
2055
+ "step": 168000
2056
+ },
2057
+ {
2058
+ "epoch": 3.87,
2059
+ "learning_rate": 3.712265742416342e-05,
2060
+ "loss": 0.3465,
2061
+ "step": 168500
2062
+ },
2063
+ {
2064
+ "epoch": 3.88,
2065
+ "learning_rate": 3.708431739410484e-05,
2066
+ "loss": 0.3431,
2067
+ "step": 169000
2068
+ },
2069
+ {
2070
+ "epoch": 3.9,
2071
+ "learning_rate": 3.704597736404625e-05,
2072
+ "loss": 0.3379,
2073
+ "step": 169500
2074
+ },
2075
+ {
2076
+ "epoch": 3.91,
2077
+ "learning_rate": 3.700771401404779e-05,
2078
+ "loss": 0.3402,
2079
+ "step": 170000
2080
+ },
2081
+ {
2082
+ "epoch": 3.92,
2083
+ "learning_rate": 3.69693739839892e-05,
2084
+ "loss": 0.3379,
2085
+ "step": 170500
2086
+ },
2087
+ {
2088
+ "epoch": 3.93,
2089
+ "learning_rate": 3.6931187314050856e-05,
2090
+ "loss": 0.3468,
2091
+ "step": 171000
2092
+ },
2093
+ {
2094
+ "epoch": 3.94,
2095
+ "learning_rate": 3.6892847283992274e-05,
2096
+ "loss": 0.3433,
2097
+ "step": 171500
2098
+ },
2099
+ {
2100
+ "epoch": 3.95,
2101
+ "learning_rate": 3.685450725393369e-05,
2102
+ "loss": 0.3305,
2103
+ "step": 172000
2104
+ },
2105
+ {
2106
+ "epoch": 3.97,
2107
+ "learning_rate": 3.68161672238751e-05,
2108
+ "loss": 0.3373,
2109
+ "step": 172500
2110
+ },
2111
+ {
2112
+ "epoch": 3.98,
2113
+ "learning_rate": 3.677782719381652e-05,
2114
+ "loss": 0.3426,
2115
+ "step": 173000
2116
+ },
2117
+ {
2118
+ "epoch": 3.99,
2119
+ "learning_rate": 3.673948716375794e-05,
2120
+ "loss": 0.3387,
2121
+ "step": 173500
2122
+ },
2123
+ {
2124
+ "epoch": 4.0,
2125
+ "learning_rate": 3.6701147133699355e-05,
2126
+ "loss": 0.3407,
2127
+ "step": 174000
2128
+ },
2129
+ {
2130
+ "epoch": 4.0,
2131
+ "eval_bleu": 58.3826,
2132
+ "eval_gen_len": 17.1623,
2133
+ "eval_loss": 0.7513108253479004,
2134
+ "eval_runtime": 7458.9074,
2135
+ "eval_samples_per_second": 11.665,
2136
+ "eval_steps_per_second": 1.458,
2137
+ "step": 174018
2138
+ }
2139
+ ],
2140
+ "logging_steps": 500,
2141
+ "max_steps": 652560,
2142
+ "num_train_epochs": 15,
2143
+ "save_steps": 500,
2144
+ "total_flos": 6.033913407575425e+18,
2145
+ "trial_name": null,
2146
+ "trial_params": null
2147
+ }
checkpoint-174018/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44b731ecd386cb51a958b6c8f191496c616dc6e80b60cf4e62402772da333973
3
+ size 4792
checkpoint-217522/config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
3
+ "_num_labels": 3,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "relu",
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": 2,
21
+ "dropout": 0.1,
22
+ "early_stopping": true,
23
+ "encoder_attention_heads": 16,
24
+ "encoder_ffn_dim": 4096,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 12,
27
+ "eos_token_id": 2,
28
+ "forced_bos_token_id": 250014,
29
+ "forced_eos_token_id": 2,
30
+ "gradient_checkpointing": false,
31
+ "id2label": {
32
+ "0": "LABEL_0",
33
+ "1": "LABEL_1",
34
+ "2": "LABEL_2"
35
+ },
36
+ "init_std": 0.02,
37
+ "is_encoder_decoder": true,
38
+ "label2id": {
39
+ "LABEL_0": 0,
40
+ "LABEL_1": 1,
41
+ "LABEL_2": 2
42
+ },
43
+ "max_length": 200,
44
+ "max_position_embeddings": 1024,
45
+ "model_type": "mbart",
46
+ "normalize_before": true,
47
+ "normalize_embedding": true,
48
+ "num_beams": 5,
49
+ "num_hidden_layers": 12,
50
+ "output_past": true,
51
+ "pad_token_id": 1,
52
+ "scale_embedding": true,
53
+ "static_position_embeddings": false,
54
+ "tokenizer_class": "MBart50Tokenizer",
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.35.2",
57
+ "use_cache": true,
58
+ "vocab_size": 250054
59
+ }
checkpoint-217522/generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 2,
4
+ "early_stopping": true,
5
+ "eos_token_id": 2,
6
+ "forced_bos_token_id": 250014,
7
+ "forced_eos_token_id": 2,
8
+ "max_length": 200,
9
+ "num_beams": 5,
10
+ "pad_token_id": 1,
11
+ "transformers_version": "4.35.2"
12
+ }
checkpoint-217522/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f483628a61a16b96d004b6c15d7a8f0196b924e9da08ab70634354fc3e31bbd
3
+ size 2444578688
checkpoint-217522/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2cc05c66900a46905498e552a47d88fbf72f824f694a3ac142b78c47a632fc7e
3
+ size 4887473903
checkpoint-217522/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0e20bd68a94dfb12a00842c4621b9aab9cb01297bcf5dfa75acaba869daa6ce4
3
+ size 15024
checkpoint-217522/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d91e7f7025b35efbebc869df10cc4e03d5923936d61c9d83522f9ae760cd35ba
3
+ size 15024
checkpoint-217522/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6d42a837bd8226393744dd25070e85314f195de1485948f0b41179b2b638de1
3
+ size 15024
checkpoint-217522/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5503ecc7d239f16aee50cab50682649946e62d726975c770d4a0259e0e5f99a7
3
+ size 15024
checkpoint-217522/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a003699053bdedc83a81d5fd5df4cbde1d80ee6dd5c97a23605d18cea26d06b9
3
+ size 1064
checkpoint-217522/sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
checkpoint-217522/special_tokens_map.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "ar_AR",
4
+ "cs_CZ",
5
+ "de_DE",
6
+ "en_XX",
7
+ "es_XX",
8
+ "et_EE",
9
+ "fi_FI",
10
+ "fr_XX",
11
+ "gu_IN",
12
+ "hi_IN",
13
+ "it_IT",
14
+ "ja_XX",
15
+ "kk_KZ",
16
+ "ko_KR",
17
+ "lt_LT",
18
+ "lv_LV",
19
+ "my_MM",
20
+ "ne_NP",
21
+ "nl_XX",
22
+ "ro_RO",
23
+ "ru_RU",
24
+ "si_LK",
25
+ "tr_TR",
26
+ "vi_VN",
27
+ "zh_CN",
28
+ "af_ZA",
29
+ "az_AZ",
30
+ "bn_IN",
31
+ "fa_IR",
32
+ "he_IL",
33
+ "hr_HR",
34
+ "id_ID",
35
+ "ka_GE",
36
+ "km_KH",
37
+ "mk_MK",
38
+ "ml_IN",
39
+ "mn_MN",
40
+ "mr_IN",
41
+ "pl_PL",
42
+ "ps_AF",
43
+ "pt_XX",
44
+ "sv_SE",
45
+ "sw_KE",
46
+ "ta_IN",
47
+ "te_IN",
48
+ "th_TH",
49
+ "tl_XX",
50
+ "uk_UA",
51
+ "ur_PK",
52
+ "xh_ZA",
53
+ "gl_ES",
54
+ "sl_SI"
55
+ ],
56
+ "bos_token": "<s>",
57
+ "cls_token": "<s>",
58
+ "eos_token": "</s>",
59
+ "mask_token": {
60
+ "content": "<mask>",
61
+ "lstrip": true,
62
+ "normalized": true,
63
+ "rstrip": false,
64
+ "single_word": false
65
+ },
66
+ "pad_token": "<pad>",
67
+ "sep_token": "</s>",
68
+ "unk_token": "<unk>"
69
+ }
checkpoint-217522/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:faf1fb0f45fdd5ae9fc44fafee492652a423fc749833ef57c2fa801358549d92
3
+ size 17110107
checkpoint-217522/tokenizer_config.json ADDED
@@ -0,0 +1,528 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "250001": {
36
+ "content": "ar_AR",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "250002": {
44
+ "content": "cs_CZ",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "250003": {
52
+ "content": "de_DE",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "250004": {
60
+ "content": "en_XX",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": true
66
+ },
67
+ "250005": {
68
+ "content": "es_XX",
69
+ "lstrip": false,
70
+ "normalized": false,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": true
74
+ },
75
+ "250006": {
76
+ "content": "et_EE",
77
+ "lstrip": false,
78
+ "normalized": false,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": true
82
+ },
83
+ "250007": {
84
+ "content": "fi_FI",
85
+ "lstrip": false,
86
+ "normalized": false,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": true
90
+ },
91
+ "250008": {
92
+ "content": "fr_XX",
93
+ "lstrip": false,
94
+ "normalized": false,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": true
98
+ },
99
+ "250009": {
100
+ "content": "gu_IN",
101
+ "lstrip": false,
102
+ "normalized": false,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": true
106
+ },
107
+ "250010": {
108
+ "content": "hi_IN",
109
+ "lstrip": false,
110
+ "normalized": false,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": true
114
+ },
115
+ "250011": {
116
+ "content": "it_IT",
117
+ "lstrip": false,
118
+ "normalized": false,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": true
122
+ },
123
+ "250012": {
124
+ "content": "ja_XX",
125
+ "lstrip": false,
126
+ "normalized": false,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": true
130
+ },
131
+ "250013": {
132
+ "content": "kk_KZ",
133
+ "lstrip": false,
134
+ "normalized": false,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": true
138
+ },
139
+ "250014": {
140
+ "content": "ko_KR",
141
+ "lstrip": false,
142
+ "normalized": false,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": true
146
+ },
147
+ "250015": {
148
+ "content": "lt_LT",
149
+ "lstrip": false,
150
+ "normalized": false,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": true
154
+ },
155
+ "250016": {
156
+ "content": "lv_LV",
157
+ "lstrip": false,
158
+ "normalized": false,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": true
162
+ },
163
+ "250017": {
164
+ "content": "my_MM",
165
+ "lstrip": false,
166
+ "normalized": false,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": true
170
+ },
171
+ "250018": {
172
+ "content": "ne_NP",
173
+ "lstrip": false,
174
+ "normalized": false,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": true
178
+ },
179
+ "250019": {
180
+ "content": "nl_XX",
181
+ "lstrip": false,
182
+ "normalized": false,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": true
186
+ },
187
+ "250020": {
188
+ "content": "ro_RO",
189
+ "lstrip": false,
190
+ "normalized": false,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": true
194
+ },
195
+ "250021": {
196
+ "content": "ru_RU",
197
+ "lstrip": false,
198
+ "normalized": false,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": true
202
+ },
203
+ "250022": {
204
+ "content": "si_LK",
205
+ "lstrip": false,
206
+ "normalized": false,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": true
210
+ },
211
+ "250023": {
212
+ "content": "tr_TR",
213
+ "lstrip": false,
214
+ "normalized": false,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": true
218
+ },
219
+ "250024": {
220
+ "content": "vi_VN",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "250025": {
228
+ "content": "zh_CN",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "250026": {
236
+ "content": "af_ZA",
237
+ "lstrip": false,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "250027": {
244
+ "content": "az_AZ",
245
+ "lstrip": false,
246
+ "normalized": false,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "250028": {
252
+ "content": "bn_IN",
253
+ "lstrip": false,
254
+ "normalized": false,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "250029": {
260
+ "content": "fa_IR",
261
+ "lstrip": false,
262
+ "normalized": false,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "250030": {
268
+ "content": "he_IL",
269
+ "lstrip": false,
270
+ "normalized": false,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": true
274
+ },
275
+ "250031": {
276
+ "content": "hr_HR",
277
+ "lstrip": false,
278
+ "normalized": false,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": true
282
+ },
283
+ "250032": {
284
+ "content": "id_ID",
285
+ "lstrip": false,
286
+ "normalized": false,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": true
290
+ },
291
+ "250033": {
292
+ "content": "ka_GE",
293
+ "lstrip": false,
294
+ "normalized": false,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": true
298
+ },
299
+ "250034": {
300
+ "content": "km_KH",
301
+ "lstrip": false,
302
+ "normalized": false,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": true
306
+ },
307
+ "250035": {
308
+ "content": "mk_MK",
309
+ "lstrip": false,
310
+ "normalized": false,
311
+ "rstrip": false,
312
+ "single_word": false,
313
+ "special": true
314
+ },
315
+ "250036": {
316
+ "content": "ml_IN",
317
+ "lstrip": false,
318
+ "normalized": false,
319
+ "rstrip": false,
320
+ "single_word": false,
321
+ "special": true
322
+ },
323
+ "250037": {
324
+ "content": "mn_MN",
325
+ "lstrip": false,
326
+ "normalized": false,
327
+ "rstrip": false,
328
+ "single_word": false,
329
+ "special": true
330
+ },
331
+ "250038": {
332
+ "content": "mr_IN",
333
+ "lstrip": false,
334
+ "normalized": false,
335
+ "rstrip": false,
336
+ "single_word": false,
337
+ "special": true
338
+ },
339
+ "250039": {
340
+ "content": "pl_PL",
341
+ "lstrip": false,
342
+ "normalized": false,
343
+ "rstrip": false,
344
+ "single_word": false,
345
+ "special": true
346
+ },
347
+ "250040": {
348
+ "content": "ps_AF",
349
+ "lstrip": false,
350
+ "normalized": false,
351
+ "rstrip": false,
352
+ "single_word": false,
353
+ "special": true
354
+ },
355
+ "250041": {
356
+ "content": "pt_XX",
357
+ "lstrip": false,
358
+ "normalized": false,
359
+ "rstrip": false,
360
+ "single_word": false,
361
+ "special": true
362
+ },
363
+ "250042": {
364
+ "content": "sv_SE",
365
+ "lstrip": false,
366
+ "normalized": false,
367
+ "rstrip": false,
368
+ "single_word": false,
369
+ "special": true
370
+ },
371
+ "250043": {
372
+ "content": "sw_KE",
373
+ "lstrip": false,
374
+ "normalized": false,
375
+ "rstrip": false,
376
+ "single_word": false,
377
+ "special": true
378
+ },
379
+ "250044": {
380
+ "content": "ta_IN",
381
+ "lstrip": false,
382
+ "normalized": false,
383
+ "rstrip": false,
384
+ "single_word": false,
385
+ "special": true
386
+ },
387
+ "250045": {
388
+ "content": "te_IN",
389
+ "lstrip": false,
390
+ "normalized": false,
391
+ "rstrip": false,
392
+ "single_word": false,
393
+ "special": true
394
+ },
395
+ "250046": {
396
+ "content": "th_TH",
397
+ "lstrip": false,
398
+ "normalized": false,
399
+ "rstrip": false,
400
+ "single_word": false,
401
+ "special": true
402
+ },
403
+ "250047": {
404
+ "content": "tl_XX",
405
+ "lstrip": false,
406
+ "normalized": false,
407
+ "rstrip": false,
408
+ "single_word": false,
409
+ "special": true
410
+ },
411
+ "250048": {
412
+ "content": "uk_UA",
413
+ "lstrip": false,
414
+ "normalized": false,
415
+ "rstrip": false,
416
+ "single_word": false,
417
+ "special": true
418
+ },
419
+ "250049": {
420
+ "content": "ur_PK",
421
+ "lstrip": false,
422
+ "normalized": false,
423
+ "rstrip": false,
424
+ "single_word": false,
425
+ "special": true
426
+ },
427
+ "250050": {
428
+ "content": "xh_ZA",
429
+ "lstrip": false,
430
+ "normalized": false,
431
+ "rstrip": false,
432
+ "single_word": false,
433
+ "special": true
434
+ },
435
+ "250051": {
436
+ "content": "gl_ES",
437
+ "lstrip": false,
438
+ "normalized": false,
