Hugging Face's logo

cx-olquinjica
/
AngXLMR-SYN

Fill-Mask
Transformers
PyTorch
5 languages
xlm-roberta
Inference Endpoints
Model card Files Community
cx-olquinjica commited on
Commit
d72acb6
•
1 Parent(s): b334ff5

Upload 7 files

Browse files
Files changed (8) hide show
  1. .gitattributes +1 -0
  2. config.json +28 -0
  3. pytorch_model.bin +3 -0
  4. special_tokens_map.json +15 -0
  5. tokenizer.json +3 -0
  6. tokenizer_config.json +19 -0
  7. trainer_state.json +2572 -0
  8. training_args.bin +3 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ 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
+ tokenizer.json filter=lfs diff=lfs merge=lfs -text
config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "xlm-roberta-base",
3
+ "architectures": [
4
+ "XLMRobertaForMaskedLM"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "gelu",
11
+ "hidden_dropout_prob": 0.1,
12
+ "hidden_size": 768,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 3072,
15
+ "layer_norm_eps": 1e-05,
16
+ "max_position_embeddings": 514,
17
+ "model_type": "xlm-roberta",
18
+ "num_attention_heads": 12,
19
+ "num_hidden_layers": 12,
20
+ "output_past": true,
21
+ "pad_token_id": 1,
22
+ "position_embedding_type": "absolute",
23
+ "torch_dtype": "float32",
24
+ "transformers_version": "4.32.1",
25
+ "type_vocab_size": 1,
26
+ "use_cache": true,
27
+ "vocab_size": 250002
28
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7bb2408968ba500de898477180102ad9c9ece41cb0597ff5ddbf1c399d39c642
3
+ size 1113250674
special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62c24cdc13d4c9952d63718d6c9fa4c287974249e16b7ade6d5a85e7bbb75626
3
+ size 17082660
tokenizer_config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "clean_up_tokenization_spaces": true,
4
+ "cls_token": "<s>",
5
+ "eos_token": "</s>",
6
+ "mask_token": {
7
+ "__type": "AddedToken",
8
+ "content": "<mask>",
9
+ "lstrip": true,
10
+ "normalized": true,
11
+ "rstrip": false,
12
+ "single_word": false
13
+ },
14
+ "model_max_length": 512,
15
+ "pad_token": "<pad>",
16
+ "sep_token": "</s>",
17
+ "tokenizer_class": "XLMRobertaTokenizer",
18
+ "unk_token": "<unk>"
19
+ }
trainer_state.json ADDED
@@ -0,0 +1,2572 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 212490,
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": 4.9882347404583746e-05,
14
+ "loss": 2.9231,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 4.9764694809167496e-05,
20
+ "loss": 2.1627,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.02,
25
+ "learning_rate": 4.964704221375124e-05,
26
+ "loss": 1.8699,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.03,
31
+ "learning_rate": 4.9529389618334984e-05,
32
+ "loss": 1.7267,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.04,
37
+ "learning_rate": 4.941173702291873e-05,
38
+ "loss": 1.6276,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.04,
43
+ "learning_rate": 4.929408442750247e-05,
44
+ "loss": 1.5576,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.05,
49
+ "learning_rate": 4.9176431832086214e-05,
50
+ "loss": 1.5107,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.06,
55
+ "learning_rate": 4.905877923666996e-05,
56
+ "loss": 1.4644,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.06,
61
+ "learning_rate": 4.894112664125371e-05,
62
+ "loss": 1.4061,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.07,
67
+ "learning_rate": 4.882347404583745e-05,
68
+ "loss": 1.3747,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.08,
73
+ "learning_rate": 4.8705821450421196e-05,
74
+ "loss": 1.3601,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.08,
79
+ "learning_rate": 4.8588168855004946e-05,
80
+ "loss": 1.3315,
81
+ "step": 6000
82
+ },
83
+ {
84
+ "epoch": 0.09,
85
+ "learning_rate": 4.847051625958869e-05,
86
+ "loss": 1.2922,
87
+ "step": 6500
88
+ },
89
+ {
90
+ "epoch": 0.1,
91
+ "learning_rate": 4.835286366417243e-05,
92
+ "loss": 1.2699,
93
+ "step": 7000
94
+ },
95
+ {
96
+ "epoch": 0.11,
97
+ "learning_rate": 4.8235211068756184e-05,
98
+ "loss": 1.2293,
99
+ "step": 7500
100
+ },
101
+ {
102
+ "epoch": 0.11,
103
+ "learning_rate": 4.811755847333993e-05,
104
+ "loss": 1.227,
105
+ "step": 8000
106
+ },
107
+ {
108
+ "epoch": 0.12,
109
+ "learning_rate": 4.799990587792367e-05,
110
+ "loss": 1.2126,
111
+ "step": 8500
112
+ },
113
+ {
114
+ "epoch": 0.13,
115
+ "learning_rate": 4.7882253282507415e-05,
116
+ "loss": 1.1798,
117
+ "step": 9000
118
+ },
119
+ {
120
+ "epoch": 0.13,
121
+ "learning_rate": 4.776460068709116e-05,
122
+ "loss": 1.1901,
123
+ "step": 9500
124
+ },
125
+ {
126
+ "epoch": 0.14,
127
+ "learning_rate": 4.76469480916749e-05,
128
+ "loss": 1.1668,
129
+ "step": 10000
130
+ },
131
+ {
132
+ "epoch": 0.15,
133
+ "learning_rate": 4.752929549625865e-05,
134
+ "loss": 1.1431,
135
+ "step": 10500
136
+ },
137
+ {
138
+ "epoch": 0.16,
139
+ "learning_rate": 4.7411642900842396e-05,
140
+ "loss": 1.1699,
141
+ "step": 11000
142
+ },
143
+ {
144
+ "epoch": 0.16,
145
+ "learning_rate": 4.729399030542614e-05,
146
+ "loss": 1.1271,
147
+ "step": 11500
148
+ },
149
+ {
150
+ "epoch": 0.17,
151
+ "learning_rate": 4.717633771000988e-05,
152
+ "loss": 1.1251,
153
+ "step": 12000
154
+ },
155
+ {
156
+ "epoch": 0.18,
157
+ "learning_rate": 4.7058685114593633e-05,
158
+ "loss": 1.0963,
159
+ "step": 12500
160
+ },
161
+ {
162
+ "epoch": 0.18,
163
+ "learning_rate": 4.694103251917738e-05,
164
+ "loss": 1.1141,
165
+ "step": 13000
166
+ },
167
+ {
168
+ "epoch": 0.19,
169
+ "learning_rate": 4.682337992376112e-05,
170
+ "loss": 1.0974,
171
+ "step": 13500
172
+ },
173
+ {
174
+ "epoch": 0.2,
175
+ "learning_rate": 4.6705727328344864e-05,
176
+ "loss": 1.0829,
177
+ "step": 14000
178
+ },
179
+ {
180
+ "epoch": 0.2,
181
+ "learning_rate": 4.658807473292861e-05,
182
+ "loss": 1.0893,
183
+ "step": 14500
184
+ },
185
+ {
186
+ "epoch": 0.21,
187
+ "learning_rate": 4.647042213751235e-05,
188
+ "loss": 1.0835,
189
+ "step": 15000
190
+ },
191
+ {
192
+ "epoch": 0.22,
193
+ "learning_rate": 4.63527695420961e-05,
194
+ "loss": 1.0583,
195
+ "step": 15500
196
+ },
197
+ {
198
+ "epoch": 0.23,
199
+ "learning_rate": 4.6235116946679846e-05,
200
+ "loss": 1.0708,
201
+ "step": 16000
202
+ },
203
+ {
204
+ "epoch": 0.23,
205
+ "learning_rate": 4.611746435126359e-05,
206
+ "loss": 1.0695,
207
+ "step": 16500
208
+ },
209
+ {
210
+ "epoch": 0.24,
211
+ "learning_rate": 4.599981175584734e-05,
212
+ "loss": 1.0373,
213
+ "step": 17000
214
+ },
215
+ {
216
+ "epoch": 0.25,
217
+ "learning_rate": 4.588215916043108e-05,
218
+ "loss": 1.02,
219
+ "step": 17500
220
+ },
221
+ {
222
+ "epoch": 0.25,
223
+ "learning_rate": 4.576450656501483e-05,
224
+ "loss": 1.0366,
225
+ "step": 18000
226
+ },
227
+ {
228
+ "epoch": 0.26,
229
+ "learning_rate": 4.564685396959858e-05,
230
+ "loss": 1.0232,
231
+ "step": 18500
232
+ },
233
+ {
234
+ "epoch": 0.27,
235
+ "learning_rate": 4.552920137418232e-05,
236
+ "loss": 0.9946,
237
+ "step": 19000
238
+ },
239
+ {
240
+ "epoch": 0.28,
241
+ "learning_rate": 4.541154877876606e-05,
242
+ "loss": 0.9895,
243
+ "step": 19500
244
+ },
245
+ {
246
+ "epoch": 0.28,
247
+ "learning_rate": 4.52938961833498e-05,
248
+ "loss": 1.0122,
249
+ "step": 20000
250
+ },
251
+ {
252
+ "epoch": 0.29,
253
