Davlan commited on
Commit
a141b9b
1 Parent(s): 18463a4

add MT model

Browse files
all_results.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_bleu": 15.6096,
4
+ "eval_gen_len": 23.7593,
5
+ "eval_loss": 2.3701624870300293,
6
+ "eval_runtime": 311.4621,
7
+ "eval_samples": 1500,
8
+ "eval_samples_per_second": 4.816,
9
+ "eval_steps_per_second": 0.482,
10
+ "predict_bleu": 20.4557,
11
+ "predict_gen_len": 22.912,
12
+ "predict_loss": 2.0914700031280518,
13
+ "predict_runtime": 309.0499,
14
+ "predict_samples": 1500,
15
+ "predict_samples_per_second": 4.854,
16
+ "predict_steps_per_second": 0.485,
17
+ "train_loss": 1.3316941024496511,
18
+ "train_runtime": 32427.393,
19
+ "train_samples": 421931,
20
+ "train_samples_per_second": 39.035,
21
+ "train_steps_per_second": 3.904
22
+ }
config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/m2m100_418M",
3
+ "activation_dropout": 0.0,
4
+ "activation_function": "relu",
5
+ "architectures": [
6
+ "M2M100ForConditionalGeneration"
7
+ ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 0,
10
+ "d_model": 1024,
11
+ "decoder_attention_heads": 16,
12
+ "decoder_ffn_dim": 4096,
13
+ "decoder_layerdrop": 0.05,
14
+ "decoder_layers": 12,
15
+ "decoder_start_token_id": 2,
16
+ "dropout": 0.1,
17
+ "early_stopping": true,
18
+ "encoder_attention_heads": 16,
19
+ "encoder_ffn_dim": 4096,
20
+ "encoder_layerdrop": 0.05,
21
+ "encoder_layers": 12,
22
+ "eos_token_id": 2,
23
+ "forced_bos_token_id": 128022,
24
+ "gradient_checkpointing": false,
25
+ "init_std": 0.02,
26
+ "is_encoder_decoder": true,
27
+ "max_length": 200,
28
+ "max_position_embeddings": 1024,
29
+ "model_type": "m2m_100",
30
+ "num_beams": 5,
31
+ "num_hidden_layers": 12,
32
+ "pad_token_id": 1,
33
+ "scale_embedding": true,
34
+ "torch_dtype": "float32",
35
+ "transformers_version": "4.10.3",
36
+ "use_cache": true,
37
+ "vocab_size": 128112
38
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_bleu": 15.6096,
4
+ "eval_gen_len": 23.7593,
5
+ "eval_loss": 2.3701624870300293,
6
+ "eval_runtime": 311.4621,
7
+ "eval_samples": 1500,
8
+ "eval_samples_per_second": 4.816,
9
+ "eval_steps_per_second": 0.482
10
+ }
generated_predictions.txt ADDED
The diff for this file is too large to render. See raw diff
 
predict_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "predict_bleu": 20.4557,
3
+ "predict_gen_len": 22.912,
4
+ "predict_loss": 2.0914700031280518,
5
+ "predict_runtime": 309.0499,
6
+ "predict_samples": 1500,
7
+ "predict_samples_per_second": 4.854,
8
+ "predict_steps_per_second": 0.485
9
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c4e9d578a02700b32c9faf425b8ac169ce9bcb1aa07a5b7d45ac84c732a641d
3
+ size 1935792071
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d8f7c76ed2a5e0822be39f0a4f95a55eb19c78f4593ce609e2edbc2aea4d380a
3
+ size 2423393
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "additional_special_tokens": ["__af__", "__am__", "__ar__", "__ast__", "__az__", "__ba__", "__be__", "__bg__", "__bn__", "__br__", "__bs__", "__ca__", "__ceb__", "__cs__", "__cy__", "__da__", "__de__", "__el__", "__en__", "__es__", "__et__", "__fa__", "__ff__", "__fi__", "__fr__", "__fy__", "__ga__", "__gd__", "__gl__", "__gu__", "__ha__", "__he__", "__hi__", "__hr__", "__ht__", "__hu__", "__hy__", "__id__", "__ig__", "__ilo__", "__is__", "__it__", "__ja__", "__jv__", "__ka__", "__kk__", "__km__", "__kn__", "__ko__", "__lb__", "__lg__", "__ln__", "__lo__", "__lt__", "__lv__", "__mg__", "__mk__", "__ml__", "__mn__", "__mr__", "__ms__", "__my__", "__ne__", "__nl__", "__no__", "__ns__", "__oc__", "__or__", "__pa__", "__pl__", "__ps__", "__pt__", "__ro__", "__ru__", "__sd__", "__si__", "__sk__", "__sl__", "__so__", "__sq__", "__sr__", "__ss__", "__su__", "__sv__", "__sw__", "__ta__", "__th__", "__tl__", "__tn__", "__tr__", "__uk__", "__ur__", "__uz__", "__vi__", "__wo__", "__xh__", "__yi__", "__yo__", "__zh__", "__zu__"]}
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"src_lang": null, "tgt_lang": null, "bos_token": "<s>", "eos_token": "</s>", "sep_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "sp_model_kwargs": {}, "special_tokens_map_file": "m2m_100_1.2B_v2/special_tokens_map.json", "tokenizer_file": null, "name_or_path": "facebook/m2m100_418M", "model_max_length": 1024, "additional_special_tokens": ["__af__", "__am__", "__ar__", "__ast__", "__az__", "__ba__", "__be__", "__bg__", "__bn__", "__br__", "__bs__", "__ca__", "__ceb__", "__cs__", "__cy__", "__da__", "__de__", "__el__", "__en__", "__es__", "__et__", "__fa__", "__ff__", "__fi__", "__fr__", "__fy__", "__ga__", "__gd__", "__gl__", "__gu__", "__ha__", "__he__", "__hi__", "__hr__", "__ht__", "__hu__", "__hy__", "__id__", "__ig__", "__ilo__", "__is__", "__it__", "__ja__", "__jv__", "__ka__", "__kk__", "__km__", "__kn__", "__ko__", "__lb__", "__lg__", "__ln__", "__lo__", "__lt__", "__lv__", "__mg__", "__mk__", "__ml__", "__mn__", "__mr__", "__ms__", "__my__", "__ne__", "__nl__", "__no__", "__ns__", "__oc__", "__or__", "__pa__", "__pl__", "__ps__", "__pt__", "__ro__", "__ru__", "__sd__", "__si__", "__sk__", "__sl__", "__so__", "__sq__", "__sr__", "__ss__", "__su__", "__sv__", "__sw__", "__ta__", "__th__", "__tl__", "__tn__", "__tr__", "__uk__", "__ur__", "__uz__", "__vi__", "__wo__", "__xh__", "__yi__", "__yo__", "__zh__", "__zu__"], "tokenizer_class": "M2M100Tokenizer"}