439
+ "rstrip": false,
440
+ "single_word": false,
441
+ "special": true
442
+ },
443
+ "250052": {
444
+ "content": "sl_SI",
445
+ "lstrip": false,
446
+ "normalized": false,
447
+ "rstrip": false,
448
+ "single_word": false,
449
+ "special": true
450
+ },
451
+ "250053": {
452
+ "content": "<mask>",
453
+ "lstrip": true,
454
+ "normalized": true,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": true
458
+ }
459
+ },
460
+ "additional_special_tokens": [
461
+ "ar_AR",
462
+ "cs_CZ",
463
+ "de_DE",
464
+ "en_XX",
465
+ "es_XX",
466
+ "et_EE",
467
+ "fi_FI",
468
+ "fr_XX",
469
+ "gu_IN",
470
+ "hi_IN",
471
+ "it_IT",
472
+ "ja_XX",
473
+ "kk_KZ",
474
+ "ko_KR",
475
+ "lt_LT",
476
+ "lv_LV",
477
+ "my_MM",
478
+ "ne_NP",
479
+ "nl_XX",
480
+ "ro_RO",
481
+ "ru_RU",
482
+ "si_LK",
483
+ "tr_TR",
484
+ "vi_VN",
485
+ "zh_CN",
486
+ "af_ZA",
487
+ "az_AZ",
488
+ "bn_IN",
489
+ "fa_IR",
490
+ "he_IL",
491
+ "hr_HR",
492
+ "id_ID",
493
+ "ka_GE",
494
+ "km_KH",
495
+ "mk_MK",
496
+ "ml_IN",
497
+ "mn_MN",
498
+ "mr_IN",
499
+ "pl_PL",
500
+ "ps_AF",
501
+ "pt_XX",
502
+ "sv_SE",
503
+ "sw_KE",
504
+ "ta_IN",
505
+ "te_IN",
506
+ "th_TH",
507
+ "tl_XX",
508
+ "uk_UA",
509
+ "ur_PK",
510
+ "xh_ZA",
511
+ "gl_ES",
512
+ "sl_SI"
513
+ ],
514
+ "bos_token": "<s>",
515
+ "clean_up_tokenization_spaces": true,
516
+ "cls_token": "<s>",
517
+ "eos_token": "</s>",
518
+ "language_codes": "ML50",
519
+ "mask_token": "<mask>",
520
+ "model_max_length": 1000000000000000019884624838656,
521
+ "pad_token": "<pad>",
522
+ "sep_token": "</s>",
523
+ "sp_model_kwargs": {},
524
+ "src_lang": "ja_XX",
525
+ "tgt_lang": "ko_KR",
526
+ "tokenizer_class": "MBart50Tokenizer",
527
+ "unk_token": "<unk>"
528
+ }
checkpoint-217522/trainer_state.json ADDED
@@ -0,0 +1,2679 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6814723610877991,
3
+ "best_model_checkpoint": "./jako_mbartLarge_100p_run1/checkpoint-87009",
4
+ "epoch": 4.999988506936064,
5
+ "eval_steps": 500,
6
+ "global_step": 217522,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 4.97e-05,
14
+ "loss": 1.9891,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.02,
19
+ "learning_rate": 4.996196669018189e-05,
20
+ "loss": 1.5354,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.03,
25
+ "learning_rate": 4.9923626660123305e-05,
26
+ "loss": 1.3833,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.05,
31
+ "learning_rate": 4.988528663006472e-05,
32
+ "loss": 1.2912,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.06,
37
+ "learning_rate": 4.984694660000614e-05,
38
+ "loss": 1.2355,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.07,
43
+ "learning_rate": 4.980860656994755e-05,
44
+ "loss": 1.1927,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.08,
49
+ "learning_rate": 4.977026653988897e-05,
50
+ "loss": 1.166,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.09,
55
+ "learning_rate": 4.9731926509830386e-05,
56
+ "loss": 1.1238,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.1,
61
+ "learning_rate": 4.9693586479771804e-05,
62
+ "loss": 1.0972,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.11,
67
+ "learning_rate": 4.9655246449713215e-05,
68
+ "loss": 1.0964,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.13,
73
+ "learning_rate": 4.961690641965463e-05,
74
+ "loss": 1.0611,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.14,
79
+ "learning_rate": 4.957871974971629e-05,
80
+ "loss": 1.0487,
81
+ "step": 6000
82
+ },
83
+ {
84
+ "epoch": 0.15,
85
+ "learning_rate": 4.9540379719657705e-05,
86
+ "loss": 1.0166,
87
+ "step": 6500
88
+ },
89
+ {
90
+ "epoch": 0.16,
91
+ "learning_rate": 4.9502039689599116e-05,
92
+ "loss": 1.0162,
93
+ "step": 7000
94
+ },
95
+ {
96
+ "epoch": 0.17,
97
+ "learning_rate": 4.946369965954053e-05,
98
+ "loss": 0.9952,
99
+ "step": 7500
100
+ },
101
+ {
102
+ "epoch": 0.18,
103
+ "learning_rate": 4.942535962948195e-05,
104
+ "loss": 0.9853,
105
+ "step": 8000
106
+ },
107
+ {
108
+ "epoch": 0.2,
109
+ "learning_rate": 4.9387096279483484e-05,
110
+ "loss": 0.9835,
111
+ "step": 8500
112
+ },
113
+ {
114
+ "epoch": 0.21,
115
+ "learning_rate": 4.9348832929485017e-05,
116
+ "loss": 0.9634,
117
+ "step": 9000
118
+ },
119
+ {
120
+ "epoch": 0.22,
121
+ "learning_rate": 4.9310492899426434e-05,
122
+ "loss": 0.9532,
123
+ "step": 9500
124
+ },
125
+ {
126
+ "epoch": 0.23,
127
+ "learning_rate": 4.927215286936785e-05,
128
+ "loss": 0.955,
129
+ "step": 10000
130
+ },
131
+ {
132
+ "epoch": 0.24,
133
+ "learning_rate": 4.923381283930927e-05,
134
+ "loss": 0.9392,
135
+ "step": 10500
136
+ },
137
+ {
138
+ "epoch": 0.25,
139
+ "learning_rate": 4.919547280925069e-05,
140
+ "loss": 0.9403,
141
+ "step": 11000
142
+ },
143
+ {
144
+ "epoch": 0.26,
145
+ "learning_rate": 4.91571327791921e-05,
146
+ "loss": 0.9177,
147
+ "step": 11500
148
+ },
149
+ {
150
+ "epoch": 0.28,
151
+ "learning_rate": 4.9118792749133516e-05,
152
+ "loss": 0.919,
153
+ "step": 12000
154
+ },
155
+ {
156
+ "epoch": 0.29,
157
+ "learning_rate": 4.908045271907493e-05,
158
+ "loss": 0.9036,
159
+ "step": 12500
160
+ },
161
+ {
162
+ "epoch": 0.3,
163
+ "learning_rate": 4.9042189369076466e-05,
164
+ "loss": 0.9019,
165
+ "step": 13000
166
+ },
167
+ {
168
+ "epoch": 0.31,
169
+ "learning_rate": 4.9003849339017884e-05,
170
+ "loss": 0.8969,
171
+ "step": 13500
172
+ },
173
+ {
174
+ "epoch": 0.32,
175
+ "learning_rate": 4.89655093089593e-05,
176
+ "loss": 0.891,
177
+ "step": 14000
178
+ },
179
+ {
180
+ "epoch": 0.33,
181
+ "learning_rate": 4.892716927890072e-05,
182
+ "loss": 0.8904,
183
+ "step": 14500
184
+ },
185
+ {
186
+ "epoch": 0.34,
187
+ "learning_rate": 4.888882924884213e-05,
188
+ "loss": 0.8833,
189
+ "step": 15000
190
+ },
191
+ {
192
+ "epoch": 0.36,
193
+ "learning_rate": 4.8850489218783554e-05,
194
+ "loss": 0.8668,
195
+ "step": 15500
196
+ },
197
+ {
198
+ "epoch": 0.37,
199
+ "learning_rate": 4.8812149188724965e-05,
200
+ "loss": 0.8692,
201
+ "step": 16000
202
+ },
203
+ {
204
+ "epoch": 0.38,
205
+ "learning_rate": 4.877380915866638e-05,
206
+ "loss": 0.8617,
207
+ "step": 16500
208
+ },
209
+ {
210
+ "epoch": 0.39,
211
+ "learning_rate": 4.8735469128607794e-05,
212
+ "loss": 0.866,
213
+ "step": 17000
214
+ },
215
+ {
216
+ "epoch": 0.4,
217
+ "learning_rate": 4.869720577860933e-05,
218
+ "loss": 0.8592,
219
+ "step": 17500
220
+ },
221
+ {
222
+ "epoch": 0.41,
223
+ "learning_rate": 4.8658865748550744e-05,
224
+ "loss": 0.8586,
225
+ "step": 18000
226
+ },
227
+ {
228
+ "epoch": 0.43,
229
+ "learning_rate": 4.862052571849217e-05,
230
+ "loss": 0.84,
231
+ "step": 18500
232
+ },
233
+ {
234
+ "epoch": 0.44,
235
+ "learning_rate": 4.8582262368493695e-05,
236
+ "loss": 0.84,
237
+ "step": 19000
238
+ },
239
+ {
240
+ "epoch": 0.45,
241
+ "learning_rate": 4.854392233843512e-05,
242
+ "loss": 0.8397,
243
+ "step": 19500
244
+ },
245
+ {
246
+ "epoch": 0.46,
247
+ "learning_rate": 4.850581234855688e-05,
248
+ "loss": 0.8338,
249
+ "step": 20000
250
+ },
251
+ {
252
+ "epoch": 0.47,
253
+ "learning_rate": 4.84674723184983e-05,
254
+ "loss": 0.838,
255
+ "step": 20500
256
+ },
257
+ {
258
+ "epoch": 0.48,
259
+ "learning_rate": 4.842913228843972e-05,
260
+ "loss": 0.8391,
261
+ "step": 21000
262
+ },
263
+ {
264
+ "epoch": 0.49,
265
+ "learning_rate": 4.8390792258381135e-05,
266
+ "loss": 0.8276,
267
+ "step": 21500
268
+ },
269
+ {
270
+ "epoch": 0.51,
271
+ "learning_rate": 4.8352452228322546e-05,
272
+ "loss": 0.8198,
273
+ "step": 22000
274
+ },
275
+ {
276
+ "epoch": 0.52,
277
+ "learning_rate": 4.8314112198263963e-05,
278
+ "loss": 0.8127,
279
+ "step": 22500
280
+ },
281
+ {
282
+ "epoch": 0.53,
283
+ "learning_rate": 4.827577216820538e-05,
284
+ "loss": 0.8184,
285
+ "step": 23000
286
+ },
287
+ {
288
+ "epoch": 0.54,
289
+ "learning_rate": 4.82374321381468e-05,
290
+ "loss": 0.817,
291
+ "step": 23500
292
+ },
293
+ {
294
+ "epoch": 0.55,
295
+ "learning_rate": 4.8199092108088216e-05,
296
+ "loss": 0.8196,
297
+ "step": 24000
298
+ },
299
+ {
300
+ "epoch": 0.56,
301
+ "learning_rate": 4.816075207802963e-05,
302
+ "loss": 0.8024,
303
+ "step": 24500
304
+ },
305
+ {
306
+ "epoch": 0.57,
307
+ "learning_rate": 4.812241204797105e-05,
308
+ "loss": 0.8133,
309
+ "step": 25000
310
+ },
311
+ {
312
+ "epoch": 0.59,
313
+ "learning_rate": 4.808407201791246e-05,
314
+ "loss": 0.8056,
315
+ "step": 25500
316
+ },
317
+ {
318
+ "epoch": 0.6,
319
+ "learning_rate": 4.804573198785388e-05,
320
+ "loss": 0.7972,
321
+ "step": 26000
322
+ },
323
+ {
324
+ "epoch": 0.61,
325
+ "learning_rate": 4.80073919577953e-05,
326
+ "loss": 0.8026,
327
+ "step": 26500
328
+ },
329
+ {
330
+ "epoch": 0.62,
331
+ "learning_rate": 4.7969051927736716e-05,
332
+ "loss": 0.7983,
333
+ "step": 27000
334
+ },
335
+ {
336
+ "epoch": 0.63,
337
+ "learning_rate": 4.793078857773824e-05,
338
+ "loss": 0.795,
339
+ "step": 27500
340
+ },
341
+ {
342
+ "epoch": 0.64,
343
+ "learning_rate": 4.7892448547679666e-05,
344
+ "loss": 0.803,
345
+ "step": 28000
346
+ },
347
+ {
348
+ "epoch": 0.66,
349
+ "learning_rate": 4.7854108517621084e-05,
350
+ "loss": 0.7911,
351
+ "step": 28500
352
+ },
353
+ {
354
+ "epoch": 0.67,
355
+ "learning_rate": 4.7815768487562494e-05,
356
+ "loss": 0.7757,
357
+ "step": 29000
358
+ },
359
+ {
360
+ "epoch": 0.68,
361
+ "learning_rate": 4.777742845750391e-05,
362
+ "loss": 0.777,
363
+ "step": 29500
364
+ },
365
+ {
366
+ "epoch": 0.69,
367
+ "learning_rate": 4.773908842744533e-05,
368
+ "loss": 0.7848,
369
+ "step": 30000
370
+ },
371
+ {
372
+ "epoch": 0.7,
373
+ "learning_rate": 4.770074839738675e-05,
374
+ "loss": 0.7834,
375
+ "step": 30500
376
+ },
377
+ {
378
+ "epoch": 0.71,
379
+ "learning_rate": 4.766240836732816e-05,
380
+ "loss": 0.7753,
381
+ "step": 31000
382
+ },
383
+ {
384
+ "epoch": 0.72,
385
+ "learning_rate": 4.7624068337269576e-05,
386
+ "loss": 0.7715,
387
+ "step": 31500
388
+ },
389
+ {
390
+ "epoch": 0.74,
391
+ "learning_rate": 4.7585728307211e-05,
392
+ "loss": 0.7714,
393
+ "step": 32000
394
+ },
395
+ {
396
+ "epoch": 0.75,
397
+ "learning_rate": 4.754738827715241e-05,
398
+ "loss": 0.7736,
399
+ "step": 32500
400
+ },
401
+ {
402
+ "epoch": 0.76,
403
+ "learning_rate": 4.750904824709383e-05,
404
+ "loss": 0.7767,
405
+ "step": 33000
406
+ },
407
+ {
408
+ "epoch": 0.77,
409
+ "learning_rate": 4.747070821703524e-05,
410
+ "loss": 0.7661,
411
+ "step": 33500
412
+ },
413
+ {
414
+ "epoch": 0.78,
415
+ "learning_rate": 4.7432368186976664e-05,
416
+ "loss": 0.7565,
417
+ "step": 34000
418
+ },
419
+ {
420
+ "epoch": 0.79,
421
+ "learning_rate": 4.739410483697819e-05,
422
+ "loss": 0.7515,
423
+ "step": 34500
424
+ },
425
+ {
426
+ "epoch": 0.8,
427
+ "learning_rate": 4.735584148697973e-05,
428
+ "loss": 0.7558,
429
+ "step": 35000
430
+ },
431
+ {
432
+ "epoch": 0.82,
433
+ "learning_rate": 4.731750145692114e-05,
434
+ "loss": 0.7642,
435
+ "step": 35500
436
+ },
437
+ {
438
+ "epoch": 0.83,
439
+ "learning_rate": 4.727916142686256e-05,
440
+ "loss": 0.7626,
441
+ "step": 36000
442
+ },
443
+ {
444
+ "epoch": 0.84,
445
+ "learning_rate": 4.7240821396803976e-05,
446
+ "loss": 0.7627,
447
+ "step": 36500
448
+ },
449
+ {
450
+ "epoch": 0.85,
451
+ "learning_rate": 4.7202481366745394e-05,
452
+ "loss": 0.7495,
453
+ "step": 37000
454
+ },
455
+ {
456
+ "epoch": 0.86,
457
+ "learning_rate": 4.716414133668681e-05,
458
+ "loss": 0.7582,
459
+ "step": 37500
460
+ },
461
+ {
462
+ "epoch": 0.87,
463
+ "learning_rate": 4.7125877986688344e-05,
464
+ "loss": 0.7481,
465
+ "step": 38000
466
+ },
467
+ {
468
+ "epoch": 0.88,
469
+ "learning_rate": 4.708753795662976e-05,
470
+ "loss": 0.7557,
471
+ "step": 38500
472
+ },
473
+ {
474
+ "epoch": 0.9,
475
+ "learning_rate": 4.704919792657117e-05,
476
+ "loss": 0.7384,
477
+ "step": 39000
478
+ },
479
+ {
480
+ "epoch": 0.91,
481
+ "learning_rate": 4.70108578965126e-05,
482
+ "loss": 0.742,
483
+ "step": 39500
484
+ },
485
+ {
486
+ "epoch": 0.92,
487
+ "learning_rate": 4.697251786645401e-05,
488
+ "loss": 0.7388,
489
+ "step": 40000
490
+ },
491
+ {
492
+ "epoch": 0.93,
493
+ "learning_rate": 4.6934177836395425e-05,
494
+ "loss": 0.7503,
495
+ "step": 40500
496
+ },
497
+ {
498
+ "epoch": 0.94,
499
+ "learning_rate": 4.689583780633684e-05,
500
+ "loss": 0.745,
501
+ "step": 41000
502
+ },
503
+ {
504
+ "epoch": 0.95,
505
+ "learning_rate": 4.685749777627826e-05,
506
+ "loss": 0.7228,
507
+ "step": 41500
508
+ },
509
+ {
510
+ "epoch": 0.97,
511
+ "learning_rate": 4.681915774621968e-05,
512
+ "loss": 0.7333,
513
+ "step": 42000
514
+ },
515
+ {
516
+ "epoch": 0.98,
517
+ "learning_rate": 4.678081771616109e-05,
518
+ "loss": 0.7371,
519
+ "step": 42500
520
+ },
521
+ {
522
+ "epoch": 0.99,
523
+ "learning_rate": 4.674247768610251e-05,
524
+ "loss": 0.733,
525
+ "step": 43000
526
+ },
527
+ {
528
+ "epoch": 1.0,
529
+ "learning_rate": 4.6704137656043925e-05,
530
+ "loss": 0.7241,
531
+ "step": 43500
532
+ },
533
+ {
534
+ "epoch": 1.0,
535
+ "eval_bleu": 56.112,
536
+ "eval_gen_len": 17.3165,
537
+ "eval_loss": 0.7177675366401672,
538
+ "eval_runtime": 7671.1849,
539
+ "eval_samples_per_second": 11.342,
540
+ "eval_steps_per_second": 1.418,
541
+ "step": 43504
542
+ },
543
+ {
544
+ "epoch": 1.01,
545
+ "learning_rate": 4.666587430604546e-05,
546
+ "loss": 0.7322,
547
+ "step": 44000
548
+ },
549
+ {
550
+ "epoch": 1.02,
551
+ "learning_rate": 4.6627534275986875e-05,
552
+ "loss": 0.7051,
553
+ "step": 44500
554
+ },
555
+ {
556
+ "epoch": 1.03,
557
+ "learning_rate": 4.658919424592829e-05,
558
+ "loss": 0.6935,
559
+ "step": 45000
560
+ },
561
+ {
562
+ "epoch": 1.05,
563
+ "learning_rate": 4.6550854215869704e-05,
564
+ "loss": 0.6831,
565
+ "step": 45500
566
+ },
567
+ {
568
+ "epoch": 1.06,
569
+ "learning_rate": 4.651259086587124e-05,
570
+ "loss": 0.6787,
571
+ "step": 46000
572
+ },
573
+ {
574
+ "epoch": 1.07,
575
+ "learning_rate": 4.6474327515872776e-05,
576
+ "loss": 0.6813,
577
+ "step": 46500
578
+ },
579
+ {
580
+ "epoch": 1.08,
581
+ "learning_rate": 4.6436140845934424e-05,
582
+ "loss": 0.6841,
583
+ "step": 47000
584
+ },
585
+ {
586
+ "epoch": 1.09,
587
+ "learning_rate": 4.639780081587584e-05,
588
+ "loss": 0.6697,
589
+ "step": 47500
590
+ },
591
+ {
592
+ "epoch": 1.1,
593
+ "learning_rate": 4.635946078581726e-05,