+ "learning_rate": 4.517624358793355e-05,
254
+ "loss": 1.0048,
255
+ "step": 20500
256
+ },
257
+ {
258
+ "epoch": 0.3,
259
+ "learning_rate": 4.5058590992517295e-05,
260
+ "loss": 0.9837,
261
+ "step": 21000
262
+ },
263
+ {
264
+ "epoch": 0.3,
265
+ "learning_rate": 4.494093839710104e-05,
266
+ "loss": 0.9891,
267
+ "step": 21500
268
+ },
269
+ {
270
+ "epoch": 0.31,
271
+ "learning_rate": 4.482328580168479e-05,
272
+ "loss": 0.9838,
273
+ "step": 22000
274
+ },
275
+ {
276
+ "epoch": 0.32,
277
+ "learning_rate": 4.470563320626853e-05,
278
+ "loss": 0.9881,
279
+ "step": 22500
280
+ },
281
+ {
282
+ "epoch": 0.32,
283
+ "learning_rate": 4.4587980610852277e-05,
284
+ "loss": 0.9387,
285
+ "step": 23000
286
+ },
287
+ {
288
+ "epoch": 0.33,
289
+ "learning_rate": 4.447032801543603e-05,
290
+ "loss": 0.9741,
291
+ "step": 23500
292
+ },
293
+ {
294
+ "epoch": 0.34,
295
+ "learning_rate": 4.435267542001977e-05,
296
+ "loss": 0.9668,
297
+ "step": 24000
298
+ },
299
+ {
300
+ "epoch": 0.35,
301
+ "learning_rate": 4.4235022824603514e-05,
302
+ "loss": 0.9816,
303
+ "step": 24500
304
+ },
305
+ {
306
+ "epoch": 0.35,
307
+ "learning_rate": 4.411737022918726e-05,
308
+ "loss": 0.9479,
309
+ "step": 25000
310
+ },
311
+ {
312
+ "epoch": 0.36,
313
+ "learning_rate": 4.3999717633771e-05,
314
+ "loss": 0.9568,
315
+ "step": 25500
316
+ },
317
+ {
318
+ "epoch": 0.37,
319
+ "learning_rate": 4.3882065038354745e-05,
320
+ "loss": 0.9476,
321
+ "step": 26000
322
+ },
323
+ {
324
+ "epoch": 0.37,
325
+ "learning_rate": 4.3764412442938495e-05,
326
+ "loss": 0.9533,
327
+ "step": 26500
328
+ },
329
+ {
330
+ "epoch": 0.38,
331
+ "learning_rate": 4.364675984752224e-05,
332
+ "loss": 0.9282,
333
+ "step": 27000
334
+ },
335
+ {
336
+ "epoch": 0.39,
337
+ "learning_rate": 4.352910725210598e-05,
338
+ "loss": 0.9454,
339
+ "step": 27500
340
+ },
341
+ {
342
+ "epoch": 0.4,
343
+ "learning_rate": 4.3411454656689726e-05,
344
+ "loss": 0.9418,
345
+ "step": 28000
346
+ },
347
+ {
348
+ "epoch": 0.4,
349
+ "learning_rate": 4.329380206127348e-05,
350
+ "loss": 0.9214,
351
+ "step": 28500
352
+ },
353
+ {
354
+ "epoch": 0.41,
355
+ "learning_rate": 4.317614946585722e-05,
356
+ "loss": 0.9193,
357
+ "step": 29000
358
+ },
359
+ {
360
+ "epoch": 0.42,
361
+ "learning_rate": 4.3058496870440964e-05,
362
+ "loss": 0.8905,
363
+ "step": 29500
364
+ },
365
+ {
366
+ "epoch": 0.42,
367
+ "learning_rate": 4.294084427502471e-05,
368
+ "loss": 0.9192,
369
+ "step": 30000
370
+ },
371
+ {
372
+ "epoch": 0.43,
373
+ "learning_rate": 4.282319167960845e-05,
374
+ "loss": 0.9106,
375
+ "step": 30500
376
+ },
377
+ {
378
+ "epoch": 0.44,
379
+ "learning_rate": 4.2705539084192195e-05,
380
+ "loss": 0.8986,
381
+ "step": 31000
382
+ },
383
+ {
384
+ "epoch": 0.44,
385
+ "learning_rate": 4.2587886488775945e-05,
386
+ "loss": 0.9049,
387
+ "step": 31500
388
+ },
389
+ {
390
+ "epoch": 0.45,
391
+ "learning_rate": 4.247023389335969e-05,
392
+ "loss": 0.8943,
393
+ "step": 32000
394
+ },
395
+ {
396
+ "epoch": 0.46,
397
+ "learning_rate": 4.235258129794343e-05,
398
+ "loss": 0.8912,
399
+ "step": 32500
400
+ },
401
+ {
402
+ "epoch": 0.47,
403
+ "learning_rate": 4.223492870252718e-05,
404
+ "loss": 0.8812,
405
+ "step": 33000
406
+ },
407
+ {
408
+ "epoch": 0.47,
409
+ "learning_rate": 4.2117276107110926e-05,
410
+ "loss": 0.8942,
411
+ "step": 33500
412
+ },
413
+ {
414
+ "epoch": 0.48,
415
+ "learning_rate": 4.199962351169467e-05,
416
+ "loss": 0.8956,
417
+ "step": 34000
418
+ },
419
+ {
420
+ "epoch": 0.49,
421
+ "learning_rate": 4.1881970916278414e-05,
422
+ "loss": 0.8793,
423
+ "step": 34500
424
+ },
425
+ {
426
+ "epoch": 0.49,
427
+ "learning_rate": 4.1764318320862164e-05,
428
+ "loss": 0.8793,
429
+ "step": 35000
430
+ },
431
+ {
432
+ "epoch": 0.5,
433
+ "learning_rate": 4.164666572544591e-05,
434
+ "loss": 0.8847,
435
+ "step": 35500
436
+ },
437
+ {
438
+ "epoch": 0.51,
439
+ "learning_rate": 4.152901313002965e-05,
440
+ "loss": 0.886,
441
+ "step": 36000
442
+ },
443
+ {
444
+ "epoch": 0.52,
445
+ "learning_rate": 4.1411360534613395e-05,
446
+ "loss": 0.8677,
447
+ "step": 36500
448
+ },
449
+ {
450
+ "epoch": 0.52,
451
+ "learning_rate": 4.129370793919714e-05,
452
+ "loss": 0.8751,
453
+ "step": 37000
454
+ },
455
+ {
456
+ "epoch": 0.53,
457
+ "learning_rate": 4.117605534378088e-05,
458
+ "loss": 0.8672,
459
+ "step": 37500
460
+ },
461
+ {
462
+ "epoch": 0.54,
463
+ "learning_rate": 4.105840274836463e-05,
464
+ "loss": 0.8774,
465
+ "step": 38000
466
+ },
467
+ {
468
+ "epoch": 0.54,
469
+ "learning_rate": 4.0940750152948376e-05,
470
+ "loss": 0.8735,
471
+ "step": 38500
472
+ },
473
+ {
474
+ "epoch": 0.55,
475
+ "learning_rate": 4.082309755753212e-05,
476
+ "loss": 0.8639,
477
+ "step": 39000
478
+ },
479
+ {
480
+ "epoch": 0.56,
481
+ "learning_rate": 4.070544496211587e-05,
482
+ "loss": 0.8609,
483
+ "step": 39500
484
+ },
485
+ {
486
+ "epoch": 0.56,
487
+ "learning_rate": 4.0587792366699614e-05,
488
+ "loss": 0.8669,
489
+ "step": 40000
490
+ },
491
+ {
492
+ "epoch": 0.57,
493
+ "learning_rate": 4.047013977128336e-05,
494
+ "loss": 0.8662,
495
+ "step": 40500
496
+ },
497
+ {
498
+ "epoch": 0.58,
499
+ "learning_rate": 4.03524871758671e-05,
500
+ "loss": 0.8522,
501
+ "step": 41000
502
+ },
503
+ {
504
+ "epoch": 0.59,
505
+ "learning_rate": 4.0234834580450845e-05,
506
+ "loss": 0.8609,
507
+ "step": 41500
508
+ },
509
+ {
510
+ "epoch": 0.59,
511
+ "learning_rate": 4.011718198503459e-05,
512
+ "loss": 0.8459,
513
+ "step": 42000
514
+ },
515
+ {
516
+ "epoch": 0.6,
517
+ "learning_rate": 3.999952938961833e-05,
518
+ "loss": 0.8384,
519
+ "step": 42500
520
+ },
521
+ {
522
+ "epoch": 0.61,
523
+ "learning_rate": 3.988187679420208e-05,
524
+ "loss": 0.8335,
525
+ "step": 43000
526
+ },
527
+ {
528
+ "epoch": 0.61,
529
+ "learning_rate": 3.9764224198785826e-05,
530
+ "loss": 0.8567,
531
+ "step": 43500
532
+ },
533
+ {
534
+ "epoch": 0.62,
535
+ "learning_rate": 3.964657160336957e-05,
536
+ "loss": 0.8567,
537
+ "step": 44000
538
+ },
539
+ {
540
+ "epoch": 0.63,
541
+ "learning_rate": 3.952891900795332e-05,
542
+ "loss": 0.8256,
543
+ "step": 44500
544
+ },
545
+ {
546
+ "epoch": 0.64,
547
+ "learning_rate": 3.9411266412537063e-05,
548
+ "loss": 0.8289,
549
+ "step": 45000
550
+ },
551
+ {
552
+ "epoch": 0.64,
553
+ "learning_rate": 3.929361381712081e-05,
554
+ "loss": 0.8406,
555
+ "step": 45500
556
+ },
557
+ {
558
+ "epoch": 0.65,
559
+ "learning_rate": 3.917596122170456e-05,
560
+ "loss": 0.8277,
561
+ "step": 46000
562
+ },
563
+ {
564
+ "epoch": 0.66,
565
+ "learning_rate": 3.90583086262883e-05,
566
+ "loss": 0.807,
567
+ "step": 46500
568
+ },
569
+ {
570
+ "epoch": 0.66,
571
+ "learning_rate": 3.8940656030872045e-05,
572
+ "loss": 0.8062,
573
+ "step": 47000
574
+ },
575
+ {
576
+ "epoch": 0.67,
577
+ "learning_rate": 3.882300343545579e-05,
578
+ "loss": 0.825,
579
+ "step": 47500
580
+ },
581
+ {
582
+ "epoch": 0.68,
583
+ "learning_rate": 3.870535084003953e-05,
584
+ "loss": 0.8083,
585
+ "step": 48000
586
+ },
587
+ {
588
+ "epoch": 0.68,
589
+ "learning_rate": 3.8587698244623275e-05,
590
+ "loss": 0.8222,
591
+ "step": 48500
592
+ },
593
+ {
594
+ "epoch": 0.69,
595
+ "learning_rate": 3.8470045649207026e-05,
596
+ "loss": 0.8262,
597
+ "step": 49000