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 1.3316941024496511,
4
+ "train_runtime": 32427.393,
5
+ "train_samples": 421931,
6
+ "train_samples_per_second": 39.035,
7
+ "train_steps_per_second": 3.904
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,1543 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "global_step": 126582,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 4.9802499565499043e-05,
13
+ "loss": 2.8425,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.02,
18
+ "learning_rate": 4.960499913099809e-05,
19
+ "loss": 2.4736,
20
+ "step": 1000
21
+ },
22
+ {
23
+ "epoch": 0.04,
24
+ "learning_rate": 4.940749869649713e-05,
25
+ "loss": 2.298,
26
+ "step": 1500
27
+ },
28
+ {
29
+ "epoch": 0.05,
30
+ "learning_rate": 4.9209998261996174e-05,
31
+ "loss": 2.235,
32
+ "step": 2000
33
+ },
34
+ {
35
+ "epoch": 0.06,
36
+ "learning_rate": 4.901249782749522e-05,
37
+ "loss": 2.1605,
38
+ "step": 2500
39
+ },
40
+ {
41
+ "epoch": 0.07,
42
+ "learning_rate": 4.881499739299427e-05,
43
+ "loss": 2.083,
44
+ "step": 3000
45
+ },
46
+ {
47
+ "epoch": 0.08,
48
+ "learning_rate": 4.861749695849331e-05,
49
+ "loss": 2.0588,
50
+ "step": 3500
51
+ },
52
+ {
53
+ "epoch": 0.09,
54
+ "learning_rate": 4.841999652399236e-05,
55
+ "loss": 2.0555,
56
+ "step": 4000
57
+ },
58
+ {
59
+ "epoch": 0.11,
60
+ "learning_rate": 4.82224960894914e-05,
61
+ "loss": 2.0182,
62
+ "step": 4500
63
+ },
64
+ {
65
+ "epoch": 0.12,
66
+ "learning_rate": 4.802499565499045e-05,
67
+ "loss": 1.9641,
68
+ "step": 5000
69
+ },
70
+ {
71
+ "epoch": 0.13,
72
+ "learning_rate": 4.782749522048949e-05,
73
+ "loss": 1.9383,
74
+ "step": 5500
75
+ },
76
+ {
77
+ "epoch": 0.14,
78
+ "learning_rate": 4.762999478598853e-05,
79
+ "loss": 1.9275,
80
+ "step": 6000
81
+ },
82
+ {
83
+ "epoch": 0.15,
84
+ "learning_rate": 4.743249435148758e-05,
85
+ "loss": 1.8923,
86
+ "step": 6500
87
+ },
88
+ {
89
+ "epoch": 0.17,
90
+ "learning_rate": 4.723499391698662e-05,
91
+ "loss": 1.8957,
92
+ "step": 7000
93
+ },
94
+ {
95
+ "epoch": 0.18,
96
+ "learning_rate": 4.703749348248566e-05,
97
+ "loss": 1.8787,
98
+ "step": 7500
99
+ },
100
+ {
101
+ "epoch": 0.19,
102
+ "learning_rate": 4.683999304798471e-05,
103
+ "loss": 1.8516,
104
+ "step": 8000
105
+ },
106
+ {
107
+ "epoch": 0.2,
108
+ "learning_rate": 4.664249261348375e-05,
109
+ "loss": 1.8373,
110
+ "step": 8500
111
+ },
112
+ {
113
+ "epoch": 0.21,
114
+ "learning_rate": 4.6444992178982796e-05,
115
+ "loss": 1.7909,
116
+ "step": 9000
117
+ },
118
+ {
119
+ "epoch": 0.23,
120
+ "learning_rate": 4.6247491744481844e-05,
121
+ "loss": 1.8034,
122
+ "step": 9500
123
+ },
124
+ {
125
+ "epoch": 0.24,
126
+ "learning_rate": 4.6049991309980885e-05,
127
+ "loss": 1.8101,
128
+ "step": 10000
129
+ },
130
+ {
131
+ "epoch": 0.25,
132
+ "learning_rate": 4.585249087547993e-05,
133
+ "loss": 1.7702,
134
+ "step": 10500
135
+ },
136
+ {
137
+ "epoch": 0.26,
138
+ "learning_rate": 4.5654990440978974e-05,
139
+ "loss": 1.7727,
140
+ "step": 11000
141
+ },
142
+ {
143
+ "epoch": 0.27,
144
+ "learning_rate": 4.5457490006478016e-05,
145
+ "loss": 1.7602,
146
+ "step": 11500
147
+ },
148
+ {
149
+ "epoch": 0.28,
150
+ "learning_rate": 4.5259989571977063e-05,
151
+ "loss": 1.761,
152
+ "step": 12000
153
+ },
154
+ {
155
+ "epoch": 0.3,
156
+ "learning_rate": 4.5062489137476105e-05,
157
+ "loss": 1.7219,
158
+ "step": 12500
159
+ },
160
+ {
161
+ "epoch": 0.31,
162
+ "learning_rate": 4.4864988702975146e-05,
163
+ "loss": 1.7497,
164
+ "step": 13000
165
+ },
166
+ {
167
+ "epoch": 0.32,
168
+ "learning_rate": 4.4667488268474193e-05,
169
+ "loss": 1.7244,
170
+ "step": 13500
171
+ },
172
+ {
173
+ "epoch": 0.33,
174
+ "learning_rate": 4.4469987833973235e-05,
175
+ "loss": 1.7144,
176
+ "step": 14000
177
+ },
178
+ {
179
+ "epoch": 0.34,
180
+ "learning_rate": 4.4272487399472276e-05,
181
+ "loss": 1.6983,
182
+ "step": 14500
183
+ },
184
+ {
185
+ "epoch": 0.36,
186
+ "learning_rate": 4.4074986964971324e-05,
187
+ "loss": 1.6779,
188
+ "step": 15000
189
+ },
190
+ {
191
+ "epoch": 0.37,
192
+ "learning_rate": 4.387748653047037e-05,
193
+ "loss": 1.676,
194
+ "step": 15500
195
+ },
196
+ {
197
+ "epoch": 0.38,
198
+ "learning_rate": 4.367998609596941e-05,
199
+ "loss": 1.6737,
200
+ "step": 16000
201
+ },
202
+ {
203
+ "epoch": 0.39,
204