594
+ "loss": 0.6605,
595
+ "step": 48000
596
+ },
597
+ {
598
+ "epoch": 1.11,
599
+ "learning_rate": 4.632112075575868e-05,
600
+ "loss": 0.675,
601
+ "step": 48500
602
+ },
603
+ {
604
+ "epoch": 1.13,
605
+ "learning_rate": 4.6282780725700094e-05,
606
+ "loss": 0.6623,
607
+ "step": 49000
608
+ },
609
+ {
610
+ "epoch": 1.14,
611
+ "learning_rate": 4.6244440695641505e-05,
612
+ "loss": 0.6589,
613
+ "step": 49500
614
+ },
615
+ {
616
+ "epoch": 1.15,
617
+ "learning_rate": 4.620610066558292e-05,
618
+ "loss": 0.6476,
619
+ "step": 50000
620
+ },
621
+ {
622
+ "epoch": 1.16,
623
+ "learning_rate": 4.616776063552434e-05,
624
+ "loss": 0.6551,
625
+ "step": 50500
626
+ },
627
+ {
628
+ "epoch": 1.17,
629
+ "learning_rate": 4.612942060546576e-05,
630
+ "loss": 0.6502,
631
+ "step": 51000
632
+ },
633
+ {
634
+ "epoch": 1.18,
635
+ "learning_rate": 4.6091080575407176e-05,
636
+ "loss": 0.6478,
637
+ "step": 51500
638
+ },
639
+ {
640
+ "epoch": 1.2,
641
+ "learning_rate": 4.605274054534859e-05,
642
+ "loss": 0.6485,
643
+ "step": 52000
644
+ },
645
+ {
646
+ "epoch": 1.21,
647
+ "learning_rate": 4.6014400515290004e-05,
648
+ "loss": 0.6412,
649
+ "step": 52500
650
+ },
651
+ {
652
+ "epoch": 1.22,
653
+ "learning_rate": 4.597606048523142e-05,
654
+ "loss": 0.6376,
655
+ "step": 53000
656
+ },
657
+ {
658
+ "epoch": 1.23,
659
+ "learning_rate": 4.593772045517284e-05,
660
+ "loss": 0.6385,
661
+ "step": 53500
662
+ },
663
+ {
664
+ "epoch": 1.24,
665
+ "learning_rate": 4.589945710517437e-05,
666
+ "loss": 0.6366,
667
+ "step": 54000
668
+ },
669
+ {
670
+ "epoch": 1.25,
671
+ "learning_rate": 4.586111707511579e-05,
672
+ "loss": 0.6444,
673
+ "step": 54500
674
+ },
675
+ {
676
+ "epoch": 1.26,
677
+ "learning_rate": 4.582277704505721e-05,
678
+ "loss": 0.6311,
679
+ "step": 55000
680
+ },
681
+ {
682
+ "epoch": 1.28,
683
+ "learning_rate": 4.578451369505874e-05,
684
+ "loss": 0.6306,
685
+ "step": 55500
686
+ },
687
+ {
688
+ "epoch": 1.29,
689
+ "learning_rate": 4.574617366500016e-05,
690
+ "loss": 0.6199,
691
+ "step": 56000
692
+ },
693
+ {
694
+ "epoch": 1.3,
695
+ "learning_rate": 4.570783363494157e-05,
696
+ "loss": 0.6268,
697
+ "step": 56500
698
+ },
699
+ {
700
+ "epoch": 1.31,
701
+ "learning_rate": 4.566949360488299e-05,
702
+ "loss": 0.623,
703
+ "step": 57000
704
+ },
705
+ {
706
+ "epoch": 1.32,
707
+ "learning_rate": 4.5631153574824404e-05,
708
+ "loss": 0.6207,
709
+ "step": 57500
710
+ },
711
+ {
712
+ "epoch": 1.33,
713
+ "learning_rate": 4.559281354476582e-05,
714
+ "loss": 0.6204,
715
+ "step": 58000
716
+ },
717
+ {
718
+ "epoch": 1.34,
719
+ "learning_rate": 4.555447351470723e-05,
720
+ "loss": 0.6215,
721
+ "step": 58500
722
+ },
723
+ {
724
+ "epoch": 1.36,
725
+ "learning_rate": 4.551613348464866e-05,
726
+ "loss": 0.6114,
727
+ "step": 59000
728
+ },
729
+ {
730
+ "epoch": 1.37,
731
+ "learning_rate": 4.5477793454590075e-05,
732
+ "loss": 0.6134,
733
+ "step": 59500
734
+ },
735
+ {
736
+ "epoch": 1.38,
737
+ "learning_rate": 4.543953010459161e-05,
738
+ "loss": 0.6142,
739
+ "step": 60000
740
+ },
741
+ {
742
+ "epoch": 1.39,
743
+ "learning_rate": 4.5401190074533025e-05,
744
+ "loss": 0.6149,
745
+ "step": 60500
746
+ },
747
+ {
748
+ "epoch": 1.4,
749
+ "learning_rate": 4.5362850044474436e-05,
750
+ "loss": 0.6123,
751
+ "step": 61000
752
+ },
753
+ {
754
+ "epoch": 1.41,
755
+ "learning_rate": 4.5324510014415854e-05,
756
+ "loss": 0.6159,
757
+ "step": 61500
758
+ },
759
+ {
760
+ "epoch": 1.43,
761
+ "learning_rate": 4.5286246664417387e-05,
762
+ "loss": 0.6001,
763
+ "step": 62000
764
+ },
765
+ {
766
+ "epoch": 1.44,
767
+ "learning_rate": 4.5247906634358804e-05,
768
+ "loss": 0.6034,
769
+ "step": 62500
770
+ },
771
+ {
772
+ "epoch": 1.45,
773
+ "learning_rate": 4.520956660430022e-05,
774
+ "loss": 0.5999,
775
+ "step": 63000
776
+ },
777
+ {
778
+ "epoch": 1.46,
779
+ "learning_rate": 4.517122657424164e-05,
780
+ "loss": 0.6011,
781
+ "step": 63500
782
+ },
783
+ {
784
+ "epoch": 1.47,
785
+ "learning_rate": 4.513288654418305e-05,
786
+ "loss": 0.6051,
787
+ "step": 64000
788
+ },
789
+ {
790
+ "epoch": 1.48,
791
+ "learning_rate": 4.509454651412447e-05,
792
+ "loss": 0.6061,
793
+ "step": 64500
794
+ },
795
+ {
796
+ "epoch": 1.49,
797
+ "learning_rate": 4.5056206484065886e-05,
798
+ "loss": 0.598,
799
+ "step": 65000
800
+ },
801
+ {
802
+ "epoch": 1.51,
803
+ "learning_rate": 4.50178664540073e-05,
804
+ "loss": 0.5939,
805
+ "step": 65500
806
+ },
807
+ {
808
+ "epoch": 1.52,
809
+ "learning_rate": 4.497952642394872e-05,
810
+ "loss": 0.5905,
811
+ "step": 66000
812
+ },
813
+ {
814
+ "epoch": 1.53,
815
+ "learning_rate": 4.494118639389013e-05,
816
+ "loss": 0.5968,
817
+ "step": 66500
818
+ },
819
+ {
820
+ "epoch": 1.54,
821
+ "learning_rate": 4.490284636383155e-05,
822
+ "loss": 0.5943,
823
+ "step": 67000
824
+ },
825
+ {
826
+ "epoch": 1.55,
827
+ "learning_rate": 4.486450633377297e-05,
828
+ "loss": 0.6025,
829
+ "step": 67500
830
+ },
831
+ {
832
+ "epoch": 1.56,
833
+ "learning_rate": 4.4826166303714385e-05,
834
+ "loss": 0.5918,
835
+ "step": 68000
836
+ },
837
+ {
838
+ "epoch": 1.57,
839
+ "learning_rate": 4.47878262736558e-05,
840
+ "loss": 0.5983,
841
+ "step": 68500
842
+ },
843
+ {
844
+ "epoch": 1.59,
845
+ "learning_rate": 4.474948624359721e-05,
846
+ "loss": 0.5907,
847
+ "step": 69000
848
+ },
849
+ {
850
+ "epoch": 1.6,
851
+ "learning_rate": 4.471114621353864e-05,
852
+ "loss": 0.5852,
853
+ "step": 69500
854
+ },
855
+ {
856
+ "epoch": 1.61,
857
+ "learning_rate": 4.4672882863540164e-05,
858
+ "loss": 0.5886,
859
+ "step": 70000
860
+ },
861
+ {
862
+ "epoch": 1.62,
863
+ "learning_rate": 4.463454283348159e-05,
864
+ "loss": 0.5895,
865
+ "step": 70500
866
+ },
867
+ {
868
+ "epoch": 1.63,
869
+ "learning_rate": 4.4596202803423e-05,
870
+ "loss": 0.5887,
871
+ "step": 71000
872
+ },
873
+ {
874
+ "epoch": 1.64,
875
+ "learning_rate": 4.455786277336442e-05,
876
+ "loss": 0.5953,
877
+ "step": 71500
878
+ },
879
+ {
880
+ "epoch": 1.66,
881
+ "learning_rate": 4.451952274330583e-05,
882
+ "loss": 0.5888,
883
+ "step": 72000
884
+ },
885
+ {
886
+ "epoch": 1.67,
887
+ "learning_rate": 4.448118271324725e-05,
888
+ "loss": 0.571,
889
+ "step": 72500
890
+ },
891
+ {
892
+ "epoch": 1.68,
893
+ "learning_rate": 4.444291936324878e-05,
894
+ "loss": 0.576,
895
+ "step": 73000
896
+ },
897
+ {
898
+ "epoch": 1.69,
899
+ "learning_rate": 4.44045793331902e-05,
900
+ "loss": 0.5828,
901
+ "step": 73500
902
+ },
903
+ {
904
+ "epoch": 1.7,
905
+ "learning_rate": 4.436631598319173e-05,
906
+ "loss": 0.5875,
907
+ "step": 74000
908
+ },
909
+ {
910
+ "epoch": 1.71,
911
+ "learning_rate": 4.432797595313315e-05,
912
+ "loss": 0.5773,
913
+ "step": 74500
914
+ },
915
+ {
916
+ "epoch": 1.72,
917
+ "learning_rate": 4.428963592307457e-05,
918
+ "loss": 0.5741,
919
+ "step": 75000
920
+ },
921
+ {
922
+ "epoch": 1.74,
923
+ "learning_rate": 4.425129589301598e-05,
924
+ "loss": 0.5721,
925
+ "step": 75500
926
+ },
927
+ {
928
+ "epoch": 1.75,
929
+ "learning_rate": 4.42129558629574e-05,
930
+ "loss": 0.5797,
931
+ "step": 76000
932
+ },
933
+ {
934
+ "epoch": 1.76,
935
+ "learning_rate": 4.417461583289882e-05,
936
+ "loss": 0.5787,
937
+ "step": 76500
938
+ },
939
+ {
940
+ "epoch": 1.77,
941
+ "learning_rate": 4.4136275802840234e-05,
942
+ "loss": 0.5731,
943
+ "step": 77000
944
+ },
945
+ {
946
+ "epoch": 1.78,
947
+ "learning_rate": 4.409801245284177e-05,
948
+ "loss": 0.5642,
949
+ "step": 77500
950
+ },
951
+ {
952
+ "epoch": 1.79,
953
+ "learning_rate": 4.4059672422783185e-05,
954
+ "loss": 0.5653,
955
+ "step": 78000
956
+ },
957
+ {
958
+ "epoch": 1.8,
959
+ "learning_rate": 4.4021332392724596e-05,
960
+ "loss": 0.5663,
961
+ "step": 78500
962
+ },
963
+ {
964
+ "epoch": 1.82,
965
+ "learning_rate": 4.398299236266601e-05,
966
+ "loss": 0.5762,
967
+ "step": 79000
968
+ },
969
+ {
970
+ "epoch": 1.83,
971
+ "learning_rate": 4.394465233260743e-05,
972
+ "loss": 0.5742,
973
+ "step": 79500
974
+ },
975
+ {
976
+ "epoch": 1.84,
977
+ "learning_rate": 4.390631230254885e-05,
978
+ "loss": 0.5739,
979
+ "step": 80000
980
+ },
981
+ {
982
+ "epoch": 1.85,
983
+ "learning_rate": 4.3867972272490266e-05,
984
+ "loss": 0.5686,
985
+ "step": 80500
986
+ },
987
+ {
988
+ "epoch": 1.86,
989
+ "learning_rate": 4.382963224243168e-05,
990
+ "loss": 0.5757,
991
+ "step": 81000
992
+ },
993
+ {
994
+ "epoch": 1.87,
995
+ "learning_rate": 4.3791292212373095e-05,
996
+ "loss": 0.5654,
997
+ "step": 81500
998
+ },
999
+ {
1000
+ "epoch": 1.88,
1001
+ "learning_rate": 4.375295218231451e-05,
1002
+ "loss": 0.5697,
1003
+ "step": 82000
1004
+ },
1005
+ {
1006
+ "epoch": 1.9,
1007
+ "learning_rate": 4.371461215225593e-05,
1008
+ "loss": 0.5582,
1009
+ "step": 82500
1010
+ },
1011
+ {
1012
+ "epoch": 1.91,
1013
+ "learning_rate": 4.367627212219735e-05,
1014
+ "loss": 0.5603,
1015
+ "step": 83000
1016
+ },
1017
+ {
1018
+ "epoch": 1.92,
1019
+ "learning_rate": 4.363793209213876e-05,
1020
+ "loss": 0.5564,
1021
+ "step": 83500
1022
+ },
1023
+ {
1024
+ "epoch": 1.93,
1025
+ "learning_rate": 4.359959206208018e-05,
1026
+ "loss": 0.5682,
1027
+ "step": 84000
1028
+ },
1029
+ {
1030
+ "epoch": 1.94,
1031
+ "learning_rate": 4.356132871208171e-05,
1032
+ "loss": 0.566,
1033
+ "step": 84500
1034
+ },
1035
+ {
1036
+ "epoch": 1.95,
1037
+ "learning_rate": 4.3522988682023133e-05,
1038
+ "loss": 0.5485,
1039
+ "step": 85000
1040
+ },
1041
+ {
1042
+ "epoch": 1.97,
1043
+ "learning_rate": 4.3484648651964544e-05,
1044
+ "loss": 0.5558,
1045
+ "step": 85500
1046
+ },
1047
+ {
1048
+ "epoch": 1.98,
1049
+ "learning_rate": 4.344630862190596e-05,
1050
+ "loss": 0.5592,
1051
+ "step": 86000
1052
+ },
1053
+ {
1054
+ "epoch": 1.99,
1055
+ "learning_rate": 4.340796859184737e-05,
1056
+ "loss": 0.5611,
1057
+ "step": 86500
1058
+ },
1059
+ {
1060
+ "epoch": 2.0,
1061
+ "learning_rate": 4.337001196208938e-05,
1062
+ "loss": 0.5523,
1063
+ "step": 87000
1064
+ },
1065
+ {
1066
+ "epoch": 2.0,
1067
+ "eval_bleu": 57.9768,
1068
+ "eval_gen_len": 17.2796,
1069
+ "eval_loss": 0.6814723610877991,
1070
+ "eval_runtime": 7592.1301,
1071
+ "eval_samples_per_second": 11.46,
1072
+ "eval_steps_per_second": 1.433,
1073
+ "step": 87009
1074
+ },
1075
+ {
1076
+ "epoch": 2.01,
1077
+ "learning_rate": 4.33316719320308e-05,
1078
+ "loss": 0.5617,
1079
+ "step": 87500
1080
+ },
1081
+ {
1082
+ "epoch": 2.02,
1083
+ "learning_rate": 4.3293331901972215e-05,
1084
+ "loss": 0.5392,
1085
+ "step": 88000
1086
+ },
1087
+ {
1088
+ "epoch": 2.03,
1089
+ "learning_rate": 4.325499187191363e-05,
1090
+ "loss": 0.5313,
1091
+ "step": 88500
1092
+ },
1093
+ {
1094
+ "epoch": 2.05,
1095
+ "learning_rate": 4.321665184185504e-05,
1096
+ "loss": 0.5247,
1097
+ "step": 89000
1098
+ },
1099
+ {
1100
+ "epoch": 2.06,
1101
+ "learning_rate": 4.317831181179646e-05,
1102
+ "loss": 0.5183,
1103
+ "step": 89500
1104
+ },
1105
+ {
1106
+ "epoch": 2.07,
1107
+ "learning_rate": 4.313997178173788e-05,
1108
+ "loss": 0.5266,
1109
+ "step": 90000
1110
+ },
1111
+ {
1112
+ "epoch": 2.08,
1113
+ "learning_rate": 4.3101631751679296e-05,
1114
+ "loss": 0.5256,
1115
+ "step": 90500
1116
+ },
1117
+ {
1118
+ "epoch": 2.09,
1119
+ "learning_rate": 4.3063291721620714e-05,
1120
+ "loss": 0.5148,
1121
+ "step": 91000
1122
+ },
1123
+ {
1124
+ "epoch": 2.1,
1125
+ "learning_rate": 4.3024951691562125e-05,
1126
+ "loss": 0.5097,
1127
+ "step": 91500
1128
+ },
1129
+ {
1130
+ "epoch": 2.11,
1131
+ "learning_rate": 4.298661166150354e-05,
1132
+ "loss": 0.5192,
1133
+ "step": 92000
1134
+ },
1135
+ {
1136
+ "epoch": 2.13,
1137
+ "learning_rate": 4.294827163144496e-05,
1138
+ "loss": 0.5125,
1139
+ "step": 92500
1140
+ },
1141
+ {
1142
+ "epoch": 2.14,
1143
+ "learning_rate": 4.290993160138638e-05,
1144
+ "loss": 0.5049,
1145
+ "step": 93000
1146
+ },
1147
+ {
1148
+ "epoch": 2.15,
1149
+ "learning_rate": 4.2871591571327795e-05,
1150
+ "loss": 0.4974,
1151
+ "step": 93500
1152
+ },
1153
+ {
1154
+ "epoch": 2.16,
1155
+ "learning_rate": 4.2833251541269206e-05,
1156
+ "loss": 0.5044,
1157
+ "step": 94000
1158
+ },
1159
+ {
1160
+ "epoch": 2.17,
1161
+ "learning_rate": 4.279491151121063e-05,
1162
+ "loss": 0.5026,
1163
+ "step": 94500
1164
+ },
1165
+ {
1166
+ "epoch": 2.18,
1167
+ "learning_rate": 4.275657148115204e-05,
1168
+ "loss": 0.5028,
1169
+ "step": 95000
1170
+ },
1171
+ {
1172
+ "epoch": 2.2,
1173
+ "learning_rate": 4.271823145109346e-05,
1174
+ "loss": 0.4992,
1175
+ "step": 95500
1176
+ },
1177
+ {
1178
+ "epoch": 2.21,
1179
+ "learning_rate": 4.267989142103488e-05,
1180
+ "loss": 0.5,
1181
+ "step": 96000
1182
+ },
1183
+ {
1184
+ "epoch": 2.22,
1185
+ "learning_rate": 4.264162807103641e-05,
1186
+ "loss": 0.4942,
1187
+ "step": 96500
1188
+ },
1189
+ {
1190
+ "epoch": 2.23,
1191
+ "learning_rate": 4.260328804097782e-05,
1192
+ "loss": 0.4942,
1193
+ "step": 97000
1194
+ },
1195
+ {
1196
+ "epoch": 2.24,
1197
+ "learning_rate": 4.2564948010919245e-05,
1198
+ "loss": 0.4952,
1199
+ "step": 97500
1200
+ },
1201
+ {
1202
+ "epoch": 2.25,
1203
+ "learning_rate": 4.252660798086066e-05,
1204
+ "loss": 0.4977,
1205
+ "step": 98000
1206
+ },
1207
+ {
1208
+ "epoch": 2.26,
1209
+ "learning_rate": 4.2488344630862195e-05,
1210
+ "loss": 0.4894,
1211
+ "step": 98500
1212
+ },
1213
+ {
1214
+ "epoch": 2.28,
1215
+ "learning_rate": 4.245008128086372e-05,
1216
+ "loss": 0.4887,
1217
+ "step": 99000
1218
+ },
1219
+ {
1220
+ "epoch": 2.29,
1221
+ "learning_rate": 4.2411741250805146e-05,
1222
+ "loss": 0.4813,
1223
+ "step": 99500
1224
+ },
1225
+ {
1226
+ "epoch": 2.3,
1227
+ "learning_rate": 4.2373401220746563e-05,
1228
+ "loss": 0.4851,
1229
+ "step": 100000
1230
+ },
1231
+ {
1232
+ "epoch": 2.31,
1233
+ "learning_rate": 4.2335061190687974e-05,
1234
+ "loss": 0.4865,
1235
+ "step": 100500
1236
+ },
1237
+ {
1238
+ "epoch": 2.32,
1239
+ "learning_rate": 4.229672116062939e-05,
1240
+ "loss": 0.4823,
1241
+ "step": 101000
1242
+ },
1243
+ {
1244
+ "epoch": 2.33,
1245
+ "learning_rate": 4.225838113057081e-05,
1246
+ "loss": 0.4806,
1247
+ "step": 101500
1248
+ },
1249
+ {
1250
+ "epoch": 2.34,
1251
+ "learning_rate": 4.222011778057234e-05,
1252
+ "loss": 0.4813,
1253
+ "step": 102000
1254
+ },
1255
+ {
1256
+ "epoch": 2.36,
1257
