higopires commited on
Commit
0725def
·
verified ·
1 Parent(s): efe86ab

Upload 12 files

Browse files
README.md CHANGED
@@ -1,3 +1,54 @@
1
- ---
2
- license: mit
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - generated_from_trainer
4
+ model-index:
5
+ - name: DeB3RTa_3_xsmall
6
+ results: []
7
+ ---
8
+
9
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
10
+ should probably proofread and complete it, then remove this comment. -->
11
+
12
+ # DeB3RTa_3_xsmall
13
+
14
+ This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
15
+
16
+ ## Model description
17
+
18
+ More information needed
19
+
20
+ ## Intended uses & limitations
21
+
22
+ More information needed
23
+
24
+ ## Training and evaluation data
25
+
26
+ More information needed
27
+
28
+ ## Training procedure
29
+
30
+ ### Training hyperparameters
31
+
32
+ The following hyperparameters were used during training:
33
+ - learning_rate: 0.0001
34
+ - train_batch_size: 192
35
+ - eval_batch_size: 8
36
+ - seed: 42
37
+ - gradient_accumulation_steps: 8
38
+ - total_train_batch_size: 1536
39
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-06
40
+ - lr_scheduler_type: linear
41
+ - lr_scheduler_warmup_ratio: 0.01
42
+ - num_epochs: 50.0
43
+ - mixed_precision_training: Native AMP
44
+
45
+ ### Training results
46
+
47
+
48
+
49
+ ### Framework versions
50
+
51
+ - Transformers 4.42.4
52
+ - Pytorch 2.3.1+cu121
53
+ - Datasets 2.20.0
54
+ - Tokenizers 0.19.1
added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "[MASK]": 128000
3
+ }
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 50.0,
3
+ "total_flos": 2.052104150815488e+18,
4
+ "train_loss": 0.04098836247254364,
5
+ "train_runtime": 10357.3823,
6
+ "train_samples": 2477405,
7
+ "train_samples_per_second": 11959.61,
8
+ "train_steps_per_second": 7.787
9
+ }
config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "DebertaV2ForMaskedLM"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "hidden_act": "gelu",
7
+ "hidden_dropout_prob": 0.1,
8
+ "hidden_size": 384,
9
+ "initializer_range": 0.02,
10
+ "intermediate_size": 1536,
11
+ "layer_norm_eps": 1e-07,
12
+ "max_position_embeddings": 512,
13
+ "max_relative_positions": -1,
14
+ "model_type": "deberta-v2",
15
+ "num_attention_heads": 6,
16
+ "num_hidden_layers": 12,
17
+ "pad_token_id": 0,
18
+ "pooler_dropout": 0,
19
+ "pooler_hidden_act": "gelu",
20
+ "pooler_hidden_size": 384,
21
+ "pos_att_type": null,
22
+ "position_biased_input": true,
23
+ "relative_attention": false,
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.42.4",
26
+ "type_vocab_size": 0,
27
+ "vocab_size": 128100
28
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cdea08c94f1b19046bb4f3a97813eeda9a7ca770e5c04724bf7579f59bd4dd7a
3
+ size 283856744
special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "[CLS]",
3
+ "cls_token": "[CLS]",
4
+ "eos_token": "[SEP]",
5
+ "mask_token": "[MASK]",
6
+ "pad_token": "[PAD]",
7
+ "sep_token": "[SEP]",
8
+ "unk_token": {
9
+ "content": "[UNK]",
10
+ "lstrip": false,
11
+ "normalized": true,
12
+ "rstrip": false,
13
+ "single_word": false
14
+ }