+ "learning_rate": 4.348248566146846e-05,
205
+ "loss": 1.6605,
206
+ "step": 16500
207
+ },
208
+ {
209
+ "epoch": 0.4,
210
+ "learning_rate": 4.32849852269675e-05,
211
+ "loss": 1.6591,
212
+ "step": 17000
213
+ },
214
+ {
215
+ "epoch": 0.41,
216
+ "learning_rate": 4.308748479246655e-05,
217
+ "loss": 1.6598,
218
+ "step": 17500
219
+ },
220
+ {
221
+ "epoch": 0.43,
222
+ "learning_rate": 4.288998435796559e-05,
223
+ "loss": 1.6579,
224
+ "step": 18000
225
+ },
226
+ {
227
+ "epoch": 0.44,
228
+ "learning_rate": 4.269248392346463e-05,
229
+ "loss": 1.6587,
230
+ "step": 18500
231
+ },
232
+ {
233
+ "epoch": 0.45,
234
+ "learning_rate": 4.249498348896368e-05,
235
+ "loss": 1.65,
236
+ "step": 19000
237
+ },
238
+ {
239
+ "epoch": 0.46,
240
+ "learning_rate": 4.229748305446272e-05,
241
+ "loss": 1.6399,
242
+ "step": 19500
243
+ },
244
+ {
245
+ "epoch": 0.47,
246
+ "learning_rate": 4.209998261996176e-05,
247
+ "loss": 1.6409,
248
+ "step": 20000
249
+ },
250
+ {
251
+ "epoch": 0.49,
252
+ "learning_rate": 4.190248218546081e-05,
253
+ "loss": 1.6167,
254
+ "step": 20500
255
+ },
256
+ {
257
+ "epoch": 0.5,
258
+ "learning_rate": 4.170498175095985e-05,
259
+ "loss": 1.6468,
260
+ "step": 21000
261
+ },
262
+ {
263
+ "epoch": 0.51,
264
+ "learning_rate": 4.15074813164589e-05,
265
+ "loss": 1.6084,
266
+ "step": 21500
267
+ },
268
+ {
269
+ "epoch": 0.52,
270
+ "learning_rate": 4.1309980881957946e-05,
271
+ "loss": 1.6214,
272
+ "step": 22000
273
+ },
274
+ {
275
+ "epoch": 0.53,
276
+ "learning_rate": 4.111248044745699e-05,
277
+ "loss": 1.6254,
278
+ "step": 22500
279
+ },
280
+ {
281
+ "epoch": 0.55,
282
+ "learning_rate": 4.091498001295603e-05,
283
+ "loss": 1.5892,
284
+ "step": 23000
285
+ },
286
+ {
287
+ "epoch": 0.56,
288
+ "learning_rate": 4.0717479578455077e-05,
289
+ "loss": 1.5746,
290
+ "step": 23500
291
+ },
292
+ {
293
+ "epoch": 0.57,
294
+ "learning_rate": 4.051997914395412e-05,
295
+ "loss": 1.5701,
296
+ "step": 24000
297
+ },
298
+ {
299
+ "epoch": 0.58,
300
+ "learning_rate": 4.0322478709453166e-05,
301
+ "loss": 1.6088,
302
+ "step": 24500
303
+ },
304
+ {
305
+ "epoch": 0.59,
306
+ "learning_rate": 4.012497827495221e-05,
307
+ "loss": 1.5952,
308
+ "step": 25000
309
+ },
310
+ {
311
+ "epoch": 0.6,
312
+ "learning_rate": 3.992747784045125e-05,
313
+ "loss": 1.5561,
314
+ "step": 25500
315
+ },
316
+ {
317
+ "epoch": 0.62,
318
+ "learning_rate": 3.9729977405950296e-05,
319
+ "loss": 1.5707,
320
+ "step": 26000
321
+ },
322
+ {
323
+ "epoch": 0.63,
324
+ "learning_rate": 3.953247697144934e-05,
325
+ "loss": 1.5582,
326
+ "step": 26500
327
+ },
328
+ {
329
+ "epoch": 0.64,
330
+ "learning_rate": 3.9334976536948385e-05,
331
+ "loss": 1.5612,
332
+ "step": 27000
333
+ },
334
+ {
335
+ "epoch": 0.65,
336
+ "learning_rate": 3.9137476102447426e-05,
337
+ "loss": 1.5552,
338
+ "step": 27500
339
+ },
340
+ {
341
+ "epoch": 0.66,
342
+ "learning_rate": 3.8939975667946474e-05,
343
+ "loss": 1.5632,
344
+ "step": 28000
345
+ },
346
+ {
347
+ "epoch": 0.68,
348
+ "learning_rate": 3.8742475233445515e-05,
349
+ "loss": 1.538,
350
+ "step": 28500
351
+ },
352
+ {
353
+ "epoch": 0.69,
354
+ "learning_rate": 3.854497479894456e-05,
355
+ "loss": 1.5257,
356
+ "step": 29000
357
+ },
358
+ {
359
+ "epoch": 0.7,
360
+ "learning_rate": 3.8347474364443604e-05,
361
+ "loss": 1.5355,
362
+ "step": 29500
363
+ },
364
+ {
365
+ "epoch": 0.71,
366
+ "learning_rate": 3.8149973929942645e-05,
367
+ "loss": 1.5233,
368
+ "step": 30000
369
+ },
370
+ {
371
+ "epoch": 0.72,
372
+ "learning_rate": 3.795247349544169e-05,
373
+ "loss": 1.5154,
374
+ "step": 30500
375
+ },
376
+ {
377
+ "epoch": 0.73,
378
+ "learning_rate": 3.7754973060940734e-05,
379
+ "loss": 1.5493,
380
+ "step": 31000
381
+ },
382
+ {
383
+ "epoch": 0.75,
384
+ "learning_rate": 3.755747262643978e-05,
385
+ "loss": 1.5327,
386
+ "step": 31500
387
+ },
388
+ {
389
+ "epoch": 0.76,
390
+ "learning_rate": 3.735997219193882e-05,
391
+ "loss": 1.5255,
392
+ "step": 32000
393
+ },
394
+ {
395
+ "epoch": 0.77,
396
+ "learning_rate": 3.7162471757437864e-05,
397
+ "loss": 1.51,
398
+ "step": 32500
399
+ },
400
+ {
401
+ "epoch": 0.78,
402
+ "learning_rate": 3.696497132293691e-05,
403
+ "loss": 1.5122,
404
+ "step": 33000
405
+ },
406
+ {
407
+ "epoch": 0.79,
408
+ "learning_rate": 3.676747088843596e-05,
409
+ "loss": 1.5006,
410
+ "step": 33500
411
+ },
412
+ {
413
+ "epoch": 0.81,
414
+ "learning_rate": 3.6569970453935e-05,
415
+ "loss": 1.5106,
416
+ "step": 34000
417
+ },
418
+ {
419
+ "epoch": 0.82,
420
+ "learning_rate": 3.637247001943405e-05,
421
+ "loss": 1.5031,
422
+ "step": 34500
423
+ },
424
+ {
425
+ "epoch": 0.83,
426
+ "learning_rate": 3.617496958493309e-05,
427
+ "loss": 1.5024,
428
+ "step": 35000
429
+ },
430
+ {
431
+ "epoch": 0.84,
432
+ "learning_rate": 3.597746915043213e-05,
433
+ "loss": 1.5231,
434