598
+ },
599
+ {
600
+ "epoch": 0.7,
601
+ "learning_rate": 3.835239305379077e-05,
602
+ "loss": 0.8231,
603
+ "step": 49500
604
+ },
605
+ {
606
+ "epoch": 0.71,
607
+ "learning_rate": 3.823474045837451e-05,
608
+ "loss": 0.8154,
609
+ "step": 50000
610
+ },
611
+ {
612
+ "epoch": 0.71,
613
+ "learning_rate": 3.811708786295826e-05,
614
+ "loss": 0.8021,
615
+ "step": 50500
616
+ },
617
+ {
618
+ "epoch": 0.72,
619
+ "learning_rate": 3.799943526754201e-05,
620
+ "loss": 0.8163,
621
+ "step": 51000
622
+ },
623
+ {
624
+ "epoch": 0.73,
625
+ "learning_rate": 3.788178267212575e-05,
626
+ "loss": 0.8021,
627
+ "step": 51500
628
+ },
629
+ {
630
+ "epoch": 0.73,
631
+ "learning_rate": 3.7764130076709494e-05,
632
+ "loss": 0.7975,
633
+ "step": 52000
634
+ },
635
+ {
636
+ "epoch": 0.74,
637
+ "learning_rate": 3.764647748129324e-05,
638
+ "loss": 0.7899,
639
+ "step": 52500
640
+ },
641
+ {
642
+ "epoch": 0.75,
643
+ "learning_rate": 3.752882488587698e-05,
644
+ "loss": 0.81,
645
+ "step": 53000
646
+ },
647
+ {
648
+ "epoch": 0.76,
649
+ "learning_rate": 3.7411172290460725e-05,
650
+ "loss": 0.8023,
651
+ "step": 53500
652
+ },
653
+ {
654
+ "epoch": 0.76,
655
+ "learning_rate": 3.7293519695044476e-05,
656
+ "loss": 0.7946,
657
+ "step": 54000
658
+ },
659
+ {
660
+ "epoch": 0.77,
661
+ "learning_rate": 3.717586709962822e-05,
662
+ "loss": 0.8181,
663
+ "step": 54500
664
+ },
665
+ {
666
+ "epoch": 0.78,
667
+ "learning_rate": 3.705821450421196e-05,
668
+ "loss": 0.8046,
669
+ "step": 55000
670
+ },
671
+ {
672
+ "epoch": 0.78,
673
+ "learning_rate": 3.694056190879571e-05,
674
+ "loss": 0.809,
675
+ "step": 55500
676
+ },
677
+ {
678
+ "epoch": 0.79,
679
+ "learning_rate": 3.682290931337946e-05,
680
+ "loss": 0.8072,
681
+ "step": 56000
682
+ },
683
+ {
684
+ "epoch": 0.8,
685
+ "learning_rate": 3.67052567179632e-05,
686
+ "loss": 0.7868,
687
+ "step": 56500
688
+ },
689
+ {
690
+ "epoch": 0.8,
691
+ "learning_rate": 3.6587604122546944e-05,
692
+ "loss": 0.7887,
693
+ "step": 57000
694
+ },
695
+ {
696
+ "epoch": 0.81,
697
+ "learning_rate": 3.6469951527130694e-05,
698
+ "loss": 0.7829,
699
+ "step": 57500
700
+ },
701
+ {
702
+ "epoch": 0.82,
703
+ "learning_rate": 3.635229893171443e-05,
704
+ "loss": 0.7685,
705
+ "step": 58000
706
+ },
707
+ {
708
+ "epoch": 0.83,
709
+ "learning_rate": 3.6234646336298175e-05,
710
+ "loss": 0.7888,
711
+ "step": 58500
712
+ },
713
+ {
714
+ "epoch": 0.83,
715
+ "learning_rate": 3.6116993740881925e-05,
716
+ "loss": 0.791,
717
+ "step": 59000
718
+ },
719
+ {
720
+ "epoch": 0.84,
721
+ "learning_rate": 3.599934114546567e-05,
722
+ "loss": 0.7961,
723
+ "step": 59500
724
+ },
725
+ {
726
+ "epoch": 0.85,
727
+ "learning_rate": 3.588168855004941e-05,
728
+ "loss": 0.7733,
729
+ "step": 60000
730
+ },
731
+ {
732
+ "epoch": 0.85,
733
+ "learning_rate": 3.576403595463316e-05,
734
+ "loss": 0.7765,
735
+ "step": 60500
736
+ },
737
+ {
738
+ "epoch": 0.86,
739
+ "learning_rate": 3.5646383359216907e-05,
740
+ "loss": 0.7793,
741
+ "step": 61000
742
+ },
743
+ {
744
+ "epoch": 0.87,
745
+ "learning_rate": 3.552873076380065e-05,
746
+ "loss": 0.7985,
747
+ "step": 61500
748
+ },
749
+ {
750
+ "epoch": 0.88,
751
+ "learning_rate": 3.54110781683844e-05,
752
+ "loss": 0.7671,
753
+ "step": 62000
754
+ },
755
+ {
756
+ "epoch": 0.88,
757
+ "learning_rate": 3.5293425572968144e-05,
758
+ "loss": 0.779,
759
+ "step": 62500
760
+ },
761
+ {
762
+ "epoch": 0.89,
763
+ "learning_rate": 3.517577297755189e-05,
764
+ "loss": 0.7888,
765
+ "step": 63000
766
+ },
767
+ {
768
+ "epoch": 0.9,
769
+ "learning_rate": 3.505812038213563e-05,
770
+ "loss": 0.7731,
771
+ "step": 63500
772
+ },
773
+ {
774
+ "epoch": 0.9,
775
+ "learning_rate": 3.4940467786719375e-05,
776
+ "loss": 0.783,
777
+ "step": 64000
778
+ },
779
+ {
780
+ "epoch": 0.91,
781
+ "learning_rate": 3.482281519130312e-05,
782
+ "loss": 0.7639,
783
+ "step": 64500
784
+ },
785
+ {
786
+ "epoch": 0.92,
787
+ "learning_rate": 3.470516259588686e-05,
788
+ "loss": 0.7659,
789
+ "step": 65000
790
+ },
791
+ {
792
+ "epoch": 0.92,
793
+ "learning_rate": 3.458751000047061e-05,
794
+ "loss": 0.7667,
795
+ "step": 65500
796
+ },
797
+ {
798
+ "epoch": 0.93,
799
+ "learning_rate": 3.4469857405054356e-05,
800
+ "loss": 0.7564,
801
+ "step": 66000
802
+ },
803
+ {
804
+ "epoch": 0.94,
805
+ "learning_rate": 3.43522048096381e-05,
806
+ "loss": 0.7697,
807
+ "step": 66500
808
+ },
809
+ {
810
+ "epoch": 0.95,
811
+ "learning_rate": 3.423455221422185e-05,
812
+ "loss": 0.765,
813
+ "step": 67000
814
+ },
815
+ {
816
+ "epoch": 0.95,
817
+ "learning_rate": 3.4116899618805594e-05,
818
+ "loss": 0.7538,
819
+ "step": 67500
820
+ },
821
+ {
822
+ "epoch": 0.96,
823
+ "learning_rate": 3.399924702338934e-05,
824
+ "loss": 0.7685,
825
+ "step": 68000
826
+ },
827
+ {
828
+ "epoch": 0.97,
829
+ "learning_rate": 3.388159442797309e-05,
830
+ "loss": 0.7635,
831
+ "step": 68500
832
+ },
833
+ {
834
+ "epoch": 0.97,
835
+ "learning_rate": 3.3763941832556825e-05,
836
+ "loss": 0.7547,
837
+ "step": 69000
838
+ },
839
+ {
840
+ "epoch": 0.98,
841
+ "learning_rate": 3.364628923714057e-05,
842
+ "loss": 0.7571,
843
+ "step": 69500
844
+ },
845
+ {
846
+ "epoch": 0.99,
847
+ "learning_rate": 3.352863664172432e-05,
848
+ "loss": 0.7492,
849
+ "step": 70000
850
+ },
851
+ {
852
+ "epoch": 1.0,
853
+ "learning_rate": 3.341098404630806e-05,
854
+ "loss": 0.7469,
855
+ "step": 70500
856
+ },
857
+ {
858
+ "epoch": 1.0,
859
+ "learning_rate": 3.3293331450891806e-05,
860
+ "loss": 0.7685,
861
+ "step": 71000
862
+ },
863
+ {
864
+ "epoch": 1.01,
865
+ "learning_rate": 3.3175678855475556e-05,
866
+ "loss": 0.7542,
867
+ "step": 71500
868
+ },
869
+ {
870
+ "epoch": 1.02,
871
+ "learning_rate": 3.30580262600593e-05,
872
+ "loss": 0.7503,
873
+ "step": 72000
874
+ },
875
+ {
876
+ "epoch": 1.02,
877
+ "learning_rate": 3.2940373664643044e-05,
878
+ "loss": 0.7487,
879
+ "step": 72500
880
+ },
881
+ {
882
+ "epoch": 1.03,
883
+ "learning_rate": 3.282272106922679e-05,
884
+ "loss": 0.7496,
885
+ "step": 73000
886
+ },
887
+ {
888
+ "epoch": 1.04,
889
+ "learning_rate": 3.270506847381054e-05,
890
+ "loss": 0.7461,
891
+ "step": 73500
892
+ },
893
+ {
894
+ "epoch": 1.04,
895
+ "learning_rate": 3.258741587839428e-05,
896
+ "loss": 0.7283,
897
+ "step": 74000
898
+ },
899
+ {
900
+ "epoch": 1.05,
901
+ "learning_rate": 3.2469763282978025e-05,
902
+ "loss": 0.7378,
903
+ "step": 74500
904
+ },
905
+ {
906
+ "epoch": 1.06,
907
+ "learning_rate": 3.235211068756177e-05,
908
+ "loss": 0.731,
909
+ "step": 75000
910
+ },
911
+ {
912
+ "epoch": 1.07,
913
+ "learning_rate": 3.223445809214551e-05,
914
+ "loss": 0.7312,
915
+ "step": 75500
916
+ },
917
+ {
918
+ "epoch": 1.07,
919
+ "learning_rate": 3.2116805496729256e-05,
920
+ "loss": 0.7316,
921
+ "step": 76000
922
+ },
923
+ {
924
+ "epoch": 1.08,
925
+ "learning_rate": 3.1999152901313006e-05,
926
+ "loss": 0.7235,
927
+ "step": 76500
928
+ },
929
+ {
930
+ "epoch": 1.09,
931
+ "learning_rate": 3.188150030589675e-05,
932
+ "loss": 0.736,
933
+ "step": 77000
934
+ },
935
+ {
936
+ "epoch": 1.09,
937
+ "learning_rate": 3.176384771048049e-05,
938
+ "loss": 0.7389,
939
+ "step": 77500
940
+ },
941
+ {
942
+ "epoch": 1.1,
943
+ "learning_rate": 3.1646195115064244e-05,