+ "learning_rate": 4.218177775051376e-05,
1258
+ "loss": 0.4734,
1259
+ "step": 102500
1260
+ },
1261
+ {
1262
+ "epoch": 2.37,
1263
+ "learning_rate": 4.214343772045518e-05,
1264
+ "loss": 0.479,
1265
+ "step": 103000
1266
+ },
1267
+ {
1268
+ "epoch": 2.38,
1269
+ "learning_rate": 4.210509769039659e-05,
1270
+ "loss": 0.4778,
1271
+ "step": 103500
1272
+ },
1273
+ {
1274
+ "epoch": 2.39,
1275
+ "learning_rate": 4.2066757660338006e-05,
1276
+ "loss": 0.4797,
1277
+ "step": 104000
1278
+ },
1279
+ {
1280
+ "epoch": 2.4,
1281
+ "learning_rate": 4.2028417630279424e-05,
1282
+ "loss": 0.474,
1283
+ "step": 104500
1284
+ },
1285
+ {
1286
+ "epoch": 2.41,
1287
+ "learning_rate": 4.199007760022084e-05,
1288
+ "loss": 0.4811,
1289
+ "step": 105000
1290
+ },
1291
+ {
1292
+ "epoch": 2.43,
1293
+ "learning_rate": 4.1951814250222374e-05,
1294
+ "loss": 0.464,
1295
+ "step": 105500
1296
+ },
1297
+ {
1298
+ "epoch": 2.44,
1299
+ "learning_rate": 4.191347422016379e-05,
1300
+ "loss": 0.4699,
1301
+ "step": 106000
1302
+ },
1303
+ {
1304
+ "epoch": 2.45,
1305
+ "learning_rate": 4.187513419010521e-05,
1306
+ "loss": 0.4655,
1307
+ "step": 106500
1308
+ },
1309
+ {
1310
+ "epoch": 2.46,
1311
+ "learning_rate": 4.183679416004662e-05,
1312
+ "loss": 0.4694,
1313
+ "step": 107000
1314
+ },
1315
+ {
1316
+ "epoch": 2.47,
1317
+ "learning_rate": 4.179845412998804e-05,
1318
+ "loss": 0.4698,
1319
+ "step": 107500
1320
+ },
1321
+ {
1322
+ "epoch": 2.48,
1323
+ "learning_rate": 4.1760114099929456e-05,
1324
+ "loss": 0.4714,
1325
+ "step": 108000
1326
+ },
1327
+ {
1328
+ "epoch": 2.49,
1329
+ "learning_rate": 4.1721774069870873e-05,
1330
+ "loss": 0.4627,
1331
+ "step": 108500
1332
+ },
1333
+ {
1334
+ "epoch": 2.51,
1335
+ "learning_rate": 4.168343403981229e-05,
1336
+ "loss": 0.4628,
1337
+ "step": 109000
1338
+ },
1339
+ {
1340
+ "epoch": 2.52,
1341
+ "learning_rate": 4.16450940097537e-05,
1342
+ "loss": 0.4565,
1343
+ "step": 109500
1344
+ },
1345
+ {
1346
+ "epoch": 2.53,
1347
+ "learning_rate": 4.1606753979695126e-05,
1348
+ "loss": 0.466,
1349
+ "step": 110000
1350
+ },
1351
+ {
1352
+ "epoch": 2.54,
1353
+ "learning_rate": 4.156841394963654e-05,
1354
+ "loss": 0.4635,
1355
+ "step": 110500
1356
+ },
1357
+ {
1358
+ "epoch": 2.55,
1359
+ "learning_rate": 4.1530073919577955e-05,
1360
+ "loss": 0.4694,
1361
+ "step": 111000
1362
+ },
1363
+ {
1364
+ "epoch": 2.56,
1365
+ "learning_rate": 4.149181056957949e-05,
1366
+ "loss": 0.4638,
1367
+ "step": 111500
1368
+ },
1369
+ {
1370
+ "epoch": 2.57,
1371
+ "learning_rate": 4.1453470539520905e-05,
1372
+ "loss": 0.4626,
1373
+ "step": 112000
1374
+ },
1375
+ {
1376
+ "epoch": 2.59,
1377
+ "learning_rate": 4.1415130509462316e-05,
1378
+ "loss": 0.4632,
1379
+ "step": 112500
1380
+ },
1381
+ {
1382
+ "epoch": 2.6,
1383
+ "learning_rate": 4.137679047940374e-05,
1384
+ "loss": 0.455,
1385
+ "step": 113000
1386
+ },
1387
+ {
1388
+ "epoch": 2.61,
1389
+ "learning_rate": 4.133845044934516e-05,
1390
+ "loss": 0.4553,
1391
+ "step": 113500
1392
+ },
1393
+ {
1394
+ "epoch": 2.62,
1395
+ "learning_rate": 4.130018709934669e-05,
1396
+ "loss": 0.4583,
1397
+ "step": 114000
1398
+ },
1399
+ {
1400
+ "epoch": 2.63,
1401
+ "learning_rate": 4.126192374934822e-05,
1402
+ "loss": 0.4596,
1403
+ "step": 114500
1404
+ },
1405
+ {
1406
+ "epoch": 2.64,
1407
+ "learning_rate": 4.1223583719289635e-05,
1408
+ "loss": 0.4634,
1409
+ "step": 115000
1410
+ },
1411
+ {
1412
+ "epoch": 2.65,
1413
+ "learning_rate": 4.118524368923106e-05,
1414
+ "loss": 0.4595,
1415
+ "step": 115500
1416
+ },
1417
+ {
1418
+ "epoch": 2.67,
1419
+ "learning_rate": 4.1146980339232585e-05,
1420
+ "loss": 0.4441,
1421
+ "step": 116000
1422
+ },
1423
+ {
1424
+ "epoch": 2.68,
1425
+ "learning_rate": 4.110864030917401e-05,
1426
+ "loss": 0.4517,
1427
+ "step": 116500
1428
+ },
1429
+ {
1430
+ "epoch": 2.69,
1431
+ "learning_rate": 4.107030027911542e-05,
1432
+ "loss": 0.4531,
1433
+ "step": 117000
1434
+ },
1435
+ {
1436
+ "epoch": 2.7,
1437
+ "learning_rate": 4.103196024905684e-05,
1438
+ "loss": 0.4556,
1439
+ "step": 117500
1440
+ },
1441
+ {
1442
+ "epoch": 2.71,
1443
+ "learning_rate": 4.099362021899825e-05,
1444
+ "loss": 0.4481,
1445
+ "step": 118000
1446
+ },
1447
+ {
1448
+ "epoch": 2.72,
1449
+ "learning_rate": 4.095528018893967e-05,
1450
+ "loss": 0.4483,
1451
+ "step": 118500
1452
+ },
1453
+ {
1454
+ "epoch": 2.74,
1455
+ "learning_rate": 4.0916940158881084e-05,
1456
+ "loss": 0.4457,
1457
+ "step": 119000
1458
+ },
1459
+ {
1460
+ "epoch": 2.75,
1461
+ "learning_rate": 4.08786001288225e-05,
1462
+ "loss": 0.452,
1463
+ "step": 119500
1464
+ },
1465
+ {
1466
+ "epoch": 2.76,
1467
+ "learning_rate": 4.084026009876392e-05,
1468
+ "loss": 0.4511,
1469
+ "step": 120000
1470
+ },
1471
+ {
1472
+ "epoch": 2.77,
1473
+ "learning_rate": 4.080192006870534e-05,
1474
+ "loss": 0.4466,
1475
+ "step": 120500
1476
+ },
1477
+ {
1478
+ "epoch": 2.78,
1479
+ "learning_rate": 4.0763580038646755e-05,
1480
+ "loss": 0.4368,
1481
+ "step": 121000
1482
+ },
1483
+ {
1484
+ "epoch": 2.79,
1485
+ "learning_rate": 4.0725240008588166e-05,
1486
+ "loss": 0.443,
1487
+ "step": 121500
1488
+ },
1489
+ {
1490
+ "epoch": 2.8,
1491
+ "learning_rate": 4.0686976658589705e-05,
1492
+ "loss": 0.441,
1493
+ "step": 122000
1494
+ },
1495
+ {
1496
+ "epoch": 2.82,
1497
+ "learning_rate": 4.0648636628531116e-05,
1498
+ "loss": 0.4503,
1499
+ "step": 122500
1500
+ },
1501
+ {
1502
+ "epoch": 2.83,
1503
+ "learning_rate": 4.0610296598472534e-05,
1504
+ "loss": 0.4487,
1505
+ "step": 123000
1506
+ },
1507
+ {
1508
+ "epoch": 2.84,
1509
+ "learning_rate": 4.057195656841395e-05,
1510
+ "loss": 0.4482,
1511
+ "step": 123500
1512
+ },
1513
+ {
1514
+ "epoch": 2.85,
1515
+ "learning_rate": 4.053361653835537e-05,
1516
+ "loss": 0.4449,
1517
+ "step": 124000
1518
+ },
1519
+ {
1520
+ "epoch": 2.86,
1521
+ "learning_rate": 4.049527650829679e-05,
1522
+ "loss": 0.4514,
1523
+ "step": 124500
1524
+ },
1525
+ {
1526
+ "epoch": 2.87,
1527
+ "learning_rate": 4.04569364782382e-05,
1528
+ "loss": 0.4407,
1529
+ "step": 125000
1530
+ },
1531
+ {
1532
+ "epoch": 2.88,
1533
+ "learning_rate": 4.041867312823974e-05,
1534
+ "loss": 0.443,
1535
+ "step": 125500
1536
+ },
1537
+ {
1538
+ "epoch": 2.9,
1539
+ "learning_rate": 4.038033309818115e-05,
1540
+ "loss": 0.434,
1541
+ "step": 126000
1542
+ },
1543
+ {
1544
+ "epoch": 2.91,
1545
+ "learning_rate": 4.0341993068122566e-05,
1546
+ "loss": 0.4384,
1547
+ "step": 126500
1548
+ },
1549
+ {
1550
+ "epoch": 2.92,
1551
+ "learning_rate": 4.030365303806398e-05,
1552
+ "loss": 0.4334,
1553
+ "step": 127000
1554
+ },
1555
+ {
1556
+ "epoch": 2.93,
1557
+ "learning_rate": 4.0265389688065516e-05,
1558
+ "loss": 0.4419,
1559
+ "step": 127500
1560
+ },
1561
+ {
1562
+ "epoch": 2.94,
1563
+ "learning_rate": 4.0227049658006934e-05,
1564
+ "loss": 0.4421,
1565
+ "step": 128000
1566
+ },
1567
+ {
1568
+ "epoch": 2.95,
1569
+ "learning_rate": 4.018870962794835e-05,
1570
+ "loss": 0.4244,
1571
+ "step": 128500
1572
+ },
1573
+ {
1574
+ "epoch": 2.97,
1575
+ "learning_rate": 4.015036959788976e-05,
1576
+ "loss": 0.4345,
1577
+ "step": 129000
1578
+ },
1579
+ {
1580
+ "epoch": 2.98,
1581
+ "learning_rate": 4.01121062478913e-05,
1582
+ "loss": 0.4374,
1583
+ "step": 129500
1584
+ },
1585
+ {
1586
+ "epoch": 2.99,
1587
+ "learning_rate": 4.007376621783271e-05,
1588
+ "loss": 0.4368,
1589
+ "step": 130000
1590
+ },
1591
+ {
1592
+ "epoch": 3.0,
1593
+ "learning_rate": 4.003542618777413e-05,
1594
+ "loss": 0.4356,
1595
+ "step": 130500
1596
+ },
1597
+ {
1598
+ "epoch": 3.0,
1599
+ "eval_bleu": 58.5493,
1600
+ "eval_gen_len": 17.1569,
1601
+ "eval_loss": 0.7014300227165222,
1602
+ "eval_runtime": 7450.9373,
1603
+ "eval_samples_per_second": 11.677,
1604
+ "eval_steps_per_second": 1.46,
1605
+ "step": 130513
1606
+ },
1607
+ {
1608
+ "epoch": 3.01,
1609
+ "learning_rate": 3.9997086157715555e-05,
1610
+ "loss": 0.4373,
1611
+ "step": 131000
1612
+ },
1613
+ {
1614
+ "epoch": 3.02,
1615
+ "learning_rate": 3.9958746127656966e-05,
1616
+ "loss": 0.42,
1617
+ "step": 131500
1618
+ },
1619
+ {
1620
+ "epoch": 3.03,
1621
+ "learning_rate": 3.992040609759838e-05,
1622
+ "loss": 0.4179,
1623
+ "step": 132000
1624
+ },
1625
+ {
1626
+ "epoch": 3.05,
1627
+ "learning_rate": 3.9882066067539794e-05,
1628
+ "loss": 0.406,
1629
+ "step": 132500
1630
+ },
1631
+ {
1632
+ "epoch": 3.06,
1633
+ "learning_rate": 3.984372603748122e-05,
1634
+ "loss": 0.4034,
1635
+ "step": 133000
1636
+ },
1637
+ {
1638
+ "epoch": 3.07,
1639
+ "learning_rate": 3.980538600742263e-05,
1640
+ "loss": 0.4103,
1641
+ "step": 133500
1642
+ },
1643
+ {
1644
+ "epoch": 3.08,
1645
+ "learning_rate": 3.976704597736405e-05,
1646
+ "loss": 0.4084,
1647
+ "step": 134000
1648
+ },
1649
+ {
1650
+ "epoch": 3.09,
1651
+ "learning_rate": 3.9728705947305465e-05,
1652
+ "loss": 0.4012,
1653
+ "step": 134500
1654
+ },
1655
+ {
1656
+ "epoch": 3.1,
1657
+ "learning_rate": 3.969036591724688e-05,
1658
+ "loss": 0.3966,
1659
+ "step": 135000
1660
+ },
1661
+ {
1662
+ "epoch": 3.11,
1663
+ "learning_rate": 3.96520258871883e-05,
1664
+ "loss": 0.4026,
1665
+ "step": 135500
1666
+ },
1667
+ {
1668
+ "epoch": 3.13,
1669
+ "learning_rate": 3.961368585712971e-05,
1670
+ "loss": 0.3983,
1671
+ "step": 136000
1672
+ },
1673
+ {
1674
+ "epoch": 3.14,
1675
+ "learning_rate": 3.957542250713125e-05,
1676
+ "loss": 0.3947,
1677
+ "step": 136500
1678
+ },
1679
+ {
1680
+ "epoch": 3.15,
1681
+ "learning_rate": 3.953708247707266e-05,
1682
+ "loss": 0.3894,
1683
+ "step": 137000
1684
+ },
1685
+ {
1686
+ "epoch": 3.16,
1687
+ "learning_rate": 3.94988191270742e-05,
1688
+ "loss": 0.3927,
1689
+ "step": 137500
1690
+ },
1691
+ {
1692
+ "epoch": 3.17,
1693
+ "learning_rate": 3.946047909701561e-05,
1694
+ "loss": 0.3919,
1695
+ "step": 138000
1696
+ },
1697
+ {
1698
+ "epoch": 3.18,
1699
+ "learning_rate": 3.942213906695703e-05,
1700
+ "loss": 0.3909,
1701
+ "step": 138500
1702
+ },
1703
+ {
1704
+ "epoch": 3.2,
1705
+ "learning_rate": 3.938379903689845e-05,
1706
+ "loss": 0.3914,
1707
+ "step": 139000
1708
+ },
1709
+ {
1710
+ "epoch": 3.21,
1711
+ "learning_rate": 3.9345459006839865e-05,
1712
+ "loss": 0.3876,
1713
+ "step": 139500
1714
+ },
1715
+ {
1716
+ "epoch": 3.22,
1717
+ "learning_rate": 3.930727233690151e-05,
1718
+ "loss": 0.3844,
1719
+ "step": 140000
1720
+ },
1721
+ {
1722
+ "epoch": 3.23,
1723
+ "learning_rate": 3.926893230684293e-05,
1724
+ "loss": 0.3871,
1725
+ "step": 140500
1726
+ },
1727
+ {
1728
+ "epoch": 3.24,
1729
+ "learning_rate": 3.923059227678435e-05,
1730
+ "loss": 0.3835,
1731
+ "step": 141000
1732
+ },
1733
+ {
1734
+ "epoch": 3.25,
1735
+ "learning_rate": 3.9192252246725766e-05,
1736
+ "loss": 0.3871,
1737
+ "step": 141500
1738
+ },
1739
+ {
1740
+ "epoch": 3.26,
1741
+ "learning_rate": 3.915391221666718e-05,
1742
+ "loss": 0.3803,
1743
+ "step": 142000
1744
+ },
1745
+ {
1746
+ "epoch": 3.28,
1747
+ "learning_rate": 3.9115572186608594e-05,
1748
+ "loss": 0.3776,
1749
+ "step": 142500
1750
+ },
1751
+ {
1752
+ "epoch": 3.29,
1753
+ "learning_rate": 3.907723215655001e-05,
1754
+ "loss": 0.3742,
1755
+ "step": 143000
1756
+ },
1757
+ {
1758
+ "epoch": 3.3,
1759
+ "learning_rate": 3.903889212649143e-05,
1760
+ "loss": 0.3778,
1761
+ "step": 143500
1762
+ },
1763
+ {
1764
+ "epoch": 3.31,
1765
+ "learning_rate": 3.900055209643285e-05,
1766
+ "loss": 0.3759,
1767
+ "step": 144000
1768
+ },
1769
+ {
1770
+ "epoch": 3.32,
1771
+ "learning_rate": 3.896221206637426e-05,
1772
+ "loss": 0.3756,
1773
+ "step": 144500
1774
+ },
1775
+ {
1776
+ "epoch": 3.33,
1777
+ "learning_rate": 3.89239487163758e-05,
1778
+ "loss": 0.3737,
1779
+ "step": 145000
1780
+ },
1781
+ {
1782
+ "epoch": 3.34,
1783
+ "learning_rate": 3.888560868631721e-05,
1784
+ "loss": 0.3749,
1785
+ "step": 145500
1786
+ },
1787
+ {
1788
+ "epoch": 3.36,
1789
+ "learning_rate": 3.884734533631875e-05,
1790
+ "loss": 0.3701,
1791
+ "step": 146000
1792
+ },
1793
+ {
1794
+ "epoch": 3.37,
1795
+ "learning_rate": 3.880900530626016e-05,
1796
+ "loss": 0.3736,
1797
+ "step": 146500
1798
+ },
1799
+ {
1800
+ "epoch": 3.38,
1801
+ "learning_rate": 3.8770665276201576e-05,
1802
+ "loss": 0.3706,
1803
+ "step": 147000
1804
+ },
1805
+ {
1806
+ "epoch": 3.39,
1807
+ "learning_rate": 3.8732325246143e-05,
1808
+ "loss": 0.3743,
1809
+ "step": 147500
1810
+ },
1811
+ {
1812
+ "epoch": 3.4,
1813
+ "learning_rate": 3.869398521608441e-05,
1814
+ "loss": 0.3665,
1815
+ "step": 148000
1816
+ },
1817
+ {
1818
+ "epoch": 3.41,
1819
+ "learning_rate": 3.865564518602583e-05,
1820
+ "loss": 0.3781,
1821
+ "step": 148500
1822
+ },
1823
+ {
1824
+ "epoch": 3.42,
1825
+ "learning_rate": 3.861738183602736e-05,
1826
+ "loss": 0.3591,
1827
+ "step": 149000
1828
+ },
1829
+ {
1830
+ "epoch": 3.44,
1831
+ "learning_rate": 3.857904180596878e-05,
1832
+ "loss": 0.365,
1833
+ "step": 149500
1834
+ },
1835
+ {
1836
+ "epoch": 3.45,
1837
+ "learning_rate": 3.854070177591019e-05,
1838
+ "loss": 0.3648,
1839
+ "step": 150000
1840
+ },
1841
+ {
1842
+ "epoch": 3.46,
1843
+ "learning_rate": 3.8502361745851615e-05,
1844
+ "loss": 0.3635,
1845
+ "step": 150500
1846
+ },
1847
+ {
1848
+ "epoch": 3.47,
1849
+ "learning_rate": 3.8464021715793026e-05,
1850
+ "loss": 0.3658,
1851
+ "step": 151000
1852
+ },
1853
+ {
1854
+ "epoch": 3.48,
1855
+ "learning_rate": 3.8425681685734444e-05,
1856
+ "loss": 0.3666,
1857
+ "step": 151500
1858
+ },
1859
+ {
1860
+ "epoch": 3.49,
1861
+ "learning_rate": 3.838734165567586e-05,
1862
+ "loss": 0.3579,
1863
+ "step": 152000
1864
+ },
1865
+ {
1866
+ "epoch": 3.51,
1867
+ "learning_rate": 3.834900162561728e-05,
1868
+ "loss": 0.3582,
1869
+ "step": 152500
1870
+ },
1871
+ {
1872
+ "epoch": 3.52,
1873
+ "learning_rate": 3.8310661595558697e-05,
1874
+ "loss": 0.3554,
1875
+ "step": 153000
1876
+ },
1877
+ {
1878
+ "epoch": 3.53,
1879
+ "learning_rate": 3.8272474925620344e-05,
1880
+ "loss": 0.3645,
1881
+ "step": 153500
1882
+ },
1883
+ {
1884
+ "epoch": 3.54,
1885
+ "learning_rate": 3.823421157562188e-05,
1886
+ "loss": 0.3605,
1887
+ "step": 154000
1888
+ },
1889
+ {
1890
+ "epoch": 3.55,
1891
+ "learning_rate": 3.8195871545563295e-05,
1892
+ "loss": 0.3657,
1893
+ "step": 154500
1894
+ },
1895
+ {
1896
+ "epoch": 3.56,
1897