+ "step": 35500
435
+ },
436
+ {
437
+ "epoch": 0.85,
438
+ "learning_rate": 3.577996871593118e-05,
439
+ "loss": 1.5271,
440
+ "step": 36000
441
+ },
442
+ {
443
+ "epoch": 0.87,
444
+ "learning_rate": 3.558246828143022e-05,
445
+ "loss": 1.508,
446
+ "step": 36500
447
+ },
448
+ {
449
+ "epoch": 0.88,
450
+ "learning_rate": 3.538496784692926e-05,
451
+ "loss": 1.4859,
452
+ "step": 37000
453
+ },
454
+ {
455
+ "epoch": 0.89,
456
+ "learning_rate": 3.518746741242831e-05,
457
+ "loss": 1.4864,
458
+ "step": 37500
459
+ },
460
+ {
461
+ "epoch": 0.9,
462
+ "learning_rate": 3.498996697792735e-05,
463
+ "loss": 1.5045,
464
+ "step": 38000
465
+ },
466
+ {
467
+ "epoch": 0.91,
468
+ "learning_rate": 3.47924665434264e-05,
469
+ "loss": 1.4773,
470
+ "step": 38500
471
+ },
472
+ {
473
+ "epoch": 0.92,
474
+ "learning_rate": 3.459496610892544e-05,
475
+ "loss": 1.4704,
476
+ "step": 39000
477
+ },
478
+ {
479
+ "epoch": 0.94,
480
+ "learning_rate": 3.439746567442449e-05,
481
+ "loss": 1.4816,
482
+ "step": 39500
483
+ },
484
+ {
485
+ "epoch": 0.95,
486
+ "learning_rate": 3.4199965239923535e-05,
487
+ "loss": 1.4899,
488
+ "step": 40000
489
+ },
490
+ {
491
+ "epoch": 0.96,
492
+ "learning_rate": 3.4002464805422576e-05,
493
+ "loss": 1.463,
494
+ "step": 40500
495
+ },
496
+ {
497
+ "epoch": 0.97,
498
+ "learning_rate": 3.380496437092162e-05,
499
+ "loss": 1.4847,
500
+ "step": 41000
501
+ },
502
+ {
503
+ "epoch": 0.98,
504
+ "learning_rate": 3.3607463936420665e-05,
505
+ "loss": 1.478,
506
+ "step": 41500
507
+ },
508
+ {
509
+ "epoch": 1.0,
510
+ "learning_rate": 3.3409963501919706e-05,
511
+ "loss": 1.4617,
512
+ "step": 42000
513
+ },
514
+ {
515
+ "epoch": 1.01,
516
+ "learning_rate": 3.321246306741875e-05,
517
+ "loss": 1.3742,
518
+ "step": 42500
519
+ },
520
+ {
521
+ "epoch": 1.02,
522
+ "learning_rate": 3.3014962632917795e-05,
523
+ "loss": 1.3036,
524
+ "step": 43000
525
+ },
526
+ {
527
+ "epoch": 1.03,
528
+ "learning_rate": 3.2817462198416836e-05,
529
+ "loss": 1.3172,
530
+ "step": 43500
531
+ },
532
+ {
533
+ "epoch": 1.04,
534
+ "learning_rate": 3.261996176391588e-05,
535
+ "loss": 1.2944,
536
+ "step": 44000
537
+ },
538
+ {
539
+ "epoch": 1.05,
540
+ "learning_rate": 3.2422461329414925e-05,
541
+ "loss": 1.319,
542
+ "step": 44500
543
+ },
544
+ {
545
+ "epoch": 1.07,
546
+ "learning_rate": 3.2224960894913966e-05,
547
+ "loss": 1.3015,
548
+ "step": 45000
549
+ },
550
+ {
551
+ "epoch": 1.08,
552
+ "learning_rate": 3.2027460460413014e-05,
553
+ "loss": 1.3011,
554
+ "step": 45500
555
+ },
556
+ {
557
+ "epoch": 1.09,
558
+ "learning_rate": 3.182996002591206e-05,
559
+ "loss": 1.295,
560
+ "step": 46000
561
+ },
562
+ {
563
+ "epoch": 1.1,
564
+ "learning_rate": 3.16324595914111e-05,
565
+ "loss": 1.2994,
566
+ "step": 46500
567
+ },
568
+ {
569
+ "epoch": 1.11,
570
+ "learning_rate": 3.143495915691015e-05,
571
+ "loss": 1.3215,
572
+ "step": 47000
573
+ },
574
+ {
575
+ "epoch": 1.13,
576
+ "learning_rate": 3.123745872240919e-05,
577
+ "loss": 1.3197,
578
+ "step": 47500
579
+ },
580
+ {
581
+ "epoch": 1.14,
582
+ "learning_rate": 3.103995828790823e-05,
583
+ "loss": 1.3114,
584
+ "step": 48000
585
+ },
586
+ {
587
+ "epoch": 1.15,
588
+ "learning_rate": 3.084245785340728e-05,
589
+ "loss": 1.2988,
590
+ "step": 48500
591
+ },
592
+ {
593
+ "epoch": 1.16,
594
+ "learning_rate": 3.064495741890632e-05,
595
+ "loss": 1.3004,
596
+ "step": 49000
597
+ },
598
+ {
599
+ "epoch": 1.17,
600
+ "learning_rate": 3.0447456984405366e-05,
601
+ "loss": 1.275,
602
+ "step": 49500
603
+ },
604
+ {
605
+ "epoch": 1.19,
606
+ "learning_rate": 3.024995654990441e-05,
607
+ "loss": 1.2822,
608
+ "step": 50000
609
+ },
610
+ {
611
+ "epoch": 1.2,
612
+ "learning_rate": 3.0052456115403455e-05,
613
+ "loss": 1.3113,
614
+ "step": 50500
615
+ },
616
+ {
617
+ "epoch": 1.21,
618
+ "learning_rate": 2.9854955680902503e-05,
619
+ "loss": 1.3087,
620
+ "step": 51000
621
+ },
622
+ {
623
+ "epoch": 1.22,
624
+ "learning_rate": 2.9657455246401544e-05,
625
+ "loss": 1.3039,
626
+ "step": 51500
627
+ },
628
+ {
629
+ "epoch": 1.23,
630
+ "learning_rate": 2.9459954811900585e-05,
631
+ "loss": 1.2772,
632
+ "step": 52000
633
+ },
634
+ {
635
+ "epoch": 1.24,
636
+ "learning_rate": 2.9262454377399633e-05,
637
+ "loss": 1.2918,
638
+ "step": 52500
639
+ },
640
+ {
641
+ "epoch": 1.26,
642
+ "learning_rate": 2.9064953942898678e-05,
643
+ "loss": 1.2815,
644
+ "step": 53000
645
+ },
646
+ {
647
+ "epoch": 1.27,
648
+ "learning_rate": 2.886745350839772e-05,
649
+ "loss": 1.3004,
650
+ "step": 53500
651
+ },
652
+ {
653
+ "epoch": 1.28,
654
+ "learning_rate": 2.8669953073896767e-05,
655
+ "loss": 1.2761,
656
+ "step": 54000
657
+ },
658
+ {
659
+ "epoch": 1.29,
660
+ "learning_rate": 2.8472452639395808e-05,
661
+ "loss": 1.2927,
662
+ "step": 54500
663
+ },
664
+ {
665
+ "epoch": 1.3,
666