15
+ }
spm.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5598d5e96f339a8d980c15f9afd405a2e5e1be7db41de3ed13b0f03fac1e8c17
3
+ size 2447305
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[CLS]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[SEP]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[UNK]",
29
+ "lstrip": false,
30
+ "normalized": true,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "128000": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "[CLS]",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "[CLS]",
47
+ "do_lower_case": false,
48
+ "eos_token": "[SEP]",
49
+ "mask_token": "[MASK]",
50
+ "model_max_length": 1000000000000000019884624838656,
51
+ "pad_token": "[PAD]",
52
+ "sep_token": "[SEP]",
53
+ "sp_model_kwargs": {},
54
+ "split_by_punct": false,
55
+ "tokenizer_class": "DebertaV2Tokenizer",
56
+ "unk_token": "[UNK]",
57
+ "vocab_type": "spm"
58
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 50.0,
3
+ "total_flos": 2.052104150815488e+18,
4
+ "train_loss": 0.04098836247254364,
5
+ "train_runtime": 10357.3823,
6
+ "train_samples": 2477405,
7
+ "train_samples_per_second": 11959.61,
8
+ "train_steps_per_second": 7.787
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,1169 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 50.0,
5
+ "eval_steps": 500,
6
+ "global_step": 80650,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.30998140111593303,
13
+ "grad_norm": 0.3543250262737274,
14
+ "learning_rate": 6.195786864931847e-05,
15
+ "loss": 9.0345,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 0.6199628022318661,
20
+ "grad_norm": 0.5106557607650757,
21
+ "learning_rate": 9.97582756158962e-05,
22
+ "loss": 6.2184,
23
+ "step": 1000
24
+ },
25
+ {
26
+ "epoch": 0.9299442033477991,
27
+ "grad_norm": 2.6617751121520996,
28
+ "learning_rate": 9.913204664153402e-05,
29
+ "loss": 5.4194,
30
+ "step": 1500
31
+ },
32
+ {
33
+ "epoch": 1.2399256044637321,
34
+ "grad_norm": 1.8096632957458496,
35
+ "learning_rate": 9.850581766717182e-05,
36
+ "loss": 3.915,
37
+ "step": 2000
38
+ },
39
+ {
40
+ "epoch": 1.5499070055796653,
41
+ "grad_norm": 1.2520173788070679,
42
+ "learning_rate": 9.787958869280964e-05,
43
+ "loss": 2.7963,
44
+ "step": 2500
45
+ },
46
+ {
47
+ "epoch": 1.8598884066955983,
48
+ "grad_norm": 0.8099603056907654,
49
+ "learning_rate": 9.725335971844745e-05,
50
+ "loss": 2.2568,
51
+ "step": 3000
52
+ },
53
+ {
54
+ "epoch": 2.1698698078115313,
55
+ "grad_norm": 0.7233591079711914,
56
+ "learning_rate": 9.662713074408527e-05,
57
+ "loss": 1.9847,
58
+ "step": 3500
59
+ },
60
+ {
61
+ "epoch": 2.4798512089274642,
62
+ "grad_norm": 0.6427165865898132,
63
+ "learning_rate": 9.600090176972308e-05,
64
+ "loss": 1.8216,
65
+ "step": 4000
66
+ },
67
+ {
68
+ "epoch": 2.7898326100433977,
69
+ "grad_norm": 0.6729193925857544,
70
+ "learning_rate": 9.53746727953609e-05,
71
+ "loss": 1.7067,
72
+ "step": 4500
73
+ },
74
+ {
75
+ "epoch": 3.0998140111593306,
76
+ "grad_norm": 0.6484789848327637,
77
+ "learning_rate": 9.47484438209987e-05,
78
+ "loss": 1.6187,
79
+ "step": 5000
80
+ },