944
+ "loss": 0.7232,
945
+ "step": 78000
946
+ },
947
+ {
948
+ "epoch": 1.11,
949
+ "learning_rate": 3.152854251964799e-05,
950
+ "loss": 0.744,
951
+ "step": 78500
952
+ },
953
+ {
954
+ "epoch": 1.12,
955
+ "learning_rate": 3.141088992423173e-05,
956
+ "loss": 0.734,
957
+ "step": 79000
958
+ },
959
+ {
960
+ "epoch": 1.12,
961
+ "learning_rate": 3.1293237328815475e-05,
962
+ "loss": 0.7253,
963
+ "step": 79500
964
+ },
965
+ {
966
+ "epoch": 1.13,
967
+ "learning_rate": 3.117558473339922e-05,
968
+ "loss": 0.7222,
969
+ "step": 80000
970
+ },
971
+ {
972
+ "epoch": 1.14,
973
+ "learning_rate": 3.105793213798296e-05,
974
+ "loss": 0.7347,
975
+ "step": 80500
976
+ },
977
+ {
978
+ "epoch": 1.14,
979
+ "learning_rate": 3.0940279542566705e-05,
980
+ "loss": 0.7022,
981
+ "step": 81000
982
+ },
983
+ {
984
+ "epoch": 1.15,
985
+ "learning_rate": 3.0822626947150456e-05,
986
+ "loss": 0.7324,
987
+ "step": 81500
988
+ },
989
+ {
990
+ "epoch": 1.16,
991
+ "learning_rate": 3.07049743517342e-05,
992
+ "loss": 0.7156,
993
+ "step": 82000
994
+ },
995
+ {
996
+ "epoch": 1.16,
997
+ "learning_rate": 3.058732175631794e-05,
998
+ "loss": 0.7259,
999
+ "step": 82500
1000
+ },
1001
+ {
1002
+ "epoch": 1.17,
1003
+ "learning_rate": 3.0469669160901693e-05,
1004
+ "loss": 0.7161,
1005
+ "step": 83000
1006
+ },
1007
+ {
1008
+ "epoch": 1.18,
1009
+ "learning_rate": 3.0352016565485437e-05,
1010
+ "loss": 0.7265,
1011
+ "step": 83500
1012
+ },
1013
+ {
1014
+ "epoch": 1.19,
1015
+ "learning_rate": 3.023436397006918e-05,
1016
+ "loss": 0.729,
1017
+ "step": 84000
1018
+ },
1019
+ {
1020
+ "epoch": 1.19,
1021
+ "learning_rate": 3.0116711374652928e-05,
1022
+ "loss": 0.7185,
1023
+ "step": 84500
1024
+ },
1025
+ {
1026
+ "epoch": 1.2,
1027
+ "learning_rate": 2.999905877923667e-05,
1028
+ "loss": 0.7164,
1029
+ "step": 85000
1030
+ },
1031
+ {
1032
+ "epoch": 1.21,
1033
+ "learning_rate": 2.9881406183820415e-05,
1034
+ "loss": 0.6989,
1035
+ "step": 85500
1036
+ },
1037
+ {
1038
+ "epoch": 1.21,
1039
+ "learning_rate": 2.9763753588404165e-05,
1040
+ "loss": 0.7204,
1041
+ "step": 86000
1042
+ },
1043
+ {
1044
+ "epoch": 1.22,
1045
+ "learning_rate": 2.964610099298791e-05,
1046
+ "loss": 0.7228,
1047
+ "step": 86500
1048
+ },
1049
+ {
1050
+ "epoch": 1.23,
1051
+ "learning_rate": 2.952844839757165e-05,
1052
+ "loss": 0.7076,
1053
+ "step": 87000
1054
+ },
1055
+ {
1056
+ "epoch": 1.24,
1057
+ "learning_rate": 2.9410795802155393e-05,
1058
+ "loss": 0.7198,
1059
+ "step": 87500
1060
+ },
1061
+ {
1062
+ "epoch": 1.24,
1063
+ "learning_rate": 2.9293143206739143e-05,
1064
+ "loss": 0.7035,
1065
+ "step": 88000
1066
+ },
1067
+ {
1068
+ "epoch": 1.25,
1069
+ "learning_rate": 2.9175490611322887e-05,
1070
+ "loss": 0.707,
1071
+ "step": 88500
1072
+ },
1073
+ {
1074
+ "epoch": 1.26,
1075
+ "learning_rate": 2.905783801590663e-05,
1076
+ "loss": 0.7194,
1077
+ "step": 89000
1078
+ },
1079
+ {
1080
+ "epoch": 1.26,
1081
+ "learning_rate": 2.8940185420490377e-05,
1082
+ "loss": 0.6997,
1083
+ "step": 89500
1084
+ },
1085
+ {
1086
+ "epoch": 1.27,
1087
+ "learning_rate": 2.882253282507412e-05,
1088
+ "loss": 0.714,
1089
+ "step": 90000
1090
+ },
1091
+ {
1092
+ "epoch": 1.28,
1093
+ "learning_rate": 2.8704880229657865e-05,
1094
+ "loss": 0.6986,
1095
+ "step": 90500
1096
+ },
1097
+ {
1098
+ "epoch": 1.28,
1099
+ "learning_rate": 2.8587227634241615e-05,
1100
+ "loss": 0.6982,
1101
+ "step": 91000
1102
+ },
1103
+ {
1104
+ "epoch": 1.29,
1105
+ "learning_rate": 2.846957503882536e-05,
1106
+ "loss": 0.6838,
1107
+ "step": 91500
1108
+ },
1109
+ {
1110
+ "epoch": 1.3,
1111
+ "learning_rate": 2.8351922443409102e-05,
1112
+ "loss": 0.711,
1113
+ "step": 92000
1114
+ },
1115
+ {
1116
+ "epoch": 1.31,
1117
+ "learning_rate": 2.823426984799285e-05,
1118
+ "loss": 0.7111,
1119
+ "step": 92500
1120
+ },
1121
+ {
1122
+ "epoch": 1.31,
1123
+ "learning_rate": 2.8116617252576593e-05,
1124
+ "loss": 0.7183,
1125
+ "step": 93000
1126
+ },
1127
+ {
1128
+ "epoch": 1.32,
1129
+ "learning_rate": 2.7998964657160337e-05,
1130
+ "loss": 0.7133,
1131
+ "step": 93500
1132
+ },
1133
+ {
1134
+ "epoch": 1.33,
1135
+ "learning_rate": 2.7881312061744087e-05,
1136
+ "loss": 0.717,
1137
+ "step": 94000
1138
+ },
1139
+ {
1140
+ "epoch": 1.33,
1141
+ "learning_rate": 2.776365946632783e-05,
1142
+ "loss": 0.7153,
1143
+ "step": 94500
1144
+ },
1145
+ {
1146
+ "epoch": 1.34,
1147
+ "learning_rate": 2.7646006870911574e-05,
1148
+ "loss": 0.7085,
1149
+ "step": 95000
1150
+ },
1151
+ {
1152
+ "epoch": 1.35,
1153
+ "learning_rate": 2.7528354275495318e-05,
1154
+ "loss": 0.7139,
1155
+ "step": 95500
1156
+ },
1157
+ {
1158
+ "epoch": 1.36,
1159
+ "learning_rate": 2.7410701680079065e-05,
1160
+ "loss": 0.6945,
1161
+ "step": 96000
1162
+ },
1163
+ {
1164
+ "epoch": 1.36,
1165
+ "learning_rate": 2.729304908466281e-05,
1166
+ "loss": 0.7022,
1167
+ "step": 96500
1168
+ },
1169
+ {
1170
+ "epoch": 1.37,
1171
+ "learning_rate": 2.7175396489246552e-05,
1172
+ "loss": 0.689,
1173
+ "step": 97000
1174
+ },
1175
+ {
1176
+ "epoch": 1.38,
1177
+ "learning_rate": 2.7057743893830302e-05,
1178
+ "loss": 0.6878,
1179
+ "step": 97500
1180
+ },
1181
+ {
1182
+ "epoch": 1.38,
1183
+ "learning_rate": 2.6940091298414043e-05,
1184
+ "loss": 0.6872,
1185
+ "step": 98000
1186
+ },
1187
+ {
1188
+ "epoch": 1.39,
1189
+ "learning_rate": 2.6822438702997786e-05,
1190
+ "loss": 0.6927,
1191
+ "step": 98500
1192
+ },
1193
+ {
1194
+ "epoch": 1.4,
1195
+ "learning_rate": 2.6704786107581537e-05,
1196
+ "loss": 0.6944,
1197
+ "step": 99000
1198
+ },
1199
+ {
1200
+ "epoch": 1.4,
1201
+ "learning_rate": 2.658713351216528e-05,
1202
+ "loss": 0.6956,
1203
+ "step": 99500
1204
+ },
1205
+ {
1206
+ "epoch": 1.41,
1207
+ "learning_rate": 2.6469480916749024e-05,
1208
+ "loss": 0.6888,
1209
+ "step": 100000
1210
+ },
1211
+ {
1212
+ "epoch": 1.42,
1213
+ "learning_rate": 2.635182832133277e-05,
1214
+ "loss": 0.6881,
1215
+ "step": 100500
1216
+ },
1217
+ {
1218
+ "epoch": 1.43,
1219
+ "learning_rate": 2.6234175725916514e-05,
1220
+ "loss": 0.6932,
1221
+ "step": 101000
1222
+ },
1223
+ {
1224
+ "epoch": 1.43,
1225
+ "learning_rate": 2.6116523130500258e-05,
1226
+ "loss": 0.691,
1227
+ "step": 101500
1228
+ },
1229
+ {
1230
+ "epoch": 1.44,
1231
+ "learning_rate": 2.599887053508401e-05,
1232
+ "loss": 0.6786,
1233
+ "step": 102000
1234
+ },
1235
+ {
1236
+ "epoch": 1.45,
1237
+ "learning_rate": 2.5881217939667752e-05,
1238
+ "loss": 0.6939,
1239
+ "step": 102500
1240
+ },
1241
+ {
1242
+ "epoch": 1.45,
1243
+ "learning_rate": 2.5763565344251496e-05,
1244
+ "loss": 0.6899,
1245
+ "step": 103000
1246
+ },
1247
+ {
1248
+ "epoch": 1.46,
1249
+ "learning_rate": 2.564591274883524e-05,
1250
+ "loss": 0.694,
1251
+ "step": 103500
1252
+ },
1253
+ {
1254
+ "epoch": 1.47,
1255
+ "learning_rate": 2.5528260153418986e-05,
1256
+ "loss": 0.6734,
1257
+ "step": 104000
1258
+ },
1259
+ {
1260
+ "epoch": 1.48,
1261
+ "learning_rate": 2.541060755800273e-05,
1262
+ "loss": 0.6901,
1263
+ "step": 104500
1264
+ },
1265
+ {
1266
+ "epoch": 1.48,
1267
+ "learning_rate": 2.5292954962586474e-05,
1268
+ "loss": 0.6726,
1269
+ "step": 105000
1270
+ },
1271
+ {
1272
+ "epoch": 1.49,
1273
+ "learning_rate": 2.5175302367170224e-05,
1274
+ "loss": 0.6855,