+ "learning_rate": 2.827495220489485e-05,
667
+ "loss": 1.2815,
668
+ "step": 55000
669
+ },
670
+ {
671
+ "epoch": 1.32,
672
+ "learning_rate": 2.8077451770393897e-05,
673
+ "loss": 1.27,
674
+ "step": 55500
675
+ },
676
+ {
677
+ "epoch": 1.33,
678
+ "learning_rate": 2.787995133589294e-05,
679
+ "loss": 1.2928,
680
+ "step": 56000
681
+ },
682
+ {
683
+ "epoch": 1.34,
684
+ "learning_rate": 2.7682450901391982e-05,
685
+ "loss": 1.2752,
686
+ "step": 56500
687
+ },
688
+ {
689
+ "epoch": 1.35,
690
+ "learning_rate": 2.748495046689103e-05,
691
+ "loss": 1.2935,
692
+ "step": 57000
693
+ },
694
+ {
695
+ "epoch": 1.36,
696
+ "learning_rate": 2.728745003239007e-05,
697
+ "loss": 1.2794,
698
+ "step": 57500
699
+ },
700
+ {
701
+ "epoch": 1.37,
702
+ "learning_rate": 2.708994959788912e-05,
703
+ "loss": 1.2956,
704
+ "step": 58000
705
+ },
706
+ {
707
+ "epoch": 1.39,
708
+ "learning_rate": 2.689244916338816e-05,
709
+ "loss": 1.2856,
710
+ "step": 58500
711
+ },
712
+ {
713
+ "epoch": 1.4,
714
+ "learning_rate": 2.6694948728887205e-05,
715
+ "loss": 1.2743,
716
+ "step": 59000
717
+ },
718
+ {
719
+ "epoch": 1.41,
720
+ "learning_rate": 2.6497448294386253e-05,
721
+ "loss": 1.2743,
722
+ "step": 59500
723
+ },
724
+ {
725
+ "epoch": 1.42,
726
+ "learning_rate": 2.6299947859885294e-05,
727
+ "loss": 1.2753,
728
+ "step": 60000
729
+ },
730
+ {
731
+ "epoch": 1.43,
732
+ "learning_rate": 2.6102447425384335e-05,
733
+ "loss": 1.2653,
734
+ "step": 60500
735
+ },
736
+ {
737
+ "epoch": 1.45,
738
+ "learning_rate": 2.5904946990883383e-05,
739
+ "loss": 1.2778,
740
+ "step": 61000
741
+ },
742
+ {
743
+ "epoch": 1.46,
744
+ "learning_rate": 2.5707446556382424e-05,
745
+ "loss": 1.2726,
746
+ "step": 61500
747
+ },
748
+ {
749
+ "epoch": 1.47,
750
+ "learning_rate": 2.550994612188147e-05,
751
+ "loss": 1.2773,
752
+ "step": 62000
753
+ },
754
+ {
755
+ "epoch": 1.48,
756
+ "learning_rate": 2.5312445687380516e-05,
757
+ "loss": 1.2632,
758
+ "step": 62500
759
+ },
760
+ {
761
+ "epoch": 1.49,
762
+ "learning_rate": 2.5114945252879557e-05,
763
+ "loss": 1.2751,
764
+ "step": 63000
765
+ },
766
+ {
767
+ "epoch": 1.5,
768
+ "learning_rate": 2.4917444818378602e-05,
769
+ "loss": 1.2607,
770
+ "step": 63500
771
+ },
772
+ {
773
+ "epoch": 1.52,
774
+ "learning_rate": 2.4719944383877646e-05,
775
+ "loss": 1.2556,
776
+ "step": 64000
777
+ },
778
+ {
779
+ "epoch": 1.53,
780
+ "learning_rate": 2.4522443949376688e-05,
781
+ "loss": 1.235,
782
+ "step": 64500
783
+ },
784
+ {
785
+ "epoch": 1.54,
786
+ "learning_rate": 2.4324943514875735e-05,
787
+ "loss": 1.2502,
788
+ "step": 65000
789
+ },
790
+ {
791
+ "epoch": 1.55,
792
+ "learning_rate": 2.412744308037478e-05,
793
+ "loss": 1.2633,
794
+ "step": 65500
795
+ },
796
+ {
797
+ "epoch": 1.56,
798
+ "learning_rate": 2.392994264587382e-05,
799
+ "loss": 1.2558,
800
+ "step": 66000
801
+ },
802
+ {
803
+ "epoch": 1.58,
804
+ "learning_rate": 2.3732442211372865e-05,
805
+ "loss": 1.2659,
806
+ "step": 66500
807
+ },
808
+ {
809
+ "epoch": 1.59,
810
+ "learning_rate": 2.353494177687191e-05,
811
+ "loss": 1.258,
812
+ "step": 67000
813
+ },
814
+ {
815
+ "epoch": 1.6,
816
+ "learning_rate": 2.3337441342370954e-05,
817
+ "loss": 1.2609,
818
+ "step": 67500
819
+ },
820
+ {
821
+ "epoch": 1.61,
822
+ "learning_rate": 2.313994090787e-05,
823
+ "loss": 1.2485,
824
+ "step": 68000
825
+ },
826
+ {
827
+ "epoch": 1.62,
828
+ "learning_rate": 2.2942440473369043e-05,
829
+ "loss": 1.2563,
830
+ "step": 68500
831
+ },
832
+ {
833
+ "epoch": 1.64,
834
+ "learning_rate": 2.2744940038868088e-05,
835
+ "loss": 1.2621,
836
+ "step": 69000
837
+ },
838
+ {
839
+ "epoch": 1.65,
840
+ "learning_rate": 2.254743960436713e-05,
841
+ "loss": 1.246,
842
+ "step": 69500
843
+ },
844
+ {
845
+ "epoch": 1.66,
846
+ "learning_rate": 2.2349939169866174e-05,
847
+ "loss": 1.2327,
848
+ "step": 70000
849
+ },
850
+ {
851
+ "epoch": 1.67,
852
+ "learning_rate": 2.2152438735365218e-05,
853
+ "loss": 1.2388,
854
+ "step": 70500
855
+ },
856
+ {
857
+ "epoch": 1.68,
858
+ "learning_rate": 2.1954938300864263e-05,
859
+ "loss": 1.2443,
860
+ "step": 71000
861
+ },
862
+ {
863
+ "epoch": 1.69,
864
+ "learning_rate": 2.1757437866363307e-05,
865
+ "loss": 1.2585,
866
+ "step": 71500
867
+ },
868
+ {
869
+ "epoch": 1.71,
870
+ "learning_rate": 2.155993743186235e-05,
871
+ "loss": 1.2376,
872
+ "step": 72000
873
+ },
874
+ {
875
+ "epoch": 1.72,
876
+ "learning_rate": 2.1362436997361396e-05,
877
+ "loss": 1.2317,
878
+ "step": 72500
879
+ },
880
+ {
881
+ "epoch": 1.73,
882
+ "learning_rate": 2.116493656286044e-05,
883
+ "loss": 1.2279,
884
+ "step": 73000
885
+ },
886
+ {
887
+ "epoch": 1.74,
888
+ "learning_rate": 2.096743612835948e-05,
889
+ "loss": 1.2444,
890
+ "step": 73500
891
+ },
892
+ {
893
+ "epoch": 1.75,
894
+ "learning_rate": 2.0769935693858526e-05,
895