+ "learning_rate": 3.8157531515504706e-05,
1898
+ "loss": 0.362,
1899
+ "step": 155000
1900
+ },
1901
+ {
1902
+ "epoch": 3.57,
1903
+ "learning_rate": 3.811919148544612e-05,
1904
+ "loss": 0.3585,
1905
+ "step": 155500
1906
+ },
1907
+ {
1908
+ "epoch": 3.59,
1909
+ "learning_rate": 3.808085145538755e-05,
1910
+ "loss": 0.3606,
1911
+ "step": 156000
1912
+ },
1913
+ {
1914
+ "epoch": 3.6,
1915
+ "learning_rate": 3.8042664785449196e-05,
1916
+ "loss": 0.3539,
1917
+ "step": 156500
1918
+ },
1919
+ {
1920
+ "epoch": 3.61,
1921
+ "learning_rate": 3.8004324755390606e-05,
1922
+ "loss": 0.3543,
1923
+ "step": 157000
1924
+ },
1925
+ {
1926
+ "epoch": 3.62,
1927
+ "learning_rate": 3.7966061405392146e-05,
1928
+ "loss": 0.3599,
1929
+ "step": 157500
1930
+ },
1931
+ {
1932
+ "epoch": 3.63,
1933
+ "learning_rate": 3.792772137533356e-05,
1934
+ "loss": 0.3613,
1935
+ "step": 158000
1936
+ },
1937
+ {
1938
+ "epoch": 3.64,
1939
+ "learning_rate": 3.7889381345274975e-05,
1940
+ "loss": 0.3616,
1941
+ "step": 158500
1942
+ },
1943
+ {
1944
+ "epoch": 3.65,
1945
+ "learning_rate": 3.78510413152164e-05,
1946
+ "loss": 0.357,
1947
+ "step": 159000
1948
+ },
1949
+ {
1950
+ "epoch": 3.67,
1951
+ "learning_rate": 3.781270128515781e-05,
1952
+ "loss": 0.3444,
1953
+ "step": 159500
1954
+ },
1955
+ {
1956
+ "epoch": 3.68,
1957
+ "learning_rate": 3.777436125509923e-05,
1958
+ "loss": 0.35,
1959
+ "step": 160000
1960
+ },
1961
+ {
1962
+ "epoch": 3.69,
1963
+ "learning_rate": 3.773602122504064e-05,
1964
+ "loss": 0.3515,
1965
+ "step": 160500
1966
+ },
1967
+ {
1968
+ "epoch": 3.7,
1969
+ "learning_rate": 3.769768119498206e-05,
1970
+ "loss": 0.3571,
1971
+ "step": 161000
1972
+ },
1973
+ {
1974
+ "epoch": 3.71,
1975
+ "learning_rate": 3.7659341164923474e-05,
1976
+ "loss": 0.3484,
1977
+ "step": 161500
1978
+ },
1979
+ {
1980
+ "epoch": 3.72,
1981
+ "learning_rate": 3.762100113486489e-05,
1982
+ "loss": 0.3467,
1983
+ "step": 162000
1984
+ },
1985
+ {
1986
+ "epoch": 3.74,
1987
+ "learning_rate": 3.758266110480631e-05,
1988
+ "loss": 0.3489,
1989
+ "step": 162500
1990
+ },
1991
+ {
1992
+ "epoch": 3.75,
1993
+ "learning_rate": 3.754432107474773e-05,
1994
+ "loss": 0.3505,
1995
+ "step": 163000
1996
+ },
1997
+ {
1998
+ "epoch": 3.76,
1999
+ "learning_rate": 3.7505981044689144e-05,
2000
+ "loss": 0.351,
2001
+ "step": 163500
2002
+ },
2003
+ {
2004
+ "epoch": 3.77,
2005
+ "learning_rate": 3.7467641014630555e-05,
2006
+ "loss": 0.3486,
2007
+ "step": 164000
2008
+ },
2009
+ {
2010
+ "epoch": 3.78,
2011
+ "learning_rate": 3.7429377664632095e-05,
2012
+ "loss": 0.3394,
2013
+ "step": 164500
2014
+ },
2015
+ {
2016
+ "epoch": 3.79,
2017
+ "learning_rate": 3.7391037634573506e-05,
2018
+ "loss": 0.3445,
2019
+ "step": 165000
2020
+ },
2021
+ {
2022
+ "epoch": 3.8,
2023
+ "learning_rate": 3.735269760451492e-05,
2024
+ "loss": 0.3417,
2025
+ "step": 165500
2026
+ },
2027
+ {
2028
+ "epoch": 3.82,
2029
+ "learning_rate": 3.731435757445634e-05,
2030
+ "loss": 0.3492,
2031
+ "step": 166000
2032
+ },
2033
+ {
2034
+ "epoch": 3.83,
2035
+ "learning_rate": 3.727601754439776e-05,
2036
+ "loss": 0.3519,
2037
+ "step": 166500
2038
+ },
2039
+ {
2040
+ "epoch": 3.84,
2041
+ "learning_rate": 3.7237677514339176e-05,
2042
+ "loss": 0.3484,
2043
+ "step": 167000
2044
+ },
2045
+ {
2046
+ "epoch": 3.85,
2047
+ "learning_rate": 3.719933748428059e-05,
2048
+ "loss": 0.3446,
2049
+ "step": 167500
2050
+ },
2051
+ {
2052
+ "epoch": 3.86,
2053
+ "learning_rate": 3.7160997454222005e-05,
2054
+ "loss": 0.3499,
2055
+ "step": 168000
2056
+ },
2057
+ {
2058
+ "epoch": 3.87,
2059
+ "learning_rate": 3.712265742416342e-05,
2060
+ "loss": 0.3465,
2061
+ "step": 168500
2062
+ },
2063
+ {
2064
+ "epoch": 3.88,
2065
+ "learning_rate": 3.708431739410484e-05,
2066
+ "loss": 0.3431,
2067
+ "step": 169000
2068
+ },
2069
+ {
2070
+ "epoch": 3.9,
2071
+ "learning_rate": 3.704597736404625e-05,
2072
+ "loss": 0.3379,
2073
+ "step": 169500
2074
+ },
2075
+ {
2076
+ "epoch": 3.91,
2077
+ "learning_rate": 3.700771401404779e-05,
2078
+ "loss": 0.3402,
2079
+ "step": 170000
2080
+ },
2081
+ {
2082
+ "epoch": 3.92,
2083
+ "learning_rate": 3.69693739839892e-05,
2084
+ "loss": 0.3379,
2085
+ "step": 170500
2086
+ },
2087
+ {
2088
+ "epoch": 3.93,
2089
+ "learning_rate": 3.6931187314050856e-05,
2090
+ "loss": 0.3468,
2091
+ "step": 171000
2092
+ },
2093
+ {
2094
+ "epoch": 3.94,
2095
+ "learning_rate": 3.6892847283992274e-05,
2096
+ "loss": 0.3433,
2097
+ "step": 171500
2098
+ },
2099
+ {
2100
+ "epoch": 3.95,
2101
+ "learning_rate": 3.685450725393369e-05,
2102
+ "loss": 0.3305,
2103
+ "step": 172000
2104
+ },
2105
+ {
2106
+ "epoch": 3.97,
2107
+ "learning_rate": 3.68161672238751e-05,
2108
+ "loss": 0.3373,
2109
+ "step": 172500
2110
+ },
2111
+ {
2112
+ "epoch": 3.98,
2113
+ "learning_rate": 3.677782719381652e-05,
2114
+ "loss": 0.3426,
2115
+ "step": 173000
2116
+ },
2117
+ {
2118
+ "epoch": 3.99,
2119
+ "learning_rate": 3.673948716375794e-05,
2120
+ "loss": 0.3387,
2121
+ "step": 173500
2122
+ },
2123
+ {
2124
+ "epoch": 4.0,
2125
+ "learning_rate": 3.6701147133699355e-05,
2126
+ "loss": 0.3407,
2127
+ "step": 174000
2128
+ },
2129
+ {
2130
+ "epoch": 4.0,
2131
+ "eval_bleu": 58.3826,
2132
+ "eval_gen_len": 17.1623,
2133
+ "eval_loss": 0.7513108253479004,
2134
+ "eval_runtime": 7458.9074,
2135
+ "eval_samples_per_second": 11.665,
2136
+ "eval_steps_per_second": 1.458,
2137
+ "step": 174018
2138
+ },
2139
+ {
2140
+ "epoch": 4.01,
2141
+ "learning_rate": 3.666280710364077e-05,
2142
+ "loss": 0.3395,
2143
+ "step": 174500
2144
+ },
2145
+ {
2146
+ "epoch": 4.02,
2147
+ "learning_rate": 3.6624467073582184e-05,
2148
+ "loss": 0.3242,
2149
+ "step": 175000
2150
+ },
2151
+ {
2152
+ "epoch": 4.03,
2153
+ "learning_rate": 3.658612704352361e-05,
2154
+ "loss": 0.3246,
2155
+ "step": 175500
2156
+ },
2157
+ {
2158
+ "epoch": 4.05,
2159
+ "learning_rate": 3.654778701346502e-05,
2160
+ "loss": 0.3132,
2161
+ "step": 176000
2162
+ },
2163
+ {
2164
+ "epoch": 4.06,
2165
+ "learning_rate": 3.6509446983406437e-05,
2166
+ "loss": 0.3129,
2167
+ "step": 176500
2168
+ },
2169
+ {
2170
+ "epoch": 4.07,
2171
+ "learning_rate": 3.6471106953347854e-05,
2172
+ "loss": 0.3162,
2173
+ "step": 177000
2174
+ },
2175
+ {
2176
+ "epoch": 4.08,
2177
+ "learning_rate": 3.643276692328927e-05,
2178
+ "loss": 0.316,
2179
+ "step": 177500
2180
+ },
2181
+ {
2182
+ "epoch": 4.09,
2183
+ "learning_rate": 3.639442689323069e-05,
2184
+ "loss": 0.3088,
2185
+ "step": 178000
2186
+ },
2187
+ {
2188
+ "epoch": 4.1,
2189
+ "learning_rate": 3.63560868631721e-05,
2190
+ "loss": 0.3048,
2191
+ "step": 178500
2192
+ },
2193
+ {
2194
+ "epoch": 4.11,
2195
+ "learning_rate": 3.631774683311352e-05,
2196
+ "loss": 0.3126,
2197
+ "step": 179000
2198
+ },
2199
+ {
2200
+ "epoch": 4.13,
2201
+ "learning_rate": 3.627948348311505e-05,
2202
+ "loss": 0.3091,
2203
+ "step": 179500
2204
+ },
2205
+ {
2206
+ "epoch": 4.14,
2207
+ "learning_rate": 3.624114345305647e-05,
2208
+ "loss": 0.3046,
2209
+ "step": 180000
2210
+ },
2211
+ {
2212
+ "epoch": 4.15,
2213
+ "learning_rate": 3.6202803422997886e-05,
2214
+ "loss": 0.3008,
2215
+ "step": 180500
2216
+ },
2217
+ {
2218
+ "epoch": 4.16,
2219
+ "learning_rate": 3.6164463392939304e-05,
2220
+ "loss": 0.3037,
2221
+ "step": 181000
2222
+ },
2223
+ {
2224
+ "epoch": 4.17,
2225
+ "learning_rate": 3.612612336288072e-05,
2226
+ "loss": 0.3034,
2227
+ "step": 181500
2228
+ },
2229
+ {
2230
+ "epoch": 4.18,
2231
+ "learning_rate": 3.608778333282213e-05,
2232
+ "loss": 0.3025,
2233
+ "step": 182000
2234
+ },
2235
+ {
2236
+ "epoch": 4.19,
2237
+ "learning_rate": 3.604944330276355e-05,
2238
+ "loss": 0.3021,
2239
+ "step": 182500
2240
+ },
2241
+ {
2242
+ "epoch": 4.21,
2243
+ "learning_rate": 3.601117995276508e-05,
2244
+ "loss": 0.3015,
2245
+ "step": 183000
2246
+ },
2247
+ {
2248
+ "epoch": 4.22,
2249
+ "learning_rate": 3.59728399227065e-05,
2250
+ "loss": 0.2977,
2251
+ "step": 183500
2252
+ },
2253
+ {
2254
+ "epoch": 4.23,
2255
+ "learning_rate": 3.593449989264792e-05,
2256
+ "loss": 0.2999,
2257
+ "step": 184000
2258
+ },
2259
+ {
2260
+ "epoch": 4.24,
2261
+ "learning_rate": 3.5896159862589336e-05,
2262
+ "loss": 0.2953,
2263
+ "step": 184500
2264
+ },
2265
+ {
2266
+ "epoch": 4.25,
2267
+ "learning_rate": 3.5857819832530747e-05,
2268
+ "loss": 0.2992,
2269
+ "step": 185000
2270
+ },
2271
+ {
2272
+ "epoch": 4.26,
2273
+ "learning_rate": 3.5819479802472164e-05,
2274
+ "loss": 0.2927,
2275
+ "step": 185500
2276
+ },
2277
+ {
2278
+ "epoch": 4.28,
2279
+ "learning_rate": 3.578113977241359e-05,
2280
+ "loss": 0.2929,
2281
+ "step": 186000
2282
+ },
2283
+ {
2284
+ "epoch": 4.29,
2285
+ "learning_rate": 3.5742799742355e-05,
2286
+ "loss": 0.2919,
2287
+ "step": 186500
2288
+ },
2289
+ {
2290
+ "epoch": 4.3,
2291
+ "learning_rate": 3.570445971229642e-05,
2292
+ "loss": 0.2953,
2293
+ "step": 187000
2294
+ },
2295
+ {
2296
+ "epoch": 4.31,
2297
+ "learning_rate": 3.566611968223783e-05,
2298
+ "loss": 0.2962,
2299
+ "step": 187500
2300
+ },
2301
+ {
2302
+ "epoch": 4.32,
2303
+ "learning_rate": 3.562777965217925e-05,
2304
+ "loss": 0.2908,
2305
+ "step": 188000
2306
+ },
2307
+ {
2308
+ "epoch": 4.33,
2309
+ "learning_rate": 3.558943962212066e-05,
2310
+ "loss": 0.2918,
2311
+ "step": 188500
2312
+ },
2313
+ {
2314
+ "epoch": 4.34,
2315
+ "learning_rate": 3.555109959206208e-05,
2316
+ "loss": 0.2898,
2317
+ "step": 189000
2318
+ },
2319
+ {
2320
+ "epoch": 4.36,
2321
+ "learning_rate": 3.5512836242063614e-05,
2322
+ "loss": 0.2876,
2323
+ "step": 189500
2324
+ },
2325
+ {
2326
+ "epoch": 4.37,
2327
+ "learning_rate": 3.547449621200503e-05,
2328
+ "loss": 0.2893,
2329
+ "step": 190000
2330
+ },
2331
+ {
2332
+ "epoch": 4.38,
2333
+ "learning_rate": 3.543615618194645e-05,
2334
+ "loss": 0.2857,
2335
+ "step": 190500
2336
+ },
2337
+ {
2338
+ "epoch": 4.39,
2339
+ "learning_rate": 3.539781615188787e-05,
2340
+ "loss": 0.2898,
2341
+ "step": 191000
2342
+ },
2343
+ {
2344
+ "epoch": 4.4,
2345
+ "learning_rate": 3.53595528018894e-05,
2346
+ "loss": 0.2834,
2347
+ "step": 191500
2348
+ },
2349
+ {
2350
+ "epoch": 4.41,
2351
+ "learning_rate": 3.532121277183082e-05,
2352
+ "loss": 0.289,
2353
+ "step": 192000
2354
+ },
2355
+ {
2356
+ "epoch": 4.42,
2357
+ "learning_rate": 3.528294942183235e-05,
2358
+ "loss": 0.2804,
2359
+ "step": 192500
2360
+ },
2361
+ {
2362
+ "epoch": 4.44,
2363
+ "learning_rate": 3.524460939177376e-05,
2364
+ "loss": 0.2839,
2365
+ "step": 193000
2366
+ },
2367
+ {
2368
+ "epoch": 4.45,
2369
+ "learning_rate": 3.5206269361715185e-05,
2370
+ "loss": 0.2785,
2371
+ "step": 193500
2372
+ },
2373
+ {
2374
+ "epoch": 4.46,
2375
+ "learning_rate": 3.5167929331656596e-05,
2376
+ "loss": 0.2802,
2377
+ "step": 194000
2378
+ },
2379
+ {
2380
+ "epoch": 4.47,
2381
+ "learning_rate": 3.5129589301598014e-05,
2382
+ "loss": 0.286,
2383
+ "step": 194500
2384
+ },
2385
+ {
2386
+ "epoch": 4.48,
2387
+ "learning_rate": 3.509124927153943e-05,
2388
+ "loss": 0.2859,
2389
+ "step": 195000
2390
+ },
2391
+ {
2392
+ "epoch": 4.49,
2393
+ "learning_rate": 3.5052985921540964e-05,
2394
+ "loss": 0.2774,
2395
+ "step": 195500
2396
+ },
2397
+ {
2398
+ "epoch": 4.51,
2399
+ "learning_rate": 3.501479925160261e-05,
2400
+ "loss": 0.2788,
2401
+ "step": 196000
2402
+ },
2403
+ {
2404
+ "epoch": 4.52,
2405
+ "learning_rate": 3.4976459221544036e-05,
2406
+ "loss": 0.2754,
2407
+ "step": 196500
2408
+ },
2409
+ {
2410
+ "epoch": 4.53,
2411
+ "learning_rate": 3.493811919148545e-05,
2412
+ "loss": 0.282,
2413
+ "step": 197000
2414
+ },
2415
+ {
2416
+ "epoch": 4.54,
2417
+ "learning_rate": 3.4899779161426865e-05,
2418
+ "loss": 0.2809,
2419
+ "step": 197500
2420
+ },
2421
+ {
2422
+ "epoch": 4.55,
2423
+ "learning_rate": 3.48615158114284e-05,
2424
+ "loss": 0.2872,
2425
+ "step": 198000
2426
+ },
2427
+ {
2428
+ "epoch": 4.56,
2429
+ "learning_rate": 3.4823175781369815e-05,
2430
+ "loss": 0.2806,
2431
+ "step": 198500
2432
+ },
2433
+ {
2434
+ "epoch": 4.57,
2435
+ "learning_rate": 3.4784835751311226e-05,
2436
+ "loss": 0.2746,
2437
+ "step": 199000
2438
+ },
2439
+ {
2440
+ "epoch": 4.59,
2441
+ "learning_rate": 3.474649572125265e-05,
2442
+ "loss": 0.2793,
2443
+ "step": 199500
2444
+ },
2445
+ {
2446
+ "epoch": 4.6,
2447
+ "learning_rate": 3.4708232371254177e-05,
2448
+ "loss": 0.2781,
2449
+ "step": 200000
2450
+ },
2451
+ {
2452
+ "epoch": 4.61,
2453
+ "learning_rate": 3.4669969021255716e-05,
2454
+ "loss": 0.2773,
2455
+ "step": 200500
2456
+ },
2457
+ {
2458
+ "epoch": 4.62,
2459
+ "learning_rate": 3.463162899119713e-05,
2460
+ "loss": 0.2782,
2461
+ "step": 201000
2462
+ },
2463
+ {
2464
+ "epoch": 4.63,
2465
+ "learning_rate": 3.4593288961138545e-05,
2466
+ "loss": 0.2777,
2467
+ "step": 201500
2468
+ },
2469
+ {
2470
+ "epoch": 4.64,
2471
+ "learning_rate": 3.455494893107996e-05,
2472
+ "loss": 0.2792,
2473
+ "step": 202000
2474
+ },
2475
+ {
2476
+ "epoch": 4.65,
2477
+ "learning_rate": 3.451660890102138e-05,
2478
+ "loss": 0.276,
2479
+ "step": 202500
2480
+ },
2481
+ {
2482
+ "epoch": 4.67,
2483
+ "learning_rate": 3.44782688709628e-05,
2484
+ "loss": 0.2663,
2485
+ "step": 203000
2486
+ },
2487
+ {
2488
+ "epoch": 4.68,
2489
+ "learning_rate": 3.4439928840904215e-05,
2490
+ "loss": 0.2697,
2491
+ "step": 203500
2492
+ },
2493
+ {
2494
+ "epoch": 4.69,
2495
+ "learning_rate": 3.440158881084563e-05,
2496
+ "loss": 0.2733,
2497
+ "step": 204000
2498
+ },
2499
+ {
2500
+ "epoch": 4.7,
2501
+ "learning_rate": 3.4363248780787044e-05,
2502
+ "loss": 0.2759,
2503
+ "step": 204500
2504
+ },
2505
+ {
2506
+ "epoch": 4.71,
2507
+ "learning_rate": 3.432490875072846e-05,
2508
+ "loss": 0.271,
2509
+ "step": 205000
2510
+ },
2511
+ {
2512
+ "epoch": 4.72,
2513
+ "learning_rate": 3.428656872066988e-05,
2514
+ "loss": 0.2669,
2515
+ "step": 205500
2516
+ },
2517
+ {
2518
+ "epoch": 4.74,
2519
+ "learning_rate": 3.42482286906113e-05,
2520
+ "loss": 0.268,
2521
+ "step": 206000
2522
+ },
2523
+ {
2524
+ "epoch": 4.75,
2525
+ "learning_rate": 3.4209888660552714e-05,
2526
+ "loss": 0.27,
2527
+ "step": 206500
2528
+ },
2529
+ {
2530
+ "epoch": 4.76,
2531
+ "learning_rate": 3.4171548630494125e-05,
2532
+ "loss": 0.2719,
2533
+ "step": 207000
2534
+ },
2535
+ {
2536
+ "epoch": 4.77,