1275
+ "step": 105500
1276
+ },
1277
+ {
1278
+ "epoch": 1.5,
1279
+ "learning_rate": 2.5057649771753968e-05,
1280
+ "loss": 0.6883,
1281
+ "step": 106000
1282
+ },
1283
+ {
1284
+ "epoch": 1.5,
1285
+ "learning_rate": 2.493999717633771e-05,
1286
+ "loss": 0.6785,
1287
+ "step": 106500
1288
+ },
1289
+ {
1290
+ "epoch": 1.51,
1291
+ "learning_rate": 2.4822344580921455e-05,
1292
+ "loss": 0.6753,
1293
+ "step": 107000
1294
+ },
1295
+ {
1296
+ "epoch": 1.52,
1297
+ "learning_rate": 2.4704691985505202e-05,
1298
+ "loss": 0.6728,
1299
+ "step": 107500
1300
+ },
1301
+ {
1302
+ "epoch": 1.52,
1303
+ "learning_rate": 2.458703939008895e-05,
1304
+ "loss": 0.6804,
1305
+ "step": 108000
1306
+ },
1307
+ {
1308
+ "epoch": 1.53,
1309
+ "learning_rate": 2.4469386794672692e-05,
1310
+ "loss": 0.6792,
1311
+ "step": 108500
1312
+ },
1313
+ {
1314
+ "epoch": 1.54,
1315
+ "learning_rate": 2.4351734199256436e-05,
1316
+ "loss": 0.6722,
1317
+ "step": 109000
1318
+ },
1319
+ {
1320
+ "epoch": 1.55,
1321
+ "learning_rate": 2.423408160384018e-05,
1322
+ "loss": 0.6629,
1323
+ "step": 109500
1324
+ },
1325
+ {
1326
+ "epoch": 1.55,
1327
+ "learning_rate": 2.4116429008423927e-05,
1328
+ "loss": 0.664,
1329
+ "step": 110000
1330
+ },
1331
+ {
1332
+ "epoch": 1.56,
1333
+ "learning_rate": 2.3998776413007674e-05,
1334
+ "loss": 0.6612,
1335
+ "step": 110500
1336
+ },
1337
+ {
1338
+ "epoch": 1.57,
1339
+ "learning_rate": 2.3881123817591417e-05,
1340
+ "loss": 0.6662,
1341
+ "step": 111000
1342
+ },
1343
+ {
1344
+ "epoch": 1.57,
1345
+ "learning_rate": 2.3763471222175164e-05,
1346
+ "loss": 0.6791,
1347
+ "step": 111500
1348
+ },
1349
+ {
1350
+ "epoch": 1.58,
1351
+ "learning_rate": 2.3645818626758908e-05,
1352
+ "loss": 0.6658,
1353
+ "step": 112000
1354
+ },
1355
+ {
1356
+ "epoch": 1.59,
1357
+ "learning_rate": 2.352816603134265e-05,
1358
+ "loss": 0.6676,
1359
+ "step": 112500
1360
+ },
1361
+ {
1362
+ "epoch": 1.6,
1363
+ "learning_rate": 2.34105134359264e-05,
1364
+ "loss": 0.6798,
1365
+ "step": 113000
1366
+ },
1367
+ {
1368
+ "epoch": 1.6,
1369
+ "learning_rate": 2.3292860840510142e-05,
1370
+ "loss": 0.6786,
1371
+ "step": 113500
1372
+ },
1373
+ {
1374
+ "epoch": 1.61,
1375
+ "learning_rate": 2.317520824509389e-05,
1376
+ "loss": 0.661,
1377
+ "step": 114000
1378
+ },
1379
+ {
1380
+ "epoch": 1.62,
1381
+ "learning_rate": 2.3057555649677633e-05,
1382
+ "loss": 0.6662,
1383
+ "step": 114500
1384
+ },
1385
+ {
1386
+ "epoch": 1.62,
1387
+ "learning_rate": 2.2939903054261376e-05,
1388
+ "loss": 0.6649,
1389
+ "step": 115000
1390
+ },
1391
+ {
1392
+ "epoch": 1.63,
1393
+ "learning_rate": 2.2822250458845123e-05,
1394
+ "loss": 0.657,
1395
+ "step": 115500
1396
+ },
1397
+ {
1398
+ "epoch": 1.64,
1399
+ "learning_rate": 2.2704597863428867e-05,
1400
+ "loss": 0.6664,
1401
+ "step": 116000
1402
+ },
1403
+ {
1404
+ "epoch": 1.64,
1405
+ "learning_rate": 2.2586945268012614e-05,
1406
+ "loss": 0.6765,
1407
+ "step": 116500
1408
+ },
1409
+ {
1410
+ "epoch": 1.65,
1411
+ "learning_rate": 2.2469292672596358e-05,
1412
+ "loss": 0.664,
1413
+ "step": 117000
1414
+ },
1415
+ {
1416
+ "epoch": 1.66,
1417
+ "learning_rate": 2.23516400771801e-05,
1418
+ "loss": 0.6664,
1419
+ "step": 117500
1420
+ },
1421
+ {
1422
+ "epoch": 1.67,
1423
+ "learning_rate": 2.2233987481763848e-05,
1424
+ "loss": 0.6532,
1425
+ "step": 118000
1426
+ },
1427
+ {
1428
+ "epoch": 1.67,
1429
+ "learning_rate": 2.2116334886347595e-05,
1430
+ "loss": 0.6687,
1431
+ "step": 118500
1432
+ },
1433
+ {
1434
+ "epoch": 1.68,
1435
+ "learning_rate": 2.199868229093134e-05,
1436
+ "loss": 0.6621,
1437
+ "step": 119000
1438
+ },
1439
+ {
1440
+ "epoch": 1.69,
1441
+ "learning_rate": 2.1881029695515086e-05,
1442
+ "loss": 0.6671,
1443
+ "step": 119500
1444
+ },
1445
+ {
1446
+ "epoch": 1.69,
1447
+ "learning_rate": 2.176337710009883e-05,
1448
+ "loss": 0.6619,
1449
+ "step": 120000
1450
+ },
1451
+ {
1452
+ "epoch": 1.7,
1453
+ "learning_rate": 2.1645724504682573e-05,
1454
+ "loss": 0.6549,
1455
+ "step": 120500
1456
+ },
1457
+ {
1458
+ "epoch": 1.71,
1459
+ "learning_rate": 2.152807190926632e-05,
1460
+ "loss": 0.6668,
1461
+ "step": 121000
1462
+ },
1463
+ {
1464
+ "epoch": 1.72,
1465
+ "learning_rate": 2.1410419313850064e-05,
1466
+ "loss": 0.6439,
1467
+ "step": 121500
1468
+ },
1469
+ {
1470
+ "epoch": 1.72,
1471
+ "learning_rate": 2.129276671843381e-05,
1472
+ "loss": 0.6554,
1473
+ "step": 122000
1474
+ },
1475
+ {
1476
+ "epoch": 1.73,
1477
+ "learning_rate": 2.1175114123017554e-05,
1478
+ "loss": 0.6395,
1479
+ "step": 122500
1480
+ },
1481
+ {
1482
+ "epoch": 1.74,
1483
+ "learning_rate": 2.1057461527601298e-05,
1484
+ "loss": 0.6482,
1485
+ "step": 123000
1486
+ },
1487
+ {
1488
+ "epoch": 1.74,
1489
+ "learning_rate": 2.0939808932185045e-05,
1490
+ "loss": 0.6607,
1491
+ "step": 123500
1492
+ },
1493
+ {
1494
+ "epoch": 1.75,
1495
+ "learning_rate": 2.082215633676879e-05,
1496
+ "loss": 0.6508,
1497
+ "step": 124000
1498
+ },
1499
+ {
1500
+ "epoch": 1.76,
1501
+ "learning_rate": 2.0704503741352536e-05,
1502
+ "loss": 0.6438,
1503
+ "step": 124500
1504
+ },
1505
+ {
1506
+ "epoch": 1.76,
1507
+ "learning_rate": 2.0586851145936283e-05,
1508
+ "loss": 0.658,
1509
+ "step": 125000
1510
+ },
1511
+ {
1512
+ "epoch": 1.77,
1513
+ "learning_rate": 2.0469198550520026e-05,
1514
+ "loss": 0.6409,
1515
+ "step": 125500
1516
+ },
1517
+ {
1518
+ "epoch": 1.78,
1519
+ "learning_rate": 2.035154595510377e-05,
1520
+ "loss": 0.6439,
1521
+ "step": 126000
1522
+ },
1523
+ {
1524
+ "epoch": 1.79,
1525
+ "learning_rate": 2.0233893359687517e-05,
1526
+ "loss": 0.6674,
1527
+ "step": 126500
1528
+ },
1529
+ {
1530
+ "epoch": 1.79,
1531
+ "learning_rate": 2.011624076427126e-05,
1532
+ "loss": 0.66,
1533
+ "step": 127000
1534
+ },
1535
+ {
1536
+ "epoch": 1.8,
1537
+ "learning_rate": 1.9998588168855007e-05,
1538
+ "loss": 0.6576,
1539
+ "step": 127500
1540
+ },
1541
+ {
1542
+ "epoch": 1.81,
1543
+ "learning_rate": 1.988093557343875e-05,
1544
+ "loss": 0.6445,
1545
+ "step": 128000
1546
+ },
1547
+ {
1548
+ "epoch": 1.81,
1549
+ "learning_rate": 1.9763282978022495e-05,
1550
+ "loss": 0.6542,
1551
+ "step": 128500
1552
+ },
1553
+ {
1554
+ "epoch": 1.82,
1555
+ "learning_rate": 1.964563038260624e-05,
1556
+ "loss": 0.6412,
1557
+ "step": 129000
1558
+ },
1559
+ {
1560
+ "epoch": 1.83,
1561
+ "learning_rate": 1.9527977787189985e-05,
1562
+ "loss": 0.6496,
1563
+ "step": 129500
1564
+ },
1565
+ {
1566
+ "epoch": 1.84,
1567
+ "learning_rate": 1.9410325191773732e-05,
1568
+ "loss": 0.6423,
1569
+ "step": 130000
1570
+ },
1571
+ {
1572
+ "epoch": 1.84,
1573
+ "learning_rate": 1.929267259635748e-05,
1574
+ "loss": 0.6437,
1575
+ "step": 130500
1576
+ },
1577
+ {
1578
+ "epoch": 1.85,
1579
+ "learning_rate": 1.917502000094122e-05,
1580
+ "loss": 0.6467,
1581
+ "step": 131000
1582
+ },
1583
+ {
1584
+ "epoch": 1.86,
1585
+ "learning_rate": 1.9057367405524967e-05,
1586
+ "loss": 0.6481,
1587
+ "step": 131500
1588
+ },
1589
+ {
1590
+ "epoch": 1.86,
1591
+ "learning_rate": 1.893971481010871e-05,
1592
+ "loss": 0.6442,
1593
+ "step": 132000
1594
+ },
1595
+ {
1596
+ "epoch": 1.87,
1597
+ "learning_rate": 1.8822062214692457e-05,
1598
+ "loss": 0.6345,
1599
+ "step": 132500
1600
+ },
1601
+ {
1602