+ "loss": 1.2488,
896
+ "step": 74000
897
+ },
898
+ {
899
+ "epoch": 1.77,
900
+ "learning_rate": 2.0572435259357574e-05,
901
+ "loss": 1.2314,
902
+ "step": 74500
903
+ },
904
+ {
905
+ "epoch": 1.78,
906
+ "learning_rate": 2.0374934824856615e-05,
907
+ "loss": 1.2385,
908
+ "step": 75000
909
+ },
910
+ {
911
+ "epoch": 1.79,
912
+ "learning_rate": 2.017743439035566e-05,
913
+ "loss": 1.2219,
914
+ "step": 75500
915
+ },
916
+ {
917
+ "epoch": 1.8,
918
+ "learning_rate": 1.9979933955854704e-05,
919
+ "loss": 1.2491,
920
+ "step": 76000
921
+ },
922
+ {
923
+ "epoch": 1.81,
924
+ "learning_rate": 1.978243352135375e-05,
925
+ "loss": 1.229,
926
+ "step": 76500
927
+ },
928
+ {
929
+ "epoch": 1.82,
930
+ "learning_rate": 1.9584933086852793e-05,
931
+ "loss": 1.2233,
932
+ "step": 77000
933
+ },
934
+ {
935
+ "epoch": 1.84,
936
+ "learning_rate": 1.9387432652351838e-05,
937
+ "loss": 1.2521,
938
+ "step": 77500
939
+ },
940
+ {
941
+ "epoch": 1.85,
942
+ "learning_rate": 1.9189932217850882e-05,
943
+ "loss": 1.2214,
944
+ "step": 78000
945
+ },
946
+ {
947
+ "epoch": 1.86,
948
+ "learning_rate": 1.8992431783349923e-05,
949
+ "loss": 1.2162,
950
+ "step": 78500
951
+ },
952
+ {
953
+ "epoch": 1.87,
954
+ "learning_rate": 1.8794931348848968e-05,
955
+ "loss": 1.2178,
956
+ "step": 79000
957
+ },
958
+ {
959
+ "epoch": 1.88,
960
+ "learning_rate": 1.8597430914348012e-05,
961
+ "loss": 1.2418,
962
+ "step": 79500
963
+ },
964
+ {
965
+ "epoch": 1.9,
966
+ "learning_rate": 1.8399930479847057e-05,
967
+ "loss": 1.2453,
968
+ "step": 80000
969
+ },
970
+ {
971
+ "epoch": 1.91,
972
+ "learning_rate": 1.82024300453461e-05,
973
+ "loss": 1.228,
974
+ "step": 80500
975
+ },
976
+ {
977
+ "epoch": 1.92,
978
+ "learning_rate": 1.8004929610845146e-05,
979
+ "loss": 1.2036,
980
+ "step": 81000
981
+ },
982
+ {
983
+ "epoch": 1.93,
984
+ "learning_rate": 1.780742917634419e-05,
985
+ "loss": 1.2078,
986
+ "step": 81500
987
+ },
988
+ {
989
+ "epoch": 1.94,
990
+ "learning_rate": 1.760992874184323e-05,
991
+ "loss": 1.2106,
992
+ "step": 82000
993
+ },
994
+ {
995
+ "epoch": 1.96,
996
+ "learning_rate": 1.7412428307342276e-05,
997
+ "loss": 1.2187,
998
+ "step": 82500
999
+ },
1000
+ {
1001
+ "epoch": 1.97,
1002
+ "learning_rate": 1.721492787284132e-05,
1003
+ "loss": 1.2267,
1004
+ "step": 83000
1005
+ },
1006
+ {
1007
+ "epoch": 1.98,
1008
+ "learning_rate": 1.7017427438340368e-05,
1009
+ "loss": 1.2262,
1010
+ "step": 83500
1011
+ },
1012
+ {
1013
+ "epoch": 1.99,
1014
+ "learning_rate": 1.681992700383941e-05,
1015
+ "loss": 1.2185,
1016
+ "step": 84000
1017
+ },
1018
+ {
1019
+ "epoch": 2.0,
1020
+ "learning_rate": 1.6622426569338454e-05,
1021
+ "loss": 1.1781,
1022
+ "step": 84500
1023
+ },
1024
+ {
1025
+ "epoch": 2.01,
1026
+ "learning_rate": 1.6424926134837498e-05,
1027
+ "loss": 1.0483,
1028
+ "step": 85000
1029
+ },
1030
+ {
1031
+ "epoch": 2.03,
1032
+ "learning_rate": 1.622742570033654e-05,
1033
+ "loss": 1.0659,
1034
+ "step": 85500
1035
+ },
1036
+ {
1037
+ "epoch": 2.04,
1038
+ "learning_rate": 1.6029925265835584e-05,
1039
+ "loss": 1.0643,
1040
+ "step": 86000
1041
+ },
1042
+ {
1043
+ "epoch": 2.05,
1044
+ "learning_rate": 1.583242483133463e-05,
1045
+ "loss": 1.0453,
1046
+ "step": 86500
1047
+ },
1048
+ {
1049
+ "epoch": 2.06,
1050
+ "learning_rate": 1.5634924396833676e-05,
1051
+ "loss": 1.0541,
1052
+ "step": 87000
1053
+ },
1054
+ {
1055
+ "epoch": 2.07,
1056
+ "learning_rate": 1.5437423962332717e-05,
1057
+ "loss": 1.0534,
1058
+ "step": 87500
1059
+ },
1060
+ {
1061
+ "epoch": 2.09,
1062
+ "learning_rate": 1.5239923527831762e-05,
1063
+ "loss": 1.0641,
1064
+ "step": 88000
1065
+ },
1066
+ {
1067
+ "epoch": 2.1,
1068
+ "learning_rate": 1.5042423093330806e-05,
1069
+ "loss": 1.0771,
1070
+ "step": 88500
1071
+ },
1072
+ {
1073
+ "epoch": 2.11,
1074
+ "learning_rate": 1.4844922658829849e-05,
1075
+ "loss": 1.0638,
1076
+ "step": 89000
1077
+ },
1078
+ {
1079
+ "epoch": 2.12,
1080
+ "learning_rate": 1.4647422224328893e-05,
1081
+ "loss": 1.0677,
1082
+ "step": 89500
1083
+ },
1084
+ {
1085
+ "epoch": 2.13,
1086
+ "learning_rate": 1.4449921789827938e-05,
1087
+ "loss": 1.0776,
1088
+ "step": 90000
1089
+ },
1090
+ {
1091
+ "epoch": 2.14,
1092
+ "learning_rate": 1.4252421355326984e-05,
1093
+ "loss": 1.064,
1094
+ "step": 90500
1095
+ },
1096
+ {
1097
+ "epoch": 2.16,
1098
+ "learning_rate": 1.4054920920826025e-05,
1099
+ "loss": 1.0548,
1100
+ "step": 91000
1101
+ },
1102
+ {
1103
+ "epoch": 2.17,
1104
+ "learning_rate": 1.3857420486325071e-05,
1105
+ "loss": 1.0606,
1106
+ "step": 91500
1107
+ },
1108
+ {
1109
+ "epoch": 2.18,
1110
+ "learning_rate": 1.3659920051824116e-05,
1111
+ "loss": 1.0699,
1112
+ "step": 92000
1113
+ },
1114
+ {
1115
+ "epoch": 2.19,
1116
+ "learning_rate": 1.3462419617323157e-05,
1117
+ "loss": 1.0721,
1118
+ "step": 92500
1119