81
+ {
82
+ "epoch": 3.4097954122752636,
83
+ "grad_norm": 0.5950448513031006,
84
+ "learning_rate": 9.412221484663653e-05,
85
+ "loss": 1.5479,
86
+ "step": 5500
87
+ },
88
+ {
89
+ "epoch": 3.7197768133911966,
90
+ "grad_norm": 0.6102598309516907,
91
+ "learning_rate": 9.349598587227433e-05,
92
+ "loss": 1.4879,
93
+ "step": 6000
94
+ },
95
+ {
96
+ "epoch": 4.02975821450713,
97
+ "grad_norm": 0.6204754710197449,
98
+ "learning_rate": 9.286975689791215e-05,
99
+ "loss": 1.4379,
100
+ "step": 6500
101
+ },
102
+ {
103
+ "epoch": 4.3397396156230625,
104
+ "grad_norm": 0.590217649936676,
105
+ "learning_rate": 9.224352792354997e-05,
106
+ "loss": 1.3926,
107
+ "step": 7000
108
+ },
109
+ {
110
+ "epoch": 4.6497210167389955,
111
+ "grad_norm": 0.6062743663787842,
112
+ "learning_rate": 9.161729894918779e-05,
113
+ "loss": 1.3553,
114
+ "step": 7500
115
+ },
116
+ {
117
+ "epoch": 4.9597024178549285,
118
+ "grad_norm": 0.5663708448410034,
119
+ "learning_rate": 9.09910699748256e-05,
120
+ "loss": 1.3201,
121
+ "step": 8000
122
+ },
123
+ {
124
+ "epoch": 5.2696838189708615,
125
+ "grad_norm": 0.5806947350502014,
126
+ "learning_rate": 9.036484100046342e-05,
127
+ "loss": 1.2904,
128
+ "step": 8500
129
+ },
130
+ {
131
+ "epoch": 5.579665220086794,
132
+ "grad_norm": 0.6131803393363953,
133
+ "learning_rate": 8.973861202610123e-05,
134
+ "loss": 1.2623,
135
+ "step": 9000
136
+ },
137
+ {
138
+ "epoch": 5.889646621202727,
139
+ "grad_norm": 0.5666236281394958,
140
+ "learning_rate": 8.911238305173905e-05,
141
+ "loss": 1.2368,
142
+ "step": 9500
143
+ },
144
+ {
145
+ "epoch": 6.199628022318661,
146
+ "grad_norm": 0.6078547239303589,
147
+ "learning_rate": 8.848615407737685e-05,
148
+ "loss": 1.212,
149
+ "step": 10000
150
+ },
151
+ {
152
+ "epoch": 6.509609423434594,
153
+ "grad_norm": 0.575513482093811,
154
+ "learning_rate": 8.785992510301467e-05,
155
+ "loss": 1.1914,
156
+ "step": 10500
157
+ },
158
+ {
159
+ "epoch": 6.819590824550527,
160
+ "grad_norm": 0.5826976895332336,
161
+ "learning_rate": 8.723369612865248e-05,
162
+ "loss": 1.1718,
163
+ "step": 11000
164
+ },
165
+ {
166
+ "epoch": 7.12957222566646,
167
+ "grad_norm": 0.544598400592804,
168
+ "learning_rate": 8.66074671542903e-05,
169
+ "loss": 1.1548,
170
+ "step": 11500
171
+ },
172
+ {
173
+ "epoch": 7.439553626782393,
174
+ "grad_norm": 0.5824791193008423,
175
+ "learning_rate": 8.598123817992811e-05,
176
+ "loss": 1.1363,
177
+ "step": 12000
178
+ },
179
+ {
180
+ "epoch": 7.749535027898326,
181
+ "grad_norm": 0.5747692584991455,
182
+ "learning_rate": 8.535500920556593e-05,
183
+ "loss": 1.1211,
184
+ "step": 12500
185
+ },
186
+ {
187
+ "epoch": 8.05951642901426,
188
+ "grad_norm": 0.5473280549049377,
189
+ "learning_rate": 8.472878023120375e-05,
190
+ "loss": 1.1077,
191
+ "step": 13000
192
+ },
193
+ {
194
+ "epoch": 8.369497830130193,
195
+ "grad_norm": 0.5574379563331604,
196
+ "learning_rate": 8.410255125684155e-05,
197
+ "loss": 1.0908,
198
+ "step": 13500
199
+ },
200
+ {
201
+ "epoch": 8.679479231246125,
202
+ "grad_norm": 0.5424452424049377,
203
+ "learning_rate": 8.347632228247937e-05,