2537
+ "learning_rate": 3.413320860043554e-05,
2538
+ "loss": 0.2723,
2539
+ "step": 207500
2540
+ },
2541
+ {
2542
+ "epoch": 4.78,
2543
+ "learning_rate": 3.409486857037696e-05,
2544
+ "loss": 0.2612,
2545
+ "step": 208000
2546
+ },
2547
+ {
2548
+ "epoch": 4.79,
2549
+ "learning_rate": 3.405652854031838e-05,
2550
+ "loss": 0.2661,
2551
+ "step": 208500
2552
+ },
2553
+ {
2554
+ "epoch": 4.8,
2555
+ "learning_rate": 3.4018188510259796e-05,
2556
+ "loss": 0.2641,
2557
+ "step": 209000
2558
+ },
2559
+ {
2560
+ "epoch": 4.82,
2561
+ "learning_rate": 3.397984848020121e-05,
2562
+ "loss": 0.2685,
2563
+ "step": 209500
2564
+ },
2565
+ {
2566
+ "epoch": 4.83,
2567
+ "learning_rate": 3.394150845014263e-05,
2568
+ "loss": 0.2716,
2569
+ "step": 210000
2570
+ },
2571
+ {
2572
+ "epoch": 4.84,
2573
+ "learning_rate": 3.390324510014416e-05,
2574
+ "loss": 0.2693,
2575
+ "step": 210500
2576
+ },
2577
+ {
2578
+ "epoch": 4.85,
2579
+ "learning_rate": 3.386490507008558e-05,
2580
+ "loss": 0.2686,
2581
+ "step": 211000
2582
+ },
2583
+ {
2584
+ "epoch": 4.86,
2585
+ "learning_rate": 3.382656504002699e-05,
2586
+ "loss": 0.2743,
2587
+ "step": 211500
2588
+ },
2589
+ {
2590
+ "epoch": 4.87,
2591
+ "learning_rate": 3.378822500996841e-05,
2592
+ "loss": 0.2672,
2593
+ "step": 212000
2594
+ },
2595
+ {
2596
+ "epoch": 4.88,
2597
+ "learning_rate": 3.374988497990982e-05,
2598
+ "loss": 0.2656,
2599
+ "step": 212500
2600
+ },
2601
+ {
2602
+ "epoch": 4.9,
2603
+ "learning_rate": 3.3711544949851245e-05,
2604
+ "loss": 0.2578,
2605
+ "step": 213000
2606
+ },
2607
+ {
2608
+ "epoch": 4.91,
2609
+ "learning_rate": 3.367320491979266e-05,
2610
+ "loss": 0.263,
2611
+ "step": 213500
2612
+ },
2613
+ {
2614
+ "epoch": 4.92,
2615
+ "learning_rate": 3.3634864889734074e-05,
2616
+ "loss": 0.2642,
2617
+ "step": 214000
2618
+ },
2619
+ {
2620
+ "epoch": 4.93,
2621
+ "learning_rate": 3.3596601539735613e-05,
2622
+ "loss": 0.2681,
2623
+ "step": 214500
2624
+ },
2625
+ {
2626
+ "epoch": 4.94,
2627
+ "learning_rate": 3.3558261509677024e-05,
2628
+ "loss": 0.2632,
2629
+ "step": 215000
2630
+ },
2631
+ {
2632
+ "epoch": 4.95,
2633
+ "learning_rate": 3.351992147961844e-05,
2634
+ "loss": 0.2538,
2635
+ "step": 215500
2636
+ },
2637
+ {
2638
+ "epoch": 4.97,
2639
+ "learning_rate": 3.348158144955986e-05,
2640
+ "loss": 0.2608,
2641
+ "step": 216000
2642
+ },
2643
+ {
2644
+ "epoch": 4.98,
2645
+ "learning_rate": 3.344324141950128e-05,
2646
+ "loss": 0.2641,
2647
+ "step": 216500
2648
+ },
2649
+ {
2650
+ "epoch": 4.99,
2651
+ "learning_rate": 3.340490138944269e-05,
2652
+ "loss": 0.2592,
2653
+ "step": 217000
2654
+ },
2655
+ {
2656
+ "epoch": 5.0,
2657
+ "learning_rate": 3.336663803944423e-05,
2658
+ "loss": 0.2605,
2659
+ "step": 217500
2660
+ },
2661
+ {
2662
+ "epoch": 5.0,
2663
+ "eval_bleu": 58.7238,
2664
+ "eval_gen_len": 17.0888,
2665
+ "eval_loss": 0.8122135400772095,
2666
+ "eval_runtime": 7422.7798,
2667
+ "eval_samples_per_second": 11.722,
2668
+ "eval_steps_per_second": 1.465,
2669
+ "step": 217522
2670
+ }
2671
+ ],
2672
+ "logging_steps": 500,
2673
+ "max_steps": 652560,
2674
+ "num_train_epochs": 15,
2675
+ "save_steps": 500,
2676
+ "total_flos": 7.542391759939043e+18,
2677
+ "trial_name": null,
2678
+ "trial_params": null
2679
+ }
checkpoint-217522/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44b731ecd386cb51a958b6c8f191496c616dc6e80b60cf4e62402772da333973
3
+ size 4792
checkpoint-87009/config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
3
+ "_num_labels": 3,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "relu",
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": 2,
21
+ "dropout": 0.1,
22
+ "early_stopping": true,
23
+ "encoder_attention_heads": 16,
24
+ "encoder_ffn_dim": 4096,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 12,
27
+ "eos_token_id": 2,
28
+ "forced_bos_token_id": 250014,
29
+ "forced_eos_token_id": 2,
30
+ "gradient_checkpointing": false,
31
+ "id2label": {
32
+ "0": "LABEL_0",
33
+ "1": "LABEL_1",
34
+ "2": "LABEL_2"
35
+ },
36
+ "init_std": 0.02,
37
+ "is_encoder_decoder": true,
38
+ "label2id": {
39
+ "LABEL_0": 0,
40
+ "LABEL_1": 1,
41
+ "LABEL_2": 2
42
+ },
43
+ "max_length": 200,
44
+ "max_position_embeddings": 1024,
45
+ "model_type": "mbart",
46
+ "normalize_before": true,
47
+ "normalize_embedding": true,
48
+ "num_beams": 5,
49
+ "num_hidden_layers": 12,
50
+ "output_past": true,
51
+ "pad_token_id": 1,
52
+ "scale_embedding": true,
53
+ "static_position_embeddings": false,
54
+ "tokenizer_class": "MBart50Tokenizer",
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.35.2",
57
+ "use_cache": true,
58
+ "vocab_size": 250054
59
+ }
checkpoint-87009/generation_config.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 0,
3
+ "decoder_start_token_id": 2,
4
+ "early_stopping": true,
5
+ "eos_token_id": 2,
6
+ "forced_bos_token_id": 250014,
7
+ "forced_eos_token_id": 2,
8
+ "max_length": 200,
9
+ "num_beams": 5,
10
+ "pad_token_id": 1,
11
+ "transformers_version": "4.35.2"
12
+ }
checkpoint-87009/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97f359d7b7a214b7d33f8b720df9fb4977922176e4c5fa16d91238daeb1027a2
3
+ size 2444578688
checkpoint-87009/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c38d5139d9300234624ea3f83ee80526c18202627fda4bc5adc19dc63638eaa0
3
+ size 4887473903
checkpoint-87009/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:46166c012e316e435016d0e3f3d98cb393b668a8468e9d0981d4420e442753f9
3
+ size 15024
checkpoint-87009/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:641db8587b2461e871f0234214d253811f15d12f7109d1fd4102ac1357a457f5
3
+ size 15024
checkpoint-87009/rng_state_2.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:66811d476fe04b02198ecdb59590d27c31d02a1b034bf1f254d86ae3475bc7b4
3
+ size 15024
checkpoint-87009/rng_state_3.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c62b319ce260583a0b18e19d711c49ef66d57af01315b2a4451ac0a4b4747245
3
+ size 15024
checkpoint-87009/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:555882394d7aacd99910ea6cc71bbe421925abef4fb1f1b252c9c17ae6ff5f5c
3
+ size 1064
checkpoint-87009/sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
checkpoint-87009/special_tokens_map.json ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "ar_AR",
4
+ "cs_CZ",
5
+ "de_DE",
6
+ "en_XX",
7
+ "es_XX",
8
+ "et_EE",
9
+ "fi_FI",
10
+ "fr_XX",
11
+ "gu_IN",
12
+ "hi_IN",
13
+ "it_IT",
14
+ "ja_XX",
15
+ "kk_KZ",
16
+ "ko_KR",
17
+ "lt_LT",
18
+ "lv_LV",
19
+ "my_MM",
20
+ "ne_NP",
21
+ "nl_XX",
22
+ "ro_RO",
23
+ "ru_RU",
24
+ "si_LK",
25
+ "tr_TR",
26
+ "vi_VN",
27
+ "zh_CN",
28
+ "af_ZA",
29
+ "az_AZ",
30
+ "bn_IN",
31
+ "fa_IR",
32
+ "he_IL",
33
+ "hr_HR",
34
+ "id_ID",
35
+ "ka_GE",
36
+ "km_KH",
37
+ "mk_MK",
38
+ "ml_IN",
39
+ "mn_MN",
40
+ "mr_IN",
41
+ "pl_PL",
42
+ "ps_AF",
43
+ "pt_XX",
44
+ "sv_SE",
45
+ "sw_KE",
46
+ "ta_IN",
47
+ "te_IN",
48
+ "th_TH",
49
+ "tl_XX",
50
+ "uk_UA",
51
+ "ur_PK",
52
+ "xh_ZA",
53
+ "gl_ES",
54
+ "sl_SI"
55
+ ],
56
+ "bos_token": "<s>",
57
+ "cls_token": "<s>",
58
+ "eos_token": "</s>",
59
+ "mask_token": {
60
+ "content": "<mask>",
61
+ "lstrip": true,
62
+ "normalized": true,
63
+ "rstrip": false,
64
+ "single_word": false
65
+ },
66
+ "pad_token": "<pad>",
67
+ "sep_token": "</s>",
68
+ "unk_token": "<unk>"
69
+ }
checkpoint-87009/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:faf1fb0f45fdd5ae9fc44fafee492652a423fc749833ef57c2fa801358549d92
3
+ size 17110107
checkpoint-87009/tokenizer_config.json ADDED
@@ -0,0 +1,528 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "250001": {
36
+ "content": "ar_AR",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "250002": {
44
+ "content": "cs_CZ",
45
+ "lstrip": false,
46
+ "normalized": false,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": true
50
+ },
51
+ "250003": {
52
+ "content": "de_DE",
53
+ "lstrip": false,
54
+ "normalized": false,
55
+ "rstrip": false,
56
+ "single_word": false,
57
+ "special": true
58
+ },
59
+ "250004": {
60
+ "content": "en_XX",
61
+ "lstrip": false,
62
+ "normalized": false,
63
+ "rstrip": false,
64
+ "single_word": false,
65
+ "special": true
66
+ },
67
+ "250005": {
68
+ "content": "es_XX",
69
+ "lstrip": false,
70
+ "normalized": false,
71
+ "rstrip": false,
72
+ "single_word": false,
73
+ "special": true
74
+ },
75
+ "250006": {
76
+ "content": "et_EE",
77
+ "lstrip": false,
78
+ "normalized": false,
79
+ "rstrip": false,
80
+ "single_word": false,
81
+ "special": true
82
+ },
83
+ "250007": {
84
+ "content": "fi_FI",
85
+ "lstrip": false,
86
+ "normalized": false,
87
+ "rstrip": false,
88
+ "single_word": false,
89
+ "special": true
90
+ },
91
+ "250008": {
92
+ "content": "fr_XX",
93
+ "lstrip": false,
94
+ "normalized": false,
95
+ "rstrip": false,
96
+ "single_word": false,
97
+ "special": true
98
+ },
99
+ "250009": {
100
+ "content": "gu_IN",
101
+ "lstrip": false,
102
+ "normalized": false,
103
+ "rstrip": false,
104
+ "single_word": false,
105
+ "special": true
106
+ },
107
+ "250010": {
108
+ "content": "hi_IN",
109
+ "lstrip": false,
110
+ "normalized": false,
111
+ "rstrip": false,
112
+ "single_word": false,
113
+ "special": true
114
+ },
115
+ "250011": {
116
+ "content": "it_IT",
117
+ "lstrip": false,
118
+ "normalized": false,
119
+ "rstrip": false,
120
+ "single_word": false,
121
+ "special": true
122
+ },
123
+ "250012": {
124
+ "content": "ja_XX",
125
+ "lstrip": false,
126
+ "normalized": false,
127
+ "rstrip": false,
128
+ "single_word": false,
129
+ "special": true
130
+ },
131
+ "250013": {
132
+ "content": "kk_KZ",
133
+ "lstrip": false,
134
+ "normalized": false,
135
+ "rstrip": false,
136
+ "single_word": false,
137
+ "special": true
138
+ },
139
+ "250014": {
140
+ "content": "ko_KR",
141
+ "lstrip": false,
142
+ "normalized": false,
143
+ "rstrip": false,
144
+ "single_word": false,
145
+ "special": true
146
+ },
147
+ "250015": {
148
+ "content": "lt_LT",
149
+ "lstrip": false,
150
+ "normalized": false,
151
+ "rstrip": false,
152
+ "single_word": false,
153
+ "special": true
154
+ },
155
+ "250016": {
156
+ "content": "lv_LV",
157
+ "lstrip": false,
158
+ "normalized": false,
159
+ "rstrip": false,
160
+ "single_word": false,
161
+ "special": true
162
+ },
163
+ "250017": {
164
+ "content": "my_MM",
165
+ "lstrip": false,
166
+ "normalized": false,
167
+ "rstrip": false,
168
+ "single_word": false,
169
+ "special": true
170
+ },
171
+ "250018": {
172
+ "content": "ne_NP",
173
+ "lstrip": false,
174
+ "normalized": false,
175
+ "rstrip": false,
176
+ "single_word": false,
177
+ "special": true
178
+ },
179
+ "250019": {
180
+ "content": "nl_XX",
181
+ "lstrip": false,
182
+ "normalized": false,
183
+ "rstrip": false,
184
+ "single_word": false,
185
+ "special": true
186
+ },
187
+ "250020": {
188
+ "content": "ro_RO",
189
+ "lstrip": false,
190
+ "normalized": false,
191
+ "rstrip": false,
192
+ "single_word": false,
193
+ "special": true
194
+ },
195
+ "250021": {
196
+ "content": "ru_RU",
197
+ "lstrip": false,
198
+ "normalized": false,
199
+ "rstrip": false,
200
+ "single_word": false,
201
+ "special": true
202
+ },
203
+ "250022": {
204
+ "content": "si_LK",
205
+ "lstrip": false,
206
+ "normalized": false,
207
+ "rstrip": false,
208
+ "single_word": false,
209
+ "special": true
210
+ },
211
+ "250023": {
212
+ "content": "tr_TR",
213
+ "lstrip": false,
214
+ "normalized": false,
215
+ "rstrip": false,
216
+ "single_word": false,
217
+ "special": true
218
+ },
219
+ "250024": {
220
+ "content": "vi_VN",
221
+ "lstrip": false,
222
+ "normalized": false,
223
+ "rstrip": false,
224
+ "single_word": false,
225
+ "special": true
226
+ },
227
+ "250025": {
228
+ "content": "zh_CN",
229
+ "lstrip": false,
230
+ "normalized": false,
231
+ "rstrip": false,
232
+ "single_word": false,
233
+ "special": true
234
+ },
235
+ "250026": {
236
+ "content": "af_ZA",
237
+ "lstrip": false,
238
+ "normalized": false,
239
+ "rstrip": false,
240
+ "single_word": false,
241
+ "special": true
242
+ },
243
+ "250027": {
244
+ "content": "az_AZ",
245
+ "lstrip": false,
246
+ "normalized": false,
247
+ "rstrip": false,
248
+ "single_word": false,
249
+ "special": true
250
+ },
251
+ "250028": {
252
+ "content": "bn_IN",
253
+ "lstrip": false,
254
+ "normalized": false,
255
+ "rstrip": false,
256
+ "single_word": false,
257
+ "special": true
258
+ },
259
+ "250029": {
260
+ "content": "fa_IR",
261
+ "lstrip": false,
262
+ "normalized": false,
263
+ "rstrip": false,
264
+ "single_word": false,
265
+ "special": true
266
+ },
267
+ "250030": {
268
+ "content": "he_IL",
269
+ "lstrip": false,
270
+ "normalized": false,
271
+ "rstrip": false,
272
+ "single_word": false,
273
+ "special": true
274
+ },
275
+ "250031": {
276
+ "content": "hr_HR",
277
+ "lstrip": false,
278
+ "normalized": false,
279
+ "rstrip": false,
280
+ "single_word": false,
281
+ "special": true
282
+ },
283
+ "250032": {
284
+ "content": "id_ID",
285
+ "lstrip": false,
286
+ "normalized": false,
287
+ "rstrip": false,
288
+ "single_word": false,
289
+ "special": true
290
+ },
291
+ "250033": {
292
+ "content": "ka_GE",
293
+ "lstrip": false,
294
+ "normalized": false,
295
+ "rstrip": false,
296
+ "single_word": false,
297
+ "special": true
298
+ },
299
+ "250034": {
300
+ "content": "km_KH",
301
+ "lstrip": false,
302
+ "normalized": false,
303
+ "rstrip": false,
304
+ "single_word": false,
305
+ "special": true
306
+ },
307
+ "250035": {
308
+ "content": "mk_MK",
309
+ "lstrip": false,
310
+ "normalized": false,
311
+ "rstrip": false,
312
+ "single_word": false,
313
+ "special": true
314
+ },
315
+ "250036": {
316
+ "content": "ml_IN",
317
+ "lstrip": false,
318
+ "normalized": false,
319
+ "rstrip": false,
320
+ "single_word": false,
321
+ "special": true
322
+ },
323
+ "250037": {
324
+ "content": "mn_MN",
325
+ "lstrip": false,
326
+ "normalized": false,
327
+ "rstrip": false,
328
+ "single_word": false,
329
+ "special": true
330
+ },
331
+ "250038": {
332
+ "content": "mr_IN",
333
+ "lstrip": false,
334
+ "normalized": false,
335
+ "rstrip": false,
336
+ "single_word": false,
337
+ "special": true
338
+ },
339
+ "250039": {
340
+ "content": "pl_PL",
341
+ "lstrip": false,
342
+ "normalized": false,
343
+ "rstrip": false,
344
+ "single_word": false,