+ },
1120
+ {
1121
+ "epoch": 2.2,
1122
+ "learning_rate": 1.3264919182822203e-05,
1123
+ "loss": 1.0438,
1124
+ "step": 93000
1125
+ },
1126
+ {
1127
+ "epoch": 2.22,
1128
+ "learning_rate": 1.3067418748321248e-05,
1129
+ "loss": 1.0425,
1130
+ "step": 93500
1131
+ },
1132
+ {
1133
+ "epoch": 2.23,
1134
+ "learning_rate": 1.2869918313820292e-05,
1135
+ "loss": 1.0501,
1136
+ "step": 94000
1137
+ },
1138
+ {
1139
+ "epoch": 2.24,
1140
+ "learning_rate": 1.2672417879319335e-05,
1141
+ "loss": 1.0307,
1142
+ "step": 94500
1143
+ },
1144
+ {
1145
+ "epoch": 2.25,
1146
+ "learning_rate": 1.247491744481838e-05,
1147
+ "loss": 1.037,
1148
+ "step": 95000
1149
+ },
1150
+ {
1151
+ "epoch": 2.26,
1152
+ "learning_rate": 1.2277417010317422e-05,
1153
+ "loss": 1.0401,
1154
+ "step": 95500
1155
+ },
1156
+ {
1157
+ "epoch": 2.28,
1158
+ "learning_rate": 1.2079916575816467e-05,
1159
+ "loss": 1.0504,
1160
+ "step": 96000
1161
+ },
1162
+ {
1163
+ "epoch": 2.29,
1164
+ "learning_rate": 1.1882416141315511e-05,
1165
+ "loss": 1.0381,
1166
+ "step": 96500
1167
+ },
1168
+ {
1169
+ "epoch": 2.3,
1170
+ "learning_rate": 1.1684915706814556e-05,
1171
+ "loss": 1.0302,
1172
+ "step": 97000
1173
+ },
1174
+ {
1175
+ "epoch": 2.31,
1176
+ "learning_rate": 1.14874152723136e-05,
1177
+ "loss": 1.0474,
1178
+ "step": 97500
1179
+ },
1180
+ {
1181
+ "epoch": 2.32,
1182
+ "learning_rate": 1.1289914837812645e-05,
1183
+ "loss": 1.0454,
1184
+ "step": 98000
1185
+ },
1186
+ {
1187
+ "epoch": 2.33,
1188
+ "learning_rate": 1.1092414403311688e-05,
1189
+ "loss": 1.0531,
1190
+ "step": 98500
1191
+ },
1192
+ {
1193
+ "epoch": 2.35,
1194
+ "learning_rate": 1.0894913968810732e-05,
1195
+ "loss": 1.0448,
1196
+ "step": 99000
1197
+ },
1198
+ {
1199
+ "epoch": 2.36,
1200
+ "learning_rate": 1.0697413534309776e-05,
1201
+ "loss": 1.0562,
1202
+ "step": 99500
1203
+ },
1204
+ {
1205
+ "epoch": 2.37,
1206
+ "learning_rate": 1.049991309980882e-05,
1207
+ "loss": 1.0432,
1208
+ "step": 100000
1209
+ },
1210
+ {
1211
+ "epoch": 2.38,
1212
+ "learning_rate": 1.0302412665307864e-05,
1213
+ "loss": 1.0436,
1214
+ "step": 100500
1215
+ },
1216
+ {
1217
+ "epoch": 2.39,
1218
+ "learning_rate": 1.0104912230806908e-05,
1219
+ "loss": 1.0419,
1220
+ "step": 101000
1221
+ },
1222
+ {
1223
+ "epoch": 2.41,
1224
+ "learning_rate": 9.907411796305953e-06,
1225
+ "loss": 1.0457,
1226
+ "step": 101500
1227
+ },
1228
+ {
1229
+ "epoch": 2.42,
1230
+ "learning_rate": 9.709911361804996e-06,
1231
+ "loss": 1.0409,
1232
+ "step": 102000
1233
+ },
1234
+ {
1235
+ "epoch": 2.43,
1236
+ "learning_rate": 9.51241092730404e-06,
1237
+ "loss": 1.031,
1238
+ "step": 102500
1239
+ },
1240
+ {
1241
+ "epoch": 2.44,
1242
+ "learning_rate": 9.314910492803085e-06,
1243
+ "loss": 1.0356,
1244
+ "step": 103000
1245
+ },
1246
+ {
1247
+ "epoch": 2.45,
1248
+ "learning_rate": 9.117410058302129e-06,
1249
+ "loss": 1.0606,
1250
+ "step": 103500
1251
+ },
1252
+ {
1253
+ "epoch": 2.46,
1254
+ "learning_rate": 8.919909623801174e-06,
1255
+ "loss": 1.0432,
1256
+ "step": 104000
1257
+ },
1258
+ {
1259
+ "epoch": 2.48,
1260
+ "learning_rate": 8.722409189300216e-06,
1261
+ "loss": 1.0254,
1262
+ "step": 104500
1263
+ },
1264
+ {
1265
+ "epoch": 2.49,
1266
+ "learning_rate": 8.52490875479926e-06,
1267
+ "loss": 1.0482,
1268
+ "step": 105000
1269
+ },
1270
+ {
1271
+ "epoch": 2.5,
1272
+ "learning_rate": 8.327408320298305e-06,
1273
+ "loss": 1.0522,
1274
+ "step": 105500
1275
+ },
1276
+ {
1277
+ "epoch": 2.51,
1278
+ "learning_rate": 8.129907885797348e-06,
1279
+ "loss": 1.0263,
1280
+ "step": 106000
1281
+ },
1282
+ {
1283
+ "epoch": 2.52,
1284
+ "learning_rate": 7.932407451296393e-06,
1285
+ "loss": 1.0546,
1286
+ "step": 106500
1287
+ },
1288
+ {
1289
+ "epoch": 2.54,
1290
+ "learning_rate": 7.734907016795437e-06,
1291
+ "loss": 1.0475,
1292
+ "step": 107000
1293
+ },
1294
+ {
1295
+ "epoch": 2.55,
1296
+ "learning_rate": 7.5374065822944816e-06,
1297
+ "loss": 1.0293,
1298
+ "step": 107500
1299
+ },
1300
+ {
1301
+ "epoch": 2.56,
1302
+ "learning_rate": 7.339906147793525e-06,
1303
+ "loss": 1.0497,
1304
+ "step": 108000
1305
+ },
1306
+ {
1307
+ "epoch": 2.57,
1308
+ "learning_rate": 7.14240571329257e-06,
1309
+ "loss": 1.0448,
1310
+ "step": 108500
1311
+ },
1312
+ {
1313
+ "epoch": 2.58,
1314
+ "learning_rate": 6.944905278791613e-06,
1315
+ "loss": 1.0275,
1316
+ "step": 109000
1317
+ },
1318
+ {
1319
+ "epoch": 2.6,
1320
+ "learning_rate": 6.747404844290659e-06,
1321
+ "loss": 1.0281,
1322
+ "step": 109500
1323
+ },
1324
+ {
1325
+ "epoch": 2.61,
1326
+ "learning_rate": 6.549904409789702e-06,
1327
+ "loss": 1.0295,
1328
+ "step": 110000
1329
+ },
1330
+ {
1331
+ "epoch": 2.62,
1332
+ "learning_rate": 6.352403975288745e-06,
1333
+ "loss": 1.0143,
1334
+ "step": 110500
1335
+ },
1336
+ {
1337
+ "epoch": 2.63,
1338
+ "learning_rate": 6.1549035407877905e-06,