+ "epoch": 1.88,
1603
+ "learning_rate": 1.8704409619276204e-05,
1604
+ "loss": 0.642,
1605
+ "step": 133000
1606
+ },
1607
+ {
1608
+ "epoch": 1.88,
1609
+ "learning_rate": 1.8586757023859948e-05,
1610
+ "loss": 0.6373,
1611
+ "step": 133500
1612
+ },
1613
+ {
1614
+ "epoch": 1.89,
1615
+ "learning_rate": 1.846910442844369e-05,
1616
+ "loss": 0.6461,
1617
+ "step": 134000
1618
+ },
1619
+ {
1620
+ "epoch": 1.9,
1621
+ "learning_rate": 1.835145183302744e-05,
1622
+ "loss": 0.6429,
1623
+ "step": 134500
1624
+ },
1625
+ {
1626
+ "epoch": 1.91,
1627
+ "learning_rate": 1.8233799237611182e-05,
1628
+ "loss": 0.6389,
1629
+ "step": 135000
1630
+ },
1631
+ {
1632
+ "epoch": 1.91,
1633
+ "learning_rate": 1.811614664219493e-05,
1634
+ "loss": 0.6395,
1635
+ "step": 135500
1636
+ },
1637
+ {
1638
+ "epoch": 1.92,
1639
+ "learning_rate": 1.7998494046778673e-05,
1640
+ "loss": 0.6502,
1641
+ "step": 136000
1642
+ },
1643
+ {
1644
+ "epoch": 1.93,
1645
+ "learning_rate": 1.7880841451362416e-05,
1646
+ "loss": 0.6417,
1647
+ "step": 136500
1648
+ },
1649
+ {
1650
+ "epoch": 1.93,
1651
+ "learning_rate": 1.7763188855946163e-05,
1652
+ "loss": 0.6376,
1653
+ "step": 137000
1654
+ },
1655
+ {
1656
+ "epoch": 1.94,
1657
+ "learning_rate": 1.7645536260529907e-05,
1658
+ "loss": 0.6368,
1659
+ "step": 137500
1660
+ },
1661
+ {
1662
+ "epoch": 1.95,
1663
+ "learning_rate": 1.7527883665113654e-05,
1664
+ "loss": 0.638,
1665
+ "step": 138000
1666
+ },
1667
+ {
1668
+ "epoch": 1.96,
1669
+ "learning_rate": 1.74102310696974e-05,
1670
+ "loss": 0.6508,
1671
+ "step": 138500
1672
+ },
1673
+ {
1674
+ "epoch": 1.96,
1675
+ "learning_rate": 1.7292578474281145e-05,
1676
+ "loss": 0.6311,
1677
+ "step": 139000
1678
+ },
1679
+ {
1680
+ "epoch": 1.97,
1681
+ "learning_rate": 1.7174925878864888e-05,
1682
+ "loss": 0.641,
1683
+ "step": 139500
1684
+ },
1685
+ {
1686
+ "epoch": 1.98,
1687
+ "learning_rate": 1.7057273283448632e-05,
1688
+ "loss": 0.638,
1689
+ "step": 140000
1690
+ },
1691
+ {
1692
+ "epoch": 1.98,
1693
+ "learning_rate": 1.693962068803238e-05,
1694
+ "loss": 0.6176,
1695
+ "step": 140500
1696
+ },
1697
+ {
1698
+ "epoch": 1.99,
1699
+ "learning_rate": 1.6821968092616126e-05,
1700
+ "loss": 0.6346,
1701
+ "step": 141000
1702
+ },
1703
+ {
1704
+ "epoch": 2.0,
1705
+ "learning_rate": 1.670431549719987e-05,
1706
+ "loss": 0.6348,
1707
+ "step": 141500
1708
+ },
1709
+ {
1710
+ "epoch": 2.0,
1711
+ "learning_rate": 1.6586662901783613e-05,
1712
+ "loss": 0.6454,
1713
+ "step": 142000
1714
+ },
1715
+ {
1716
+ "epoch": 2.01,
1717
+ "learning_rate": 1.6469010306367357e-05,
1718
+ "loss": 0.6237,
1719
+ "step": 142500
1720
+ },
1721
+ {
1722
+ "epoch": 2.02,
1723
+ "learning_rate": 1.6351357710951104e-05,
1724
+ "loss": 0.622,
1725
+ "step": 143000
1726
+ },
1727
+ {
1728
+ "epoch": 2.03,
1729
+ "learning_rate": 1.623370511553485e-05,
1730
+ "loss": 0.6112,
1731
+ "step": 143500
1732
+ },
1733
+ {
1734
+ "epoch": 2.03,
1735
+ "learning_rate": 1.6116052520118594e-05,
1736
+ "loss": 0.6288,
1737
+ "step": 144000
1738
+ },
1739
+ {
1740
+ "epoch": 2.04,
1741
+ "learning_rate": 1.599839992470234e-05,
1742
+ "loss": 0.641,
1743
+ "step": 144500
1744
+ },
1745
+ {
1746
+ "epoch": 2.05,
1747
+ "learning_rate": 1.5880747329286085e-05,
1748
+ "loss": 0.6298,
1749
+ "step": 145000
1750
+ },
1751
+ {
1752
+ "epoch": 2.05,
1753
+ "learning_rate": 1.576309473386983e-05,
1754
+ "loss": 0.6144,
1755
+ "step": 145500
1756
+ },
1757
+ {
1758
+ "epoch": 2.06,
1759
+ "learning_rate": 1.5645442138453575e-05,
1760
+ "loss": 0.6261,
1761
+ "step": 146000
1762
+ },
1763
+ {
1764
+ "epoch": 2.07,
1765
+ "learning_rate": 1.552778954303732e-05,
1766
+ "loss": 0.6195,
1767
+ "step": 146500
1768
+ },
1769
+ {
1770
+ "epoch": 2.08,
1771
+ "learning_rate": 1.5410136947621066e-05,
1772
+ "loss": 0.6249,
1773
+ "step": 147000
1774
+ },
1775
+ {
1776
+ "epoch": 2.08,
1777
+ "learning_rate": 1.529248435220481e-05,
1778
+ "loss": 0.6395,
1779
+ "step": 147500
1780
+ },
1781
+ {
1782
+ "epoch": 2.09,
1783
+ "learning_rate": 1.5174831756788555e-05,
1784
+ "loss": 0.6273,
1785
+ "step": 148000
1786
+ },
1787
+ {
1788
+ "epoch": 2.1,
1789
+ "learning_rate": 1.50571791613723e-05,
1790
+ "loss": 0.6279,
1791
+ "step": 148500
1792
+ },
1793
+ {
1794
+ "epoch": 2.1,
1795
+ "learning_rate": 1.4939526565956047e-05,
1796
+ "loss": 0.6157,
1797
+ "step": 149000
1798
+ },
1799
+ {
1800
+ "epoch": 2.11,
1801
+ "learning_rate": 1.4821873970539791e-05,
1802
+ "loss": 0.6238,
1803
+ "step": 149500
1804
+ },
1805
+ {
1806
+ "epoch": 2.12,
1807
+ "learning_rate": 1.4704221375123536e-05,
1808
+ "loss": 0.6222,
1809
+ "step": 150000
1810
+ },
1811
+ {
1812
+ "epoch": 2.12,
1813
+ "learning_rate": 1.458656877970728e-05,
1814
+ "loss": 0.6292,
1815
+ "step": 150500
1816
+ },
1817
+ {
1818
+ "epoch": 2.13,
1819
+ "learning_rate": 1.4468916184291025e-05,
1820
+ "loss": 0.6159,
1821
+ "step": 151000
1822
+ },
1823
+ {
1824
+ "epoch": 2.14,
1825
+ "learning_rate": 1.4351263588874772e-05,
1826
+ "loss": 0.6197,
1827
+ "step": 151500
1828
+ },
1829
+ {
1830
+ "epoch": 2.15,
1831
+ "learning_rate": 1.4233610993458516e-05,
1832
+ "loss": 0.6247,
1833
+ "step": 152000
1834
+ },
1835
+ {
1836
+ "epoch": 2.15,
1837
+ "learning_rate": 1.4115958398042261e-05,
1838
+ "loss": 0.6152,
1839
+ "step": 152500
1840
+ },
1841
+ {
1842
+ "epoch": 2.16,
1843
+ "learning_rate": 1.3998305802626008e-05,
1844
+ "loss": 0.6279,
1845
+ "step": 153000
1846
+ },
1847
+ {
1848
+ "epoch": 2.17,
1849
+ "learning_rate": 1.3880653207209752e-05,
1850
+ "loss": 0.6107,
1851
+ "step": 153500
1852
+ },
1853
+ {
1854
+ "epoch": 2.17,
1855
+ "learning_rate": 1.3763000611793497e-05,
1856
+ "loss": 0.62,
1857
+ "step": 154000
1858
+ },
1859
+ {
1860
+ "epoch": 2.18,
1861
+ "learning_rate": 1.364534801637724e-05,
1862
+ "loss": 0.6009,
1863
+ "step": 154500
1864
+ },
1865
+ {
1866
+ "epoch": 2.19,
1867
+ "learning_rate": 1.3527695420960988e-05,
1868
+ "loss": 0.6217,
1869
+ "step": 155000
1870
+ },
1871
+ {
1872
+ "epoch": 2.2,
1873
+ "learning_rate": 1.3410042825544733e-05,
1874
+ "loss": 0.613,
1875
+ "step": 155500
1876
+ },
1877
+ {
1878
+ "epoch": 2.2,
1879
+ "learning_rate": 1.3292390230128477e-05,
1880
+ "loss": 0.6257,
1881
+ "step": 156000
1882
+ },
1883
+ {
1884
+ "epoch": 2.21,
1885
+ "learning_rate": 1.3174737634712222e-05,
1886
+ "loss": 0.6257,
1887
+ "step": 156500
1888
+ },
1889
+ {
1890
+ "epoch": 2.22,
1891
+ "learning_rate": 1.3057085039295969e-05,
1892
+ "loss": 0.6184,
1893
+ "step": 157000
1894
+ },
1895
+ {
1896
+ "epoch": 2.22,
1897
+ "learning_rate": 1.2939432443879713e-05,
1898
+ "loss": 0.6118,
1899
+ "step": 157500
1900
+ },
1901
+ {
1902
+ "epoch": 2.23,
1903
+ "learning_rate": 1.2821779848463458e-05,
1904
+ "loss": 0.5928,
1905
+ "step": 158000
1906
+ },
1907
+ {
1908
+ "epoch": 2.24,
1909
+ "learning_rate": 1.2704127253047201e-05,
1910
+ "loss": 0.6092,
1911
+ "step": 158500
1912
+ },
1913
+ {
1914
+ "epoch": 2.24,
1915
+ "learning_rate": 1.2586474657630948e-05,
1916
+ "loss": 0.6055,
1917
+ "step": 159000
1918
+ },
1919
+ {
1920
+ "epoch": 2.25,
1921
+ "learning_rate": 1.2468822062214692e-05,
1922
+ "loss": 0.6102,
1923
+ "step": 159500
1924
+ },
1925
+ {
1926
+ "epoch": 2.26,
1927
+ "learning_rate": 1.2351169466798439e-05,