345
+ "special": true
346
+ },
347
+ "250040": {
348
+ "content": "ps_AF",
349
+ "lstrip": false,
350
+ "normalized": false,
351
+ "rstrip": false,
352
+ "single_word": false,
353
+ "special": true
354
+ },
355
+ "250041": {
356
+ "content": "pt_XX",
357
+ "lstrip": false,
358
+ "normalized": false,
359
+ "rstrip": false,
360
+ "single_word": false,
361
+ "special": true
362
+ },
363
+ "250042": {
364
+ "content": "sv_SE",
365
+ "lstrip": false,
366
+ "normalized": false,
367
+ "rstrip": false,
368
+ "single_word": false,
369
+ "special": true
370
+ },
371
+ "250043": {
372
+ "content": "sw_KE",
373
+ "lstrip": false,
374
+ "normalized": false,
375
+ "rstrip": false,
376
+ "single_word": false,
377
+ "special": true
378
+ },
379
+ "250044": {
380
+ "content": "ta_IN",
381
+ "lstrip": false,
382
+ "normalized": false,
383
+ "rstrip": false,
384
+ "single_word": false,
385
+ "special": true
386
+ },
387
+ "250045": {
388
+ "content": "te_IN",
389
+ "lstrip": false,
390
+ "normalized": false,
391
+ "rstrip": false,
392
+ "single_word": false,
393
+ "special": true
394
+ },
395
+ "250046": {
396
+ "content": "th_TH",
397
+ "lstrip": false,
398
+ "normalized": false,
399
+ "rstrip": false,
400
+ "single_word": false,
401
+ "special": true
402
+ },
403
+ "250047": {
404
+ "content": "tl_XX",
405
+ "lstrip": false,
406
+ "normalized": false,
407
+ "rstrip": false,
408
+ "single_word": false,
409
+ "special": true
410
+ },
411
+ "250048": {
412
+ "content": "uk_UA",
413
+ "lstrip": false,
414
+ "normalized": false,
415
+ "rstrip": false,
416
+ "single_word": false,
417
+ "special": true
418
+ },
419
+ "250049": {
420
+ "content": "ur_PK",
421
+ "lstrip": false,
422
+ "normalized": false,
423
+ "rstrip": false,
424
+ "single_word": false,
425
+ "special": true
426
+ },
427
+ "250050": {
428
+ "content": "xh_ZA",
429
+ "lstrip": false,
430
+ "normalized": false,
431
+ "rstrip": false,
432
+ "single_word": false,
433
+ "special": true
434
+ },
435
+ "250051": {
436
+ "content": "gl_ES",
437
+ "lstrip": false,
438
+ "normalized": false,
439
+ "rstrip": false,
440
+ "single_word": false,
441
+ "special": true
442
+ },
443
+ "250052": {
444
+ "content": "sl_SI",
445
+ "lstrip": false,
446
+ "normalized": false,
447
+ "rstrip": false,
448
+ "single_word": false,
449
+ "special": true
450
+ },
451
+ "250053": {
452
+ "content": "<mask>",
453
+ "lstrip": true,
454
+ "normalized": true,
455
+ "rstrip": false,
456
+ "single_word": false,
457
+ "special": true
458
+ }
459
+ },
460
+ "additional_special_tokens": [
461
+ "ar_AR",
462
+ "cs_CZ",
463
+ "de_DE",
464
+ "en_XX",
465
+ "es_XX",
466
+ "et_EE",
467
+ "fi_FI",
468
+ "fr_XX",
469
+ "gu_IN",
470
+ "hi_IN",
471
+ "it_IT",
472
+ "ja_XX",
473
+ "kk_KZ",
474
+ "ko_KR",
475
+ "lt_LT",
476
+ "lv_LV",
477
+ "my_MM",
478
+ "ne_NP",
479
+ "nl_XX",
480
+ "ro_RO",
481
+ "ru_RU",
482
+ "si_LK",
483
+ "tr_TR",
484
+ "vi_VN",
485
+ "zh_CN",
486
+ "af_ZA",
487
+ "az_AZ",
488
+ "bn_IN",
489
+ "fa_IR",
490
+ "he_IL",
491
+ "hr_HR",
492
+ "id_ID",
493
+ "ka_GE",
494
+ "km_KH",
495
+ "mk_MK",
496
+ "ml_IN",
497
+ "mn_MN",
498
+ "mr_IN",
499
+ "pl_PL",
500
+ "ps_AF",
501
+ "pt_XX",
502
+ "sv_SE",
503
+ "sw_KE",
504
+ "ta_IN",
505
+ "te_IN",
506
+ "th_TH",
507
+ "tl_XX",
508
+ "uk_UA",
509
+ "ur_PK",
510
+ "xh_ZA",
511
+ "gl_ES",
512
+ "sl_SI"
513
+ ],
514
+ "bos_token": "<s>",
515
+ "clean_up_tokenization_spaces": true,
516
+ "cls_token": "<s>",
517
+ "eos_token": "</s>",
518
+ "language_codes": "ML50",
519
+ "mask_token": "<mask>",
520
+ "model_max_length": 1000000000000000019884624838656,
521
+ "pad_token": "<pad>",
522
+ "sep_token": "</s>",
523
+ "sp_model_kwargs": {},
524
+ "src_lang": "ja_XX",
525
+ "tgt_lang": "ko_KR",
526
+ "tokenizer_class": "MBart50Tokenizer",
527
+ "unk_token": "<unk>"
528
+ }
checkpoint-87009/trainer_state.json ADDED
@@ -0,0 +1,1083 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6814723610877991,
3
+ "best_model_checkpoint": "./jako_mbartLarge_100p_run1/checkpoint-87009",
4
+ "epoch": 2.0,
5
+ "eval_steps": 500,
6
+ "global_step": 87009,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.01,
13
+ "learning_rate": 4.97e-05,
14
+ "loss": 1.9891,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.02,
19
+ "learning_rate": 4.996196669018189e-05,
20
+ "loss": 1.5354,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.03,
25
+ "learning_rate": 4.9923626660123305e-05,
26
+ "loss": 1.3833,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.05,
31
+ "learning_rate": 4.988528663006472e-05,
32
+ "loss": 1.2912,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.06,
37
+ "learning_rate": 4.984694660000614e-05,
38
+ "loss": 1.2355,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.07,
43
+ "learning_rate": 4.980860656994755e-05,
44
+ "loss": 1.1927,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.08,
49
+ "learning_rate": 4.977026653988897e-05,
50
+ "loss": 1.166,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.09,
55
+ "learning_rate": 4.9731926509830386e-05,
56
+ "loss": 1.1238,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.1,
61
+ "learning_rate": 4.9693586479771804e-05,
62
+ "loss": 1.0972,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.11,
67
+ "learning_rate": 4.9655246449713215e-05,
68
+ "loss": 1.0964,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.13,
73
+ "learning_rate": 4.961690641965463e-05,
74
+ "loss": 1.0611,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.14,
79
+ "learning_rate": 4.957871974971629e-05,
80
+ "loss": 1.0487,
81
+ "step": 6000
82
+ },
83
+ {
84
+ "epoch": 0.15,
85
+ "learning_rate": 4.9540379719657705e-05,
86
+ "loss": 1.0166,
87
+ "step": 6500
88
+ },
89
+ {
90
+ "epoch": 0.16,
91
+ "learning_rate": 4.9502039689599116e-05,
92
+ "loss": 1.0162,
93
+ "step": 7000
94
+ },
95
+ {
96
+ "epoch": 0.17,
97
+ "learning_rate": 4.946369965954053e-05,
98
+ "loss": 0.9952,
99
+ "step": 7500
100
+ },
101
+ {
102
+ "epoch": 0.18,
103
+ "learning_rate": 4.942535962948195e-05,
104
+ "loss": 0.9853,
105
+ "step": 8000
106
+ },
107
+ {
108
+ "epoch": 0.2,
109
+ "learning_rate": 4.9387096279483484e-05,
110
+ "loss": 0.9835,
111
+ "step": 8500
112
+ },
113
+ {
114
+ "epoch": 0.21,
115
+ "learning_rate": 4.9348832929485017e-05,
116
+ "loss": 0.9634,
117
+ "step": 9000
118
+ },
119
+ {
120
+ "epoch": 0.22,
121
+ "learning_rate": 4.9310492899426434e-05,
122
+ "loss": 0.9532,
123
+ "step": 9500
124
+ },
125
+ {
126
+ "epoch": 0.23,
127
+ "learning_rate": 4.927215286936785e-05,
128
+ "loss": 0.955,
129
+ "step": 10000
130
+ },
131
+ {
132
+ "epoch": 0.24,
133
+ "learning_rate": 4.923381283930927e-05,
134
+ "loss": 0.9392,
135
+ "step": 10500
136
+ },
137
+ {
138
+ "epoch": 0.25,
139
+ "learning_rate": 4.919547280925069e-05,
140
+ "loss": 0.9403,
141
+ "step": 11000
142
+ },
143
+ {
144
+ "epoch": 0.26,
145
+ "learning_rate": 4.91571327791921e-05,
146
+ "loss": 0.9177,
147
+ "step": 11500
148
+ },
149
+ {
150
+ "epoch": 0.28,
151
+ "learning_rate": 4.9118792749133516e-05,
152
+ "loss": 0.919,
153
+ "step": 12000
154
+ },
155
+ {
156
+ "epoch": 0.29,
157
+ "learning_rate": 4.908045271907493e-05,
158
+ "loss": 0.9036,
159
+ "step": 12500
160
+ },
161
+ {
162
+ "epoch": 0.3,
163
+ "learning_rate": 4.9042189369076466e-05,
164
+ "loss": 0.9019,
165
+ "step": 13000
166
+ },
167
+ {
168
+ "epoch": 0.31,
169
+ "learning_rate": 4.9003849339017884e-05,
170
+ "loss": 0.8969,
171
+ "step": 13500
172
+ },
173
+ {
174
+ "epoch": 0.32,
175
+ "learning_rate": 4.89655093089593e-05,
176
+ "loss": 0.891,
177
+ "step": 14000
178
+ },
179
+ {
180
+ "epoch": 0.33,
181
+ "learning_rate": 4.892716927890072e-05,
182
+ "loss": 0.8904,
183
+ "step": 14500
184
+ },
185
+ {
186
+ "epoch": 0.34,
187
+ "learning_rate": 4.888882924884213e-05,
188
+ "loss": 0.8833,
189
+ "step": 15000
190
+ },
191
+ {
192
+ "epoch": 0.36,
193
+ "learning_rate": 4.8850489218783554e-05,
194
+ "loss": 0.8668,
195
+ "step": 15500
196
+ },
197
+ {
198
+ "epoch": 0.37,
199
+ "learning_rate": 4.8812149188724965e-05,
200
+ "loss": 0.8692,
201
+ "step": 16000
202
+ },
203
+ {
204
+ "epoch": 0.38,
205
+ "learning_rate": 4.877380915866638e-05,
206
+ "loss": 0.8617,
207
+ "step": 16500
208
+ },
209
+ {
210
+ "epoch": 0.39,
211
+ "learning_rate": 4.8735469128607794e-05,
212
+ "loss": 0.866,
213
+ "step": 17000
214
+ },
215
+ {
216
+ "epoch": 0.4,
217
+ "learning_rate": 4.869720577860933e-05,
218
+ "loss": 0.8592,
219
+ "step": 17500
220
+ },
221
+ {
222
+ "epoch": 0.41,
223
+ "learning_rate": 4.8658865748550744e-05,
224
+ "loss": 0.8586,
225
+ "step": 18000
226
+ },
227
+ {
228
+ "epoch": 0.43,
229
+ "learning_rate": 4.862052571849217e-05,
230
+ "loss": 0.84,
231
+ "step": 18500
232
+ },
233
+ {
234
+ "epoch": 0.44,
235
+ "learning_rate": 4.8582262368493695e-05,
236
+ "loss": 0.84,
237
+ "step": 19000
238
+ },
239
+ {
240
+ "epoch": 0.45,
241
+ "learning_rate": 4.854392233843512e-05,
242
+ "loss": 0.8397,
243
+ "step": 19500
244
+ },
245
+ {
246
+ "epoch": 0.46,
247
+ "learning_rate": 4.850581234855688e-05,
248
+ "loss": 0.8338,
249
+ "step": 20000
250
+ },
251
+ {
252
+ "epoch": 0.47,
253
+ "learning_rate": 4.84674723184983e-05,
254
+ "loss": 0.838,
255
+ "step": 20500
256
+ },
257
+ {
258
+ "epoch": 0.48,
259
+ "learning_rate": 4.842913228843972e-05,
260
+ "loss": 0.8391,
261
+ "step": 21000
262
+ },
263
+ {
264
+ "epoch": 0.49,
265
+ "learning_rate": 4.8390792258381135e-05,
266
+ "loss": 0.8276,
267
+ "step": 21500
268
+ },
269
+ {
270
+ "epoch": 0.51,
271
+ "learning_rate": 4.8352452228322546e-05,
272
+ "loss": 0.8198,
273
+ "step": 22000
274
+ },
275
+ {
276
+ "epoch": 0.52,
277
+ "learning_rate": 4.8314112198263963e-05,
278
+ "loss": 0.8127,
279
+ "step": 22500
280
+ },
281
+ {
282
+ "epoch": 0.53,
283
+ "learning_rate": 4.827577216820538e-05,
284
+ "loss": 0.8184,
285
+ "step": 23000
286
+ },
287
+ {
288
+ "epoch": 0.54,
289
+ "learning_rate": 4.82374321381468e-05,
290
+ "loss": 0.817,
291
+ "step": 23500
292
+ },
293
+ {
294
+ "epoch": 0.55,
295
+ "learning_rate": 4.8199092108088216e-05,
296
+ "loss": 0.8196,
297
+ "step": 24000
298
+ },
299
+ {
300
+ "epoch": 0.56,
301
+ "learning_rate": 4.816075207802963e-05,
302
+ "loss": 0.8024,
303
+ "step": 24500
304
+ },
305
+ {
306
+ "epoch": 0.57,
307
+ "learning_rate": 4.812241204797105e-05,
308
+ "loss": 0.8133,
309
+ "step": 25000
310
+ },
311
+ {
312
+ "epoch": 0.59,
313
+ "learning_rate": 4.808407201791246e-05,
314
+ "loss": 0.8056,
315
+ "step": 25500
316
+ },
317
+ {
318
+ "epoch": 0.6,
319
+ "learning_rate": 4.804573198785388e-05,
320
+ "loss": 0.7972,
321
+ "step": 26000
322
+ },
323
+ {
324
+ "epoch": 0.61,
325
+ "learning_rate": 4.80073919577953e-05,
326
+ "loss": 0.8026,
327
+ "step": 26500
328
+ },
329
+ {
330
+ "epoch": 0.62,
331
+ "learning_rate": 4.7969051927736716e-05,
332
+ "loss": 0.7983,
333
+ "step": 27000
334
+ },
335
+ {
336
+ "epoch": 0.63,
337
+ "learning_rate": 4.793078857773824e-05,
338
+ "loss": 0.795,
339
+ "step": 27500
340
+ },
341
+ {
342
+ "epoch": 0.64,
343
+ "learning_rate": 4.7892448547679666e-05,
344
+ "loss": 0.803,
345
+ "step": 28000
346
+ },
347
+ {
348
+ "epoch": 0.66,
349
+ "learning_rate": 4.7854108517621084e-05,
350
+ "loss": 0.7911,
351
+ "step": 28500
352
+ },
353
+ {
354
+ "epoch": 0.67,
355
+ "learning_rate": 4.7815768487562494e-05,
356
+ "loss": 0.7757,
357
+ "step": 29000
358
+ },
359
+ {
360
+ "epoch": 0.68,
361
+ "learning_rate": 4.777742845750391e-05,
362
+ "loss": 0.777,
363
+ "step": 29500
364
+ },
365
+ {
366
+ "epoch": 0.69,
367
+ "learning_rate": 4.773908842744533e-05,
368
+ "loss": 0.7848,
369
+ "step": 30000
370
+ },
371
+ {
372
+ "epoch": 0.7,
373
+ "learning_rate": 4.770074839738675e-05,
374
+ "loss": 0.7834,
375
+ "step": 30500
376
+ },
377
+ {
378
+ "epoch": 0.71,
379
+ "learning_rate": 4.766240836732816e-05,
380
+ "loss": 0.7753,
381
+ "step": 31000
382
+ },
383
+ {
384
+ "epoch": 0.72,
385
+ "learning_rate": 4.7624068337269576e-05,
386
+ "loss": 0.7715,
387
+ "step": 31500
388
+ },
389
+ {
390
+ "epoch": 0.74,
391
+ "learning_rate": 4.7585728307211e-05,
392
+ "loss": 0.7714,
393
+ "step": 32000
394
+ },
395
+ {
396
+ "epoch": 0.75,
397
+ "learning_rate": 4.754738827715241e-05,
398
+ "loss": 0.7736,
399
+ "step": 32500
400
+ },
401
+ {
402
+ "epoch": 0.76,
403
+ "learning_rate": 4.750904824709383e-05,
404
+ "loss": 0.7767,
405
+ "step": 33000
406
+ },
407
+ {
408
+ "epoch": 0.77,
409
+ "learning_rate": 4.747070821703524e-05,
410
+ "loss": 0.7661,
411
+ "step": 33500
412
+ },
413
+ {
414
+ "epoch": 0.78,
415
+ "learning_rate": 4.7432368186976664e-05,
416
+ "loss": 0.7565,
417
+ "step": 34000
418
+ },
419
+ {
420
+ "epoch": 0.79,
421
+ "learning_rate": 4.739410483697819e-05,
422
+ "loss": 0.7515,
423
+ "step": 34500
424
+ },
425
+ {
426
+ "epoch": 0.8,
427
+ "learning_rate": 4.735584148697973e-05,
428
+ "loss": 0.7558,
429
+ "step": 35000
430
+ },
431
+ {
432
+ "epoch": 0.82,
433
+ "learning_rate": 4.731750145692114e-05,
434
+ "loss": 0.7642,
435
+ "step": 35500
436
+ },
437
+ {
438
+ "epoch": 0.83,
439
+ "learning_rate": 4.727916142686256e-05,
440
+ "loss": 0.7626,
441
+ "step": 36000
442
+ },
443
+ {
444
+ "epoch": 0.84,
445
+ "learning_rate": 4.7240821396803976e-05,
446
+ "loss": 0.7627,
447
+ "step": 36500
448
+ },
449
+ {
450
+ "epoch": 0.85,
451
+ "learning_rate": 4.7202481366745394e-05,
452
+ "loss": 0.7495,
453
+ "step": 37000
454
+ },
455
+ {
456
+ "epoch": 0.86,
457
+ "learning_rate": 4.716414133668681e-05,
458
+ "loss": 0.7582,
459
+ "step": 37500
460
+ },
461
+ {
462
+ "epoch": 0.87,
463
+ "learning_rate": 4.7125877986688344e-05,
464
+ "loss": 0.7481,
465
+ "step": 38000
466
+ },
467
+ {
468
+ "epoch": 0.88,
469
+ "learning_rate": 4.708753795662976e-05,
470
+ "loss": 0.7557,
471
+ "step": 38500
472
+ },
473
+ {
474
+ "epoch": 0.9,
475
+ "learning_rate": 4.704919792657117e-05,
476
+ "loss": 0.7384,
477
+ "step": 39000
478
+ },
479
+ {
480
+ "epoch": 0.91,
481
+ "learning_rate": 4.70108578965126e-05,
482
+ "loss": 0.742,
483
+ "step": 39500
484
+ },
485
+ {
486
+ "epoch": 0.92,
487
+ "learning_rate": 4.697251786645401e-05,
488
+ "loss": 0.7388,
489
+ "step": 40000
490
+ },
491
+ {
492
+ "epoch": 0.93,
493
+ "learning_rate": 4.6934177836395425e-05,
494
+ "loss": 0.7503,
495
+ "step": 40500
496
+ },
497
+ {
498
+ "epoch": 0.94,