1339
+ "loss": 1.0277,
1340
+ "step": 111000
1341
+ },
1342
+ {
1343
+ "epoch": 2.64,
1344
+ "learning_rate": 5.957403106286834e-06,
1345
+ "loss": 1.0266,
1346
+ "step": 111500
1347
+ },
1348
+ {
1349
+ "epoch": 2.65,
1350
+ "learning_rate": 5.759902671785878e-06,
1351
+ "loss": 1.0502,
1352
+ "step": 112000
1353
+ },
1354
+ {
1355
+ "epoch": 2.67,
1356
+ "learning_rate": 5.562402237284922e-06,
1357
+ "loss": 1.0312,
1358
+ "step": 112500
1359
+ },
1360
+ {
1361
+ "epoch": 2.68,
1362
+ "learning_rate": 5.364901802783967e-06,
1363
+ "loss": 1.0444,
1364
+ "step": 113000
1365
+ },
1366
+ {
1367
+ "epoch": 2.69,
1368
+ "learning_rate": 5.16740136828301e-06,
1369
+ "loss": 1.0257,
1370
+ "step": 113500
1371
+ },
1372
+ {
1373
+ "epoch": 2.7,
1374
+ "learning_rate": 4.969900933782055e-06,
1375
+ "loss": 1.0199,
1376
+ "step": 114000
1377
+ },
1378
+ {
1379
+ "epoch": 2.71,
1380
+ "learning_rate": 4.7724004992810985e-06,
1381
+ "loss": 1.0178,
1382
+ "step": 114500
1383
+ },
1384
+ {
1385
+ "epoch": 2.73,
1386
+ "learning_rate": 4.574900064780143e-06,
1387
+ "loss": 1.0197,
1388
+ "step": 115000
1389
+ },
1390
+ {
1391
+ "epoch": 2.74,
1392
+ "learning_rate": 4.377399630279187e-06,
1393
+ "loss": 1.0181,
1394
+ "step": 115500
1395
+ },
1396
+ {
1397
+ "epoch": 2.75,
1398
+ "learning_rate": 4.179899195778231e-06,
1399
+ "loss": 1.0492,
1400
+ "step": 116000
1401
+ },
1402
+ {
1403
+ "epoch": 2.76,
1404
+ "learning_rate": 3.982398761277275e-06,
1405
+ "loss": 1.035,
1406
+ "step": 116500
1407
+ },
1408
+ {
1409
+ "epoch": 2.77,
1410
+ "learning_rate": 3.7848983267763193e-06,
1411
+ "loss": 1.0256,
1412
+ "step": 117000
1413
+ },
1414
+ {
1415
+ "epoch": 2.78,
1416
+ "learning_rate": 3.5873978922753633e-06,
1417
+ "loss": 1.0031,
1418
+ "step": 117500
1419
+ },
1420
+ {
1421
+ "epoch": 2.8,
1422
+ "learning_rate": 3.3898974577744074e-06,
1423
+ "loss": 1.0208,
1424
+ "step": 118000
1425
+ },
1426
+ {
1427
+ "epoch": 2.81,
1428
+ "learning_rate": 3.1923970232734515e-06,
1429
+ "loss": 1.0215,
1430
+ "step": 118500
1431
+ },
1432
+ {
1433
+ "epoch": 2.82,
1434
+ "learning_rate": 2.9948965887724956e-06,
1435
+ "loss": 1.0196,
1436
+ "step": 119000
1437
+ },
1438
+ {
1439
+ "epoch": 2.83,
1440
+ "learning_rate": 2.7973961542715396e-06,
1441
+ "loss": 1.0366,
1442
+ "step": 119500
1443
+ },
1444
+ {
1445
+ "epoch": 2.84,
1446
+ "learning_rate": 2.5998957197705837e-06,
1447
+ "loss": 1.0268,
1448
+ "step": 120000
1449
+ },
1450
+ {
1451
+ "epoch": 2.86,
1452
+ "learning_rate": 2.4023952852696278e-06,
1453
+ "loss": 1.0141,
1454
+ "step": 120500
1455
+ },
1456
+ {
1457
+ "epoch": 2.87,
1458
+ "learning_rate": 2.204894850768672e-06,
1459
+ "loss": 1.0083,
1460
+ "step": 121000
1461
+ },
1462
+ {
1463
+ "epoch": 2.88,
1464
+ "learning_rate": 2.007394416267716e-06,
1465
+ "loss": 1.0266,
1466
+ "step": 121500
1467
+ },
1468
+ {
1469
+ "epoch": 2.89,
1470
+ "learning_rate": 1.80989398176676e-06,
1471
+ "loss": 1.0216,
1472
+ "step": 122000
1473
+ },
1474
+ {
1475
+ "epoch": 2.9,
1476
+ "learning_rate": 1.612393547265804e-06,
1477
+ "loss": 1.0168,
1478
+ "step": 122500
1479
+ },
1480
+ {
1481
+ "epoch": 2.92,
1482
+ "learning_rate": 1.414893112764848e-06,
1483
+ "loss": 1.0004,
1484
+ "step": 123000
1485
+ },
1486
+ {
1487
+ "epoch": 2.93,
1488
+ "learning_rate": 1.2173926782638924e-06,
1489
+ "loss": 1.026,
1490
+ "step": 123500
1491
+ },
1492
+ {
1493
+ "epoch": 2.94,
1494
+ "learning_rate": 1.0198922437629362e-06,
1495
+ "loss": 1.0042,
1496
+ "step": 124000
1497
+ },
1498
+ {
1499
+ "epoch": 2.95,
1500
+ "learning_rate": 8.223918092619805e-07,
1501
+ "loss": 0.9972,
1502
+ "step": 124500
1503
+ },
1504
+ {
1505
+ "epoch": 2.96,
1506
+ "learning_rate": 6.248913747610245e-07,
1507
+ "loss": 1.0221,
1508
+ "step": 125000
1509
+ },
1510
+ {
1511
+ "epoch": 2.97,
1512
+ "learning_rate": 4.273909402600686e-07,
1513
+ "loss": 1.02,
1514
+ "step": 125500
1515
+ },
1516
+ {
1517
+ "epoch": 2.99,
1518
+ "learning_rate": 2.2989050575911268e-07,
1519
+ "loss": 1.0402,
1520
+ "step": 126000
1521
+ },
1522
+ {
1523
+ "epoch": 3.0,
1524
+ "learning_rate": 3.239007125815677e-08,
1525
+ "loss": 1.0188,
1526
+ "step": 126500
1527
+ },
1528
+ {
1529
+ "epoch": 3.0,
1530
+ "step": 126582,
1531
+ "total_flos": 1.9038716367888384e+17,
1532
+ "train_loss": 1.3316941024496511,
1533
+ "train_runtime": 32427.393,
1534
+ "train_samples_per_second": 39.035,
1535
+ "train_steps_per_second": 3.904
1536
+ }
1537
+ ],
1538
+ "max_steps": 126582,
1539
+ "num_train_epochs": 3,
1540
+ "total_flos": 1.9038716367888384e+17,
1541
+ "trial_name": null,
1542
+ "trial_params": null
1543
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73d7d5636a3a1b734380e29cee84402bb43ece15c310b04317be27eda09a41bf
3
+ size 2799
vocab.json ADDED
The diff for this file is too large to render. See raw diff