1928
+ "loss": 0.6194,
1929
+ "step": 160000
1930
+ },
1931
+ {
1932
+ "epoch": 2.27,
1933
+ "learning_rate": 1.2233516871382183e-05,
1934
+ "loss": 0.6108,
1935
+ "step": 160500
1936
+ },
1937
+ {
1938
+ "epoch": 2.27,
1939
+ "learning_rate": 1.2115864275965928e-05,
1940
+ "loss": 0.6091,
1941
+ "step": 161000
1942
+ },
1943
+ {
1944
+ "epoch": 2.28,
1945
+ "learning_rate": 1.1998211680549673e-05,
1946
+ "loss": 0.6019,
1947
+ "step": 161500
1948
+ },
1949
+ {
1950
+ "epoch": 2.29,
1951
+ "learning_rate": 1.1880559085133419e-05,
1952
+ "loss": 0.6176,
1953
+ "step": 162000
1954
+ },
1955
+ {
1956
+ "epoch": 2.29,
1957
+ "learning_rate": 1.1762906489717164e-05,
1958
+ "loss": 0.6149,
1959
+ "step": 162500
1960
+ },
1961
+ {
1962
+ "epoch": 2.3,
1963
+ "learning_rate": 1.164525389430091e-05,
1964
+ "loss": 0.5845,
1965
+ "step": 163000
1966
+ },
1967
+ {
1968
+ "epoch": 2.31,
1969
+ "learning_rate": 1.1527601298884653e-05,
1970
+ "loss": 0.6121,
1971
+ "step": 163500
1972
+ },
1973
+ {
1974
+ "epoch": 2.32,
1975
+ "learning_rate": 1.14099487034684e-05,
1976
+ "loss": 0.6092,
1977
+ "step": 164000
1978
+ },
1979
+ {
1980
+ "epoch": 2.32,
1981
+ "learning_rate": 1.1292296108052145e-05,
1982
+ "loss": 0.606,
1983
+ "step": 164500
1984
+ },
1985
+ {
1986
+ "epoch": 2.33,
1987
+ "learning_rate": 1.1174643512635889e-05,
1988
+ "loss": 0.6011,
1989
+ "step": 165000
1990
+ },
1991
+ {
1992
+ "epoch": 2.34,
1993
+ "learning_rate": 1.1056990917219634e-05,
1994
+ "loss": 0.6015,
1995
+ "step": 165500
1996
+ },
1997
+ {
1998
+ "epoch": 2.34,
1999
+ "learning_rate": 1.093933832180338e-05,
2000
+ "loss": 0.6132,
2001
+ "step": 166000
2002
+ },
2003
+ {
2004
+ "epoch": 2.35,
2005
+ "learning_rate": 1.0821685726387125e-05,
2006
+ "loss": 0.6025,
2007
+ "step": 166500
2008
+ },
2009
+ {
2010
+ "epoch": 2.36,
2011
+ "learning_rate": 1.070403313097087e-05,
2012
+ "loss": 0.6099,
2013
+ "step": 167000
2014
+ },
2015
+ {
2016
+ "epoch": 2.36,
2017
+ "learning_rate": 1.0586380535554614e-05,
2018
+ "loss": 0.6172,
2019
+ "step": 167500
2020
+ },
2021
+ {
2022
+ "epoch": 2.37,
2023
+ "learning_rate": 1.046872794013836e-05,
2024
+ "loss": 0.6132,
2025
+ "step": 168000
2026
+ },
2027
+ {
2028
+ "epoch": 2.38,
2029
+ "learning_rate": 1.0351075344722106e-05,
2030
+ "loss": 0.5979,
2031
+ "step": 168500
2032
+ },
2033
+ {
2034
+ "epoch": 2.39,
2035
+ "learning_rate": 1.023342274930585e-05,
2036
+ "loss": 0.6012,
2037
+ "step": 169000
2038
+ },
2039
+ {
2040
+ "epoch": 2.39,
2041
+ "learning_rate": 1.0115770153889595e-05,
2042
+ "loss": 0.5921,
2043
+ "step": 169500
2044
+ },
2045
+ {
2046
+ "epoch": 2.4,
2047
+ "learning_rate": 9.998117558473342e-06,
2048
+ "loss": 0.6039,
2049
+ "step": 170000
2050
+ },
2051
+ {
2052
+ "epoch": 2.41,
2053
+ "learning_rate": 9.880464963057086e-06,
2054
+ "loss": 0.6064,
2055
+ "step": 170500
2056
+ },
2057
+ {
2058
+ "epoch": 2.41,
2059
+ "learning_rate": 9.76281236764083e-06,
2060
+ "loss": 0.6084,
2061
+ "step": 171000
2062
+ },
2063
+ {
2064
+ "epoch": 2.42,
2065
+ "learning_rate": 9.645159772224576e-06,
2066
+ "loss": 0.5933,
2067
+ "step": 171500
2068
+ },
2069
+ {
2070
+ "epoch": 2.43,
2071
+ "learning_rate": 9.52750717680832e-06,
2072
+ "loss": 0.6008,
2073
+ "step": 172000
2074
+ },
2075
+ {
2076
+ "epoch": 2.44,
2077
+ "learning_rate": 9.409854581392067e-06,
2078
+ "loss": 0.6048,
2079
+ "step": 172500
2080
+ },
2081
+ {
2082
+ "epoch": 2.44,
2083
+ "learning_rate": 9.29220198597581e-06,
2084
+ "loss": 0.5933,
2085
+ "step": 173000
2086
+ },
2087
+ {
2088
+ "epoch": 2.45,
2089
+ "learning_rate": 9.174549390559556e-06,
2090
+ "loss": 0.5908,
2091
+ "step": 173500
2092
+ },
2093
+ {
2094
+ "epoch": 2.46,
2095
+ "learning_rate": 9.056896795143301e-06,
2096
+ "loss": 0.5921,
2097
+ "step": 174000
2098
+ },
2099
+ {
2100
+ "epoch": 2.46,
2101
+ "learning_rate": 8.939244199727046e-06,
2102
+ "loss": 0.6012,
2103
+ "step": 174500
2104
+ },
2105
+ {
2106
+ "epoch": 2.47,
2107
+ "learning_rate": 8.821591604310792e-06,
2108
+ "loss": 0.5988,
2109
+ "step": 175000
2110
+ },
2111
+ {
2112
+ "epoch": 2.48,
2113
+ "learning_rate": 8.703939008894537e-06,
2114
+ "loss": 0.6019,
2115
+ "step": 175500
2116
+ },
2117
+ {
2118
+ "epoch": 2.48,
2119
+ "learning_rate": 8.58628641347828e-06,
2120
+ "loss": 0.5915,
2121
+ "step": 176000
2122
+ },
2123
+ {
2124
+ "epoch": 2.49,
2125
+ "learning_rate": 8.468633818062028e-06,
2126
+ "loss": 0.6104,
2127
+ "step": 176500
2128
+ },
2129
+ {
2130
+ "epoch": 2.5,
2131
+ "learning_rate": 8.350981222645773e-06,
2132
+ "loss": 0.6093,
2133
+ "step": 177000
2134
+ },
2135
+ {
2136
+ "epoch": 2.51,
2137
+ "learning_rate": 8.233328627229516e-06,
2138
+ "loss": 0.6084,
2139
+ "step": 177500
2140
+ },
2141
+ {
2142
+ "epoch": 2.51,
2143
+ "learning_rate": 8.115676031813262e-06,
2144
+ "loss": 0.5989,
2145
+ "step": 178000
2146
+ },
2147
+ {
2148
+ "epoch": 2.52,
2149
+ "learning_rate": 7.998023436397007e-06,
2150
+ "loss": 0.6088,
2151
+ "step": 178500
2152
+ },
2153
+ {
2154
+ "epoch": 2.53,
2155
+ "learning_rate": 7.880370840980752e-06,
2156
+ "loss": 0.5889,
2157
+ "step": 179000
2158
+ },
2159
+ {
2160
+ "epoch": 2.53,
2161
+ "learning_rate": 7.762718245564498e-06,
2162
+ "loss": 0.5819,
2163
+ "step": 179500
2164
+ },
2165
+ {
2166
+ "epoch": 2.54,
2167
+ "learning_rate": 7.645065650148241e-06,
2168
+ "loss": 0.5901,
2169
+ "step": 180000
2170
+ },
2171
+ {
2172
+ "epoch": 2.55,
2173
+ "learning_rate": 7.527413054731988e-06,
2174
+ "loss": 0.58,
2175
+ "step": 180500
2176
+ },
2177
+ {
2178
+ "epoch": 2.56,
2179
+ "learning_rate": 7.409760459315733e-06,
2180
+ "loss": 0.5963,
2181
+ "step": 181000
2182
+ },
2183
+ {
2184
+ "epoch": 2.56,
2185
+ "learning_rate": 7.292107863899478e-06,
2186
+ "loss": 0.604,
2187
+ "step": 181500
2188
+ },
2189
+ {
2190
+ "epoch": 2.57,
2191
+ "learning_rate": 7.174455268483223e-06,
2192
+ "loss": 0.5984,
2193
+ "step": 182000
2194
+ },
2195
+ {
2196
+ "epoch": 2.58,
2197
+ "learning_rate": 7.056802673066969e-06,
2198
+ "loss": 0.591,
2199
+ "step": 182500
2200
+ },
2201
+ {
2202
+ "epoch": 2.58,
2203
+ "learning_rate": 6.939150077650713e-06,
2204
+ "loss": 0.5932,
2205
+ "step": 183000
2206
+ },
2207
+ {
2208
+ "epoch": 2.59,
2209
+ "learning_rate": 6.8214974822344585e-06,
2210
+ "loss": 0.5996,
2211
+ "step": 183500
2212
+ },
2213
+ {
2214
+ "epoch": 2.6,
2215
+ "learning_rate": 6.703844886818203e-06,
2216
+ "loss": 0.5987,
2217
+ "step": 184000
2218
+ },
2219
+ {
2220
+ "epoch": 2.6,
2221
+ "learning_rate": 6.586192291401949e-06,
2222
+ "loss": 0.5859,
2223
+ "step": 184500
2224
+ },
2225
+ {
2226
+ "epoch": 2.61,
2227
+ "learning_rate": 6.468539695985694e-06,
2228
+ "loss": 0.594,
2229
+ "step": 185000
2230
+ },
2231
+ {
2232
+ "epoch": 2.62,
2233
+ "learning_rate": 6.350887100569439e-06,
2234
+ "loss": 0.5836,
2235
+ "step": 185500
2236
+ },
2237
+ {
2238
+ "epoch": 2.63,
2239
+ "learning_rate": 6.233234505153184e-06,
2240
+ "loss": 0.5789,
2241
+ "step": 186000
2242
+ },
2243
+ {
2244
+ "epoch": 2.63,
2245
+ "learning_rate": 6.115581909736929e-06,
2246
+ "loss": 0.595,
2247
+ "step": 186500
2248
+ },
2249
+ {
2250
+ "epoch": 2.64,
2251
+ "learning_rate": 5.997929314320674e-06,
2252
+ "loss": 0.5817,
2253
+ "step": 187000
2254
+ },
2255
+ {
2256
+ "epoch": 2.65,