499
+ "learning_rate": 4.689583780633684e-05,
500
+ "loss": 0.745,
501
+ "step": 41000
502
+ },
503
+ {
504
+ "epoch": 0.95,
505
+ "learning_rate": 4.685749777627826e-05,
506
+ "loss": 0.7228,
507
+ "step": 41500
508
+ },
509
+ {
510
+ "epoch": 0.97,
511
+ "learning_rate": 4.681915774621968e-05,
512
+ "loss": 0.7333,
513
+ "step": 42000
514
+ },
515
+ {
516
+ "epoch": 0.98,
517
+ "learning_rate": 4.678081771616109e-05,
518
+ "loss": 0.7371,
519
+ "step": 42500
520
+ },
521
+ {
522
+ "epoch": 0.99,
523
+ "learning_rate": 4.674247768610251e-05,
524
+ "loss": 0.733,
525
+ "step": 43000
526
+ },
527
+ {
528
+ "epoch": 1.0,
529
+ "learning_rate": 4.6704137656043925e-05,
530
+ "loss": 0.7241,
531
+ "step": 43500
532
+ },
533
+ {
534
+ "epoch": 1.0,
535
+ "eval_bleu": 56.112,
536
+ "eval_gen_len": 17.3165,
537
+ "eval_loss": 0.7177675366401672,
538
+ "eval_runtime": 7671.1849,
539
+ "eval_samples_per_second": 11.342,
540
+ "eval_steps_per_second": 1.418,
541
+ "step": 43504
542
+ },
543
+ {
544
+ "epoch": 1.01,
545
+ "learning_rate": 4.666587430604546e-05,
546
+ "loss": 0.7322,
547
+ "step": 44000
548
+ },
549
+ {
550
+ "epoch": 1.02,
551
+ "learning_rate": 4.6627534275986875e-05,
552
+ "loss": 0.7051,
553
+ "step": 44500
554
+ },
555
+ {
556
+ "epoch": 1.03,
557
+ "learning_rate": 4.658919424592829e-05,
558
+ "loss": 0.6935,
559
+ "step": 45000
560
+ },
561
+ {
562
+ "epoch": 1.05,
563
+ "learning_rate": 4.6550854215869704e-05,
564
+ "loss": 0.6831,
565
+ "step": 45500
566
+ },
567
+ {
568
+ "epoch": 1.06,
569
+ "learning_rate": 4.651259086587124e-05,
570
+ "loss": 0.6787,
571
+ "step": 46000
572
+ },
573
+ {
574
+ "epoch": 1.07,
575
+ "learning_rate": 4.6474327515872776e-05,
576
+ "loss": 0.6813,
577
+ "step": 46500
578
+ },
579
+ {
580
+ "epoch": 1.08,
581
+ "learning_rate": 4.6436140845934424e-05,
582
+ "loss": 0.6841,
583
+ "step": 47000
584
+ },
585
+ {
586
+ "epoch": 1.09,
587
+ "learning_rate": 4.639780081587584e-05,
588
+ "loss": 0.6697,
589
+ "step": 47500
590
+ },
591
+ {
592
+ "epoch": 1.1,
593
+ "learning_rate": 4.635946078581726e-05,
594
+ "loss": 0.6605,
595
+ "step": 48000
596
+ },
597
+ {
598
+ "epoch": 1.11,
599
+ "learning_rate": 4.632112075575868e-05,
600
+ "loss": 0.675,
601
+ "step": 48500
602
+ },
603
+ {
604
+ "epoch": 1.13,
605
+ "learning_rate": 4.6282780725700094e-05,
606
+ "loss": 0.6623,
607
+ "step": 49000
608
+ },
609
+ {
610
+ "epoch": 1.14,
611
+ "learning_rate": 4.6244440695641505e-05,
612
+ "loss": 0.6589,
613
+ "step": 49500
614
+ },
615
+ {
616
+ "epoch": 1.15,
617
+ "learning_rate": 4.620610066558292e-05,
618
+ "loss": 0.6476,
619
+ "step": 50000
620
+ },
621
+ {
622
+ "epoch": 1.16,
623
+ "learning_rate": 4.616776063552434e-05,
624
+ "loss": 0.6551,
625
+ "step": 50500
626
+ },
627
+ {
628
+ "epoch": 1.17,
629
+ "learning_rate": 4.612942060546576e-05,
630
+ "loss": 0.6502,
631
+ "step": 51000
632
+ },
633
+ {
634
+ "epoch": 1.18,
635
+ "learning_rate": 4.6091080575407176e-05,
636
+ "loss": 0.6478,
637
+ "step": 51500
638
+ },
639
+ {
640
+ "epoch": 1.2,
641
+ "learning_rate": 4.605274054534859e-05,
642
+ "loss": 0.6485,
643
+ "step": 52000
644
+ },
645
+ {
646
+ "epoch": 1.21,
647
+ "learning_rate": 4.6014400515290004e-05,
648
+ "loss": 0.6412,
649
+ "step": 52500
650
+ },
651
+ {
652
+ "epoch": 1.22,
653
+ "learning_rate": 4.597606048523142e-05,
654
+ "loss": 0.6376,
655
+ "step": 53000
656
+ },
657
+ {
658
+ "epoch": 1.23,
659
+ "learning_rate": 4.593772045517284e-05,
660
+ "loss": 0.6385,
661
+ "step": 53500
662
+ },
663
+ {
664
+ "epoch": 1.24,
665
+ "learning_rate": 4.589945710517437e-05,
666
+ "loss": 0.6366,
667
+ "step": 54000
668
+ },
669
+ {
670
+ "epoch": 1.25,
671
+ "learning_rate": 4.586111707511579e-05,
672
+ "loss": 0.6444,
673
+ "step": 54500
674
+ },
675
+ {
676
+ "epoch": 1.26,
677
+ "learning_rate": 4.582277704505721e-05,
678
+ "loss": 0.6311,
679
+ "step": 55000
680
+ },
681
+ {
682
+ "epoch": 1.28,
683
+ "learning_rate": 4.578451369505874e-05,
684
+ "loss": 0.6306,
685
+ "step": 55500
686
+ },
687
+ {
688
+ "epoch": 1.29,
689
+ "learning_rate": 4.574617366500016e-05,
690
+ "loss": 0.6199,
691
+ "step": 56000
692
+ },
693
+ {
694
+ "epoch": 1.3,
695
+ "learning_rate": 4.570783363494157e-05,
696
+ "loss": 0.6268,
697
+ "step": 56500
698
+ },
699
+ {
700
+ "epoch": 1.31,
701
+ "learning_rate": 4.566949360488299e-05,
702
+ "loss": 0.623,
703
+ "step": 57000
704
+ },
705
+ {
706
+ "epoch": 1.32,
707
+ "learning_rate": 4.5631153574824404e-05,
708
+ "loss": 0.6207,
709
+ "step": 57500
710
+ },
711
+ {
712
+ "epoch": 1.33,
713
+ "learning_rate": 4.559281354476582e-05,
714
+ "loss": 0.6204,
715
+ "step": 58000
716
+ },
717
+ {
718
+ "epoch": 1.34,
719
+ "learning_rate": 4.555447351470723e-05,
720
+ "loss": 0.6215,
721
+ "step": 58500
722
+ },
723
+ {
724
+ "epoch": 1.36,
725
+ "learning_rate": 4.551613348464866e-05,
726
+ "loss": 0.6114,
727
+ "step": 59000
728
+ },
729
+ {
730
+ "epoch": 1.37,
731
+ "learning_rate": 4.5477793454590075e-05,
732
+ "loss": 0.6134,
733
+ "step": 59500
734
+ },
735
+ {
736
+ "epoch": 1.38,
737
+ "learning_rate": 4.543953010459161e-05,
738
+ "loss": 0.6142,
739
+ "step": 60000
740
+ },
741
+ {
742
+ "epoch": 1.39,
743
+ "learning_rate": 4.5401190074533025e-05,
744
+ "loss": 0.6149,
745
+ "step": 60500
746
+ },
747
+ {
748
+ "epoch": 1.4,
749
+ "learning_rate": 4.5362850044474436e-05,
750
+ "loss": 0.6123,
751
+ "step": 61000
752
+ },
753
+ {
754
+ "epoch": 1.41,
755
+ "learning_rate": 4.5324510014415854e-05,
756
+ "loss": 0.6159,
757
+ "step": 61500
758
+ },
759
+ {
760
+ "epoch": 1.43,
761
+ "learning_rate": 4.5286246664417387e-05,
762
+ "loss": 0.6001,
763
+ "step": 62000
764
+ },
765
+ {
766
+ "epoch": 1.44,
767
+ "learning_rate": 4.5247906634358804e-05,
768
+ "loss": 0.6034,
769
+ "step": 62500
770
+ },
771
+ {
772
+ "epoch": 1.45,
773
+ "learning_rate": 4.520956660430022e-05,
774
+ "loss": 0.5999,
775
+ "step": 63000
776
+ },
777
+ {
778
+ "epoch": 1.46,
779
+ "learning_rate": 4.517122657424164e-05,
780
+ "loss": 0.6011,
781
+ "step": 63500
782
+ },
783
+ {
784
+ "epoch": 1.47,
785
+ "learning_rate": 4.513288654418305e-05,
786
+ "loss": 0.6051,
787
+ "step": 64000
788
+ },
789
+ {
790
+ "epoch": 1.48,
791
+ "learning_rate": 4.509454651412447e-05,
792
+ "loss": 0.6061,
793
+ "step": 64500
794
+ },
795
+ {
796
+ "epoch": 1.49,
797
+ "learning_rate": 4.5056206484065886e-05,
798
+ "loss": 0.598,
799
+ "step": 65000
800
+ },
801
+ {
802
+ "epoch": 1.51,
803
+ "learning_rate": 4.50178664540073e-05,
804
+ "loss": 0.5939,
805
+ "step": 65500
806
+ },
807
+ {
808
+ "epoch": 1.52,
809
+ "learning_rate": 4.497952642394872e-05,
810
+ "loss": 0.5905,
811
+ "step": 66000
812
+ },
813
+ {
814
+ "epoch": 1.53,
815
+ "learning_rate": 4.494118639389013e-05,
816
+ "loss": 0.5968,
817
+ "step": 66500
818
+ },
819
+ {
820
+ "epoch": 1.54,
821
+ "learning_rate": 4.490284636383155e-05,
822
+ "loss": 0.5943,
823
+ "step": 67000
824
+ },
825
+ {
826
+ "epoch": 1.55,
827
+ "learning_rate": 4.486450633377297e-05,
828
+ "loss": 0.6025,
829
+ "step": 67500
830
+ },
831
+ {
832
+ "epoch": 1.56,
833
+ "learning_rate": 4.4826166303714385e-05,
834
+ "loss": 0.5918,
835
+ "step": 68000
836
+ },
837
+ {
838
+ "epoch": 1.57,
839
+ "learning_rate": 4.47878262736558e-05,
840
+ "loss": 0.5983,
841
+ "step": 68500
842
+ },
843
+ {
844
+ "epoch": 1.59,
845
+ "learning_rate": 4.474948624359721e-05,
846
+ "loss": 0.5907,
847
+ "step": 69000
848
+ },
849
+ {
850
+ "epoch": 1.6,
851
+ "learning_rate": 4.471114621353864e-05,
852
+ "loss": 0.5852,
853
+ "step": 69500
854
+ },
855
+ {
856
+ "epoch": 1.61,
857
+ "learning_rate": 4.4672882863540164e-05,
858
+ "loss": 0.5886,
859
+ "step": 70000
860
+ },
861
+ {
862
+ "epoch": 1.62,
863
+ "learning_rate": 4.463454283348159e-05,
864
+ "loss": 0.5895,
865
+ "step": 70500
866
+ },
867
+ {
868
+ "epoch": 1.63,
869
+ "learning_rate": 4.4596202803423e-05,
870
+ "loss": 0.5887,
871
+ "step": 71000
872
+ },
873
+ {
874
+ "epoch": 1.64,
875
+ "learning_rate": 4.455786277336442e-05,
876
+ "loss": 0.5953,
877
+ "step": 71500
878
+ },
879
+ {
880
+ "epoch": 1.66,
881
+ "learning_rate": 4.451952274330583e-05,
882
+ "loss": 0.5888,
883
+ "step": 72000
884
+ },
885
+ {
886
+ "epoch": 1.67,
887
+ "learning_rate": 4.448118271324725e-05,
888
+ "loss": 0.571,
889
+ "step": 72500
890
+ },
891
+ {
892
+ "epoch": 1.68,
893
+ "learning_rate": 4.444291936324878e-05,
894
+ "loss": 0.576,
895
+ "step": 73000
896
+ },
897
+ {
898
+ "epoch": 1.69,
899
+ "learning_rate": 4.44045793331902e-05,
900
+ "loss": 0.5828,
901
+ "step": 73500
902
+ },
903
+ {
904
+ "epoch": 1.7,
905
+ "learning_rate": 4.436631598319173e-05,
906
+ "loss": 0.5875,
907
+ "step": 74000
908
+ },
909
+ {
910
+ "epoch": 1.71,
911
+ "learning_rate": 4.432797595313315e-05,
912
+ "loss": 0.5773,
913
+ "step": 74500
914
+ },
915
+ {
916
+ "epoch": 1.72,
917
+ "learning_rate": 4.428963592307457e-05,
918
+ "loss": 0.5741,
919
+ "step": 75000
920
+ },
921
+ {
922
+ "epoch": 1.74,
923
+ "learning_rate": 4.425129589301598e-05,
924
+ "loss": 0.5721,
925
+ "step": 75500
926
+ },
927
+ {
928
+ "epoch": 1.75,
929
+ "learning_rate": 4.42129558629574e-05,
930
+ "loss": 0.5797,
931
+ "step": 76000
932
+ },
933
+ {
934
+ "epoch": 1.76,
935
+ "learning_rate": 4.417461583289882e-05,
936
+ "loss": 0.5787,
937
+ "step": 76500
938
+ },
939
+ {
940
+ "epoch": 1.77,
941
+ "learning_rate": 4.4136275802840234e-05,
942
+ "loss": 0.5731,
943
+ "step": 77000
944
+ },
945
+ {
946
+ "epoch": 1.78,
947
+ "learning_rate": 4.409801245284177e-05,
948
+ "loss": 0.5642,
949
+ "step": 77500
950
+ },
951
+ {
952
+ "epoch": 1.79,
953
+ "learning_rate": 4.4059672422783185e-05,
954
+ "loss": 0.5653,
955
+ "step": 78000
956
+ },
957
+ {
958
+ "epoch": 1.8,
959
+ "learning_rate": 4.4021332392724596e-05,
960
+ "loss": 0.5663,
961
+ "step": 78500
962
+ },
963
+ {
964
+ "epoch": 1.82,
965
+ "learning_rate": 4.398299236266601e-05,
966
+ "loss": 0.5762,
967
+ "step": 79000
968
+ },
969
+ {
970
+ "epoch": 1.83,
971
+ "learning_rate": 4.394465233260743e-05,
972
+ "loss": 0.5742,
973
+ "step": 79500
974
+ },
975
+ {
976
+ "epoch": 1.84,
977
+ "learning_rate": 4.390631230254885e-05,
978
+ "loss": 0.5739,
979
+ "step": 80000
980
+ },
981
+ {
982
+ "epoch": 1.85,
983
+ "learning_rate": 4.3867972272490266e-05,
984
+ "loss": 0.5686,
985
+ "step": 80500
986
+ },
987
+ {
988
+ "epoch": 1.86,
989
+ "learning_rate": 4.382963224243168e-05,
990
+ "loss": 0.5757,
991
+ "step": 81000
992
+ },
993
+ {
994
+ "epoch": 1.87,
995
+ "learning_rate": 4.3791292212373095e-05,
996
+ "loss": 0.5654,
997
+ "step": 81500
998
+ },
999
+ {
1000
+ "epoch": 1.88,
1001
+ "learning_rate": 4.375295218231451e-05,
1002
+ "loss": 0.5697,
1003
+ "step": 82000
1004
+ },
1005
+ {
1006
+ "epoch": 1.9,
1007
+ "learning_rate": 4.371461215225593e-05,
1008
+ "loss": 0.5582,
1009
+ "step": 82500
1010
+ },
1011
+ {
1012
+ "epoch": 1.91,
1013
+ "learning_rate": 4.367627212219735e-05,
1014
+ "loss": 0.5603,
1015
+ "step": 83000
1016
+ },
1017
+ {
1018
+ "epoch": 1.92,
1019
+ "learning_rate": 4.363793209213876e-05,
1020
+ "loss": 0.5564,
1021
+ "step": 83500
1022
+ },
1023
+ {
1024
+ "epoch": 1.93,
1025
+ "learning_rate": 4.359959206208018e-05,
1026
+ "loss": 0.5682,
1027
+ "step": 84000
1028
+ },
1029
+ {
1030
+ "epoch": 1.94,
1031
+ "learning_rate": 4.356132871208171e-05,
1032
+ "loss": 0.566,
1033
+ "step": 84500
1034
+ },
1035
+ {
1036
+ "epoch": 1.95,
1037
+ "learning_rate": 4.3522988682023133e-05,
1038
+ "loss": 0.5485,
1039
+ "step": 85000
1040
+ },
1041
+ {
1042
+ "epoch": 1.97,
1043
+ "learning_rate": 4.3484648651964544e-05,
1044
+ "loss": 0.5558,
1045
+ "step": 85500
1046
+ },
1047
+ {
1048
+ "epoch": 1.98,
1049
+ "learning_rate": 4.344630862190596e-05,
1050
+ "loss": 0.5592,
1051
+ "step": 86000
1052
+ },
1053
+ {
1054
+ "epoch": 1.99,
1055
+ "learning_rate": 4.340796859184737e-05,
1056
+ "loss": 0.5611,
1057
+ "step": 86500
1058
+ },
1059
+ {
1060
+ "epoch": 2.0,
1061
+ "learning_rate": 4.337001196208938e-05,
1062
+ "loss": 0.5523,
1063
+ "step": 87000
1064
+ },
1065
+ {
1066
+ "epoch": 2.0,
1067
+ "eval_bleu": 57.9768,
1068
+ "eval_gen_len": 17.2796,
1069
+ "eval_loss": 0.6814723610877991,
1070
+ "eval_runtime": 7592.1301,
1071
+ "eval_samples_per_second": 11.46,
1072
+ "eval_steps_per_second": 1.433,
1073
+ "step": 87009
1074
+ }
1075
+ ],
1076
+ "logging_steps": 500,
1077
+ "max_steps": 652560,
1078
+ "num_train_epochs": 15,
1079
+ "save_steps": 500,
1080
+ "total_flos": 3.016956704056148e+18,
1081
+ "trial_name": null,
1082
+ "trial_params": null
1083
+ }
checkpoint-87009/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:44b731ecd386cb51a958b6c8f191496c616dc6e80b60cf4e62402772da333973
3
+ size 4792
config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/mbart-large-50-many-to-many-mmt",
3
+ "_num_labels": 3,
4
+ "activation_dropout": 0.0,
5
+ "activation_function": "relu",
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": 2,
21
+ "dropout": 0.1,
22
+ "early_stopping": true,
23
+ "encoder_attention_heads": 16,
24
+ "encoder_ffn_dim": 4096,
25
+ "encoder_layerdrop": 0.0,
26
+ "encoder_layers": 12,
27
+ "eos_token_id": 2,
28
+ "forced_bos_token_id": 250014,
29
+ "forced_eos_token_id": 2,
30
+ "gradient_checkpointing": false,
31
+ "id2label": {
32
+ "0": "LABEL_0",
33
+ "1": "LABEL_1",
34
+ "2": "LABEL_2"
35
+ },
36
+ "init_std": 0.02,
37
+ "is_encoder_decoder": true,
38
+ "label2id": {
39
+ "LABEL_0": 0,
40
+ "LABEL_1": 1,
41
+ "LABEL_2": 2
42
+ },
43
+ "max_length": 200,
44
+ "max_position_embeddings": 1024,
45
+ "model_type": "mbart",
46
+ "normalize_before": true,
47
+ "normalize_embedding": true,
48
+ "num_beams": 5,
49
+ "num_hidden_layers": 12,
50
+ "output_past": true,
51
+ "pad_token_id": 1,
52
+ "scale_embedding": true,
53
+ "static_position_embeddings": false,
54
+ "tokenizer_class": "MBart50Tokenizer",
55
+ "torch_dtype": "float32",
56
+ "transformers_version": "4.35.2",
57
+ "use_cache": true,
58
+ "vocab_size": 250054
59
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 5.0,
3
+ "eval_bleu": 57.9768,
4
+ "eval_gen_len": 17.2796,
5
+ "eval_loss": 0.6814723610877991,
6
+ "eval_runtime": 7593.4193,
7
+ "eval_samples": 87008,
8
+ "eval_samples_per_second": 11.458,
9
+ "eval_steps_per_second": 1.432
10
+ }