2257
+ "learning_rate": 5.880276718904419e-06,
2258
+ "loss": 0.5926,
2259
+ "step": 187500
2260
+ },
2261
+ {
2262
+ "epoch": 2.65,
2263
+ "learning_rate": 5.762624123488165e-06,
2264
+ "loss": 0.6029,
2265
+ "step": 188000
2266
+ },
2267
+ {
2268
+ "epoch": 2.66,
2269
+ "learning_rate": 5.644971528071909e-06,
2270
+ "loss": 0.5888,
2271
+ "step": 188500
2272
+ },
2273
+ {
2274
+ "epoch": 2.67,
2275
+ "learning_rate": 5.527318932655655e-06,
2276
+ "loss": 0.5843,
2277
+ "step": 189000
2278
+ },
2279
+ {
2280
+ "epoch": 2.68,
2281
+ "learning_rate": 5.4096663372394e-06,
2282
+ "loss": 0.5794,
2283
+ "step": 189500
2284
+ },
2285
+ {
2286
+ "epoch": 2.68,
2287
+ "learning_rate": 5.292013741823145e-06,
2288
+ "loss": 0.5829,
2289
+ "step": 190000
2290
+ },
2291
+ {
2292
+ "epoch": 2.69,
2293
+ "learning_rate": 5.1743611464068895e-06,
2294
+ "loss": 0.5903,
2295
+ "step": 190500
2296
+ },
2297
+ {
2298
+ "epoch": 2.7,
2299
+ "learning_rate": 5.056708550990636e-06,
2300
+ "loss": 0.5742,
2301
+ "step": 191000
2302
+ },
2303
+ {
2304
+ "epoch": 2.7,
2305
+ "learning_rate": 4.93905595557438e-06,
2306
+ "loss": 0.5948,
2307
+ "step": 191500
2308
+ },
2309
+ {
2310
+ "epoch": 2.71,
2311
+ "learning_rate": 4.821403360158125e-06,
2312
+ "loss": 0.5807,
2313
+ "step": 192000
2314
+ },
2315
+ {
2316
+ "epoch": 2.72,
2317
+ "learning_rate": 4.703750764741871e-06,
2318
+ "loss": 0.5824,
2319
+ "step": 192500
2320
+ },
2321
+ {
2322
+ "epoch": 2.72,
2323
+ "learning_rate": 4.586098169325616e-06,
2324
+ "loss": 0.5818,
2325
+ "step": 193000
2326
+ },
2327
+ {
2328
+ "epoch": 2.73,
2329
+ "learning_rate": 4.4684455739093605e-06,
2330
+ "loss": 0.5871,
2331
+ "step": 193500
2332
+ },
2333
+ {
2334
+ "epoch": 2.74,
2335
+ "learning_rate": 4.350792978493106e-06,
2336
+ "loss": 0.5814,
2337
+ "step": 194000
2338
+ },
2339
+ {
2340
+ "epoch": 2.75,
2341
+ "learning_rate": 4.233140383076851e-06,
2342
+ "loss": 0.5851,
2343
+ "step": 194500
2344
+ },
2345
+ {
2346
+ "epoch": 2.75,
2347
+ "learning_rate": 4.1154877876605964e-06,
2348
+ "loss": 0.5794,
2349
+ "step": 195000
2350
+ },
2351
+ {
2352
+ "epoch": 2.76,
2353
+ "learning_rate": 3.997835192244341e-06,
2354
+ "loss": 0.5914,
2355
+ "step": 195500
2356
+ },
2357
+ {
2358
+ "epoch": 2.77,
2359
+ "learning_rate": 3.880182596828086e-06,
2360
+ "loss": 0.5792,
2361
+ "step": 196000
2362
+ },
2363
+ {
2364
+ "epoch": 2.77,
2365
+ "learning_rate": 3.762530001411831e-06,
2366
+ "loss": 0.5941,
2367
+ "step": 196500
2368
+ },
2369
+ {
2370
+ "epoch": 2.78,
2371
+ "learning_rate": 3.644877405995577e-06,
2372
+ "loss": 0.5673,
2373
+ "step": 197000
2374
+ },
2375
+ {
2376
+ "epoch": 2.79,
2377
+ "learning_rate": 3.5272248105793217e-06,
2378
+ "loss": 0.5818,
2379
+ "step": 197500
2380
+ },
2381
+ {
2382
+ "epoch": 2.8,
2383
+ "learning_rate": 3.409572215163067e-06,
2384
+ "loss": 0.5828,
2385
+ "step": 198000
2386
+ },
2387
+ {
2388
+ "epoch": 2.8,
2389
+ "learning_rate": 3.291919619746812e-06,
2390
+ "loss": 0.5798,
2391
+ "step": 198500
2392
+ },
2393
+ {
2394
+ "epoch": 2.81,
2395
+ "learning_rate": 3.1742670243305572e-06,
2396
+ "loss": 0.5714,
2397
+ "step": 199000
2398
+ },
2399
+ {
2400
+ "epoch": 2.82,
2401
+ "learning_rate": 3.056614428914302e-06,
2402
+ "loss": 0.5866,
2403
+ "step": 199500
2404
+ },
2405
+ {
2406
+ "epoch": 2.82,
2407
+ "learning_rate": 2.938961833498047e-06,
2408
+ "loss": 0.578,
2409
+ "step": 200000
2410
+ },
2411
+ {
2412
+ "epoch": 2.83,
2413
+ "learning_rate": 2.8213092380817923e-06,
2414
+ "loss": 0.5924,
2415
+ "step": 200500
2416
+ },
2417
+ {
2418
+ "epoch": 2.84,
2419
+ "learning_rate": 2.703656642665537e-06,
2420
+ "loss": 0.5854,
2421
+ "step": 201000
2422
+ },
2423
+ {
2424
+ "epoch": 2.84,
2425
+ "learning_rate": 2.5860040472492825e-06,
2426
+ "loss": 0.5865,
2427
+ "step": 201500
2428
+ },
2429
+ {
2430
+ "epoch": 2.85,
2431
+ "learning_rate": 2.4683514518330274e-06,
2432
+ "loss": 0.5933,
2433
+ "step": 202000
2434
+ },
2435
+ {
2436
+ "epoch": 2.86,
2437
+ "learning_rate": 2.3506988564167727e-06,
2438
+ "loss": 0.5748,
2439
+ "step": 202500
2440
+ },
2441
+ {
2442
+ "epoch": 2.87,
2443
+ "learning_rate": 2.233046261000518e-06,
2444
+ "loss": 0.5913,
2445
+ "step": 203000
2446
+ },
2447
+ {
2448
+ "epoch": 2.87,
2449
+ "learning_rate": 2.115393665584263e-06,
2450
+ "loss": 0.5811,
2451
+ "step": 203500
2452
+ },
2453
+ {
2454
+ "epoch": 2.88,
2455
+ "learning_rate": 1.997741070168008e-06,
2456
+ "loss": 0.5835,
2457
+ "step": 204000
2458
+ },
2459
+ {
2460
+ "epoch": 2.89,
2461
+ "learning_rate": 1.880088474751753e-06,
2462
+ "loss": 0.5836,
2463
+ "step": 204500
2464
+ },
2465
+ {
2466
+ "epoch": 2.89,
2467
+ "learning_rate": 1.7624358793354984e-06,
2468
+ "loss": 0.5737,
2469
+ "step": 205000
2470
+ },
2471
+ {
2472
+ "epoch": 2.9,
2473
+ "learning_rate": 1.6447832839192435e-06,
2474
+ "loss": 0.5813,
2475
+ "step": 205500
2476
+ },
2477
+ {
2478
+ "epoch": 2.91,
2479
+ "learning_rate": 1.5271306885029886e-06,
2480
+ "loss": 0.5837,
2481
+ "step": 206000
2482
+ },
2483
+ {
2484
+ "epoch": 2.92,
2485
+ "learning_rate": 1.4094780930867337e-06,
2486
+ "loss": 0.5816,
2487
+ "step": 206500
2488
+ },
2489
+ {
2490
+ "epoch": 2.92,
2491
+ "learning_rate": 1.2918254976704786e-06,
2492
+ "loss": 0.5871,
2493
+ "step": 207000
2494
+ },
2495
+ {
2496
+ "epoch": 2.93,
2497
+ "learning_rate": 1.1741729022542237e-06,
2498
+ "loss": 0.5657,
2499
+ "step": 207500
2500
+ },
2501
+ {
2502
+ "epoch": 2.94,
2503
+ "learning_rate": 1.0565203068379688e-06,
2504
+ "loss": 0.584,
2505
+ "step": 208000
2506
+ },
2507
+ {
2508
+ "epoch": 2.94,
2509
+ "learning_rate": 9.38867711421714e-07,
2510
+ "loss": 0.5672,
2511
+ "step": 208500
2512
+ },
2513
+ {
2514
+ "epoch": 2.95,
2515
+ "learning_rate": 8.212151160054591e-07,
2516
+ "loss": 0.5814,
2517
+ "step": 209000
2518
+ },
2519
+ {
2520
+ "epoch": 2.96,
2521
+ "learning_rate": 7.035625205892043e-07,
2522
+ "loss": 0.579,
2523
+ "step": 209500
2524
+ },
2525
+ {
2526
+ "epoch": 2.96,
2527
+ "learning_rate": 5.859099251729494e-07,
2528
+ "loss": 0.5969,
2529
+ "step": 210000
2530
+ },
2531
+ {
2532
+ "epoch": 2.97,
2533
+ "learning_rate": 4.6825732975669443e-07,
2534
+ "loss": 0.5796,
2535
+ "step": 210500
2536
+ },
2537
+ {
2538
+ "epoch": 2.98,
2539
+ "learning_rate": 3.506047343404396e-07,
2540
+ "loss": 0.5699,
2541
+ "step": 211000
2542
+ },
2543
+ {
2544
+ "epoch": 2.99,
2545
+ "learning_rate": 2.3295213892418468e-07,
2546
+ "loss": 0.5749,
2547
+ "step": 211500
2548
+ },
2549
+ {
2550
+ "epoch": 2.99,
2551
+ "learning_rate": 1.152995435079298e-07,
2552
+ "loss": 0.5678,
2553
+ "step": 212000
2554
+ },
2555
+ {
2556
+ "epoch": 3.0,
2557
+ "step": 212490,
2558
+ "total_flos": 4.485693256017408e+17,
2559
+ "train_loss": 0.7508337134063662,
2560
+ "train_runtime": 150510.9458,
2561
+ "train_samples_per_second": 11.294,
2562
+ "train_steps_per_second": 1.412
2563
+ }
2564
+ ],
2565
+ "logging_steps": 500,
2566
+ "max_steps": 212490,
2567
+ "num_train_epochs": 3,
2568
+ "save_steps": 100000,
2569
+ "total_flos": 4.485693256017408e+17,
2570
+ "trial_name": null,
2571
+ "trial_params": null
2572
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7935bb1acb760e27d09b3850c4075f1e37f513abea836f5e7d48b0e2c01345e1
3
+ size 4472