204
+ "loss": 1.0785,
205
+ "step": 14000
206
+ },
207
+ {
208
+ "epoch": 8.989460632362059,
209
+ "grad_norm": 0.5508283376693726,
210
+ "learning_rate": 8.285009330811718e-05,
211
+ "loss": 1.0683,
212
+ "step": 14500
213
+ },
214
+ {
215
+ "epoch": 9.299442033477991,
216
+ "grad_norm": 0.5519115924835205,
217
+ "learning_rate": 8.2223864333755e-05,
218
+ "loss": 1.0537,
219
+ "step": 15000
220
+ },
221
+ {
222
+ "epoch": 9.609423434593925,
223
+ "grad_norm": 0.5510475039482117,
224
+ "learning_rate": 8.159763535939281e-05,
225
+ "loss": 1.0443,
226
+ "step": 15500
227
+ },
228
+ {
229
+ "epoch": 9.919404835709857,
230
+ "grad_norm": 0.5631123185157776,
231
+ "learning_rate": 8.097140638503063e-05,
232
+ "loss": 1.0339,
233
+ "step": 16000
234
+ },
235
+ {
236
+ "epoch": 10.22938623682579,
237
+ "grad_norm": 0.5705382823944092,
238
+ "learning_rate": 8.034517741066844e-05,
239
+ "loss": 1.0217,
240
+ "step": 16500
241
+ },
242
+ {
243
+ "epoch": 10.539367637941723,
244
+ "grad_norm": 0.5316577553749084,
245
+ "learning_rate": 7.971894843630626e-05,
246
+ "loss": 1.0151,
247
+ "step": 17000
248
+ },
249
+ {
250
+ "epoch": 10.849349039057657,
251
+ "grad_norm": 0.5557442307472229,
252
+ "learning_rate": 7.909271946194406e-05,
253
+ "loss": 1.0043,
254
+ "step": 17500
255
+ },
256
+ {
257
+ "epoch": 11.159330440173589,
258
+ "grad_norm": 0.5498985648155212,
259
+ "learning_rate": 7.846649048758188e-05,
260
+ "loss": 0.9951,
261
+ "step": 18000
262
+ },
263
+ {
264
+ "epoch": 11.469311841289523,
265
+ "grad_norm": 0.552780032157898,
266
+ "learning_rate": 7.784026151321969e-05,
267
+ "loss": 0.9855,
268
+ "step": 18500
269
+ },
270
+ {
271
+ "epoch": 11.779293242405455,
272
+ "grad_norm": 0.5406888127326965,
273
+ "learning_rate": 7.721403253885752e-05,
274
+ "loss": 0.9795,
275
+ "step": 19000
276
+ },
277
+ {
278
+ "epoch": 12.089274643521389,
279
+ "grad_norm": 0.537375271320343,
280
+ "learning_rate": 7.658780356449533e-05,
281
+ "loss": 0.971,
282
+ "step": 19500
283
+ },
284
+ {
285
+ "epoch": 12.399256044637323,
286
+ "grad_norm": 0.5666614174842834,
287
+ "learning_rate": 7.596157459013315e-05,
288
+ "loss": 0.9643,
289
+ "step": 20000
290
+ },
291
+ {
292
+ "epoch": 12.709237445753255,
293
+ "grad_norm": 0.5302731990814209,
294
+ "learning_rate": 7.533659807371968e-05,
295
+ "loss": 0.9582,
296
+ "step": 20500
297
+ },
298
+ {
299
+ "epoch": 13.019218846869189,
300
+ "grad_norm": 0.5608243346214294,
301
+ "learning_rate": 7.471036909935749e-05,
302
+ "loss": 0.9512,
303
+ "step": 21000
304
+ },
305
+ {
306
+ "epoch": 13.32920024798512,
307
+ "grad_norm": 0.5309119820594788,
308
+ "learning_rate": 7.408414012499531e-05,
309
+ "loss": 0.9424,
310
+ "step": 21500
311
+ },
312
+ {
313
+ "epoch": 13.639181649101054,
314
+ "grad_norm": 0.5380939245223999,
315
+ "learning_rate": 7.345791115063312e-05,
316
+ "loss": 0.9383,
317
+ "step": 22000
318
+ },
319
+ {
320
+ "epoch": 13.949163050216987,
321
+ "grad_norm": 0.5440984964370728,
322
+ "learning_rate": 7.283168217627094e-05,
323
+ "loss": 0.9298,
324
+ "step": 22500
325
+ },
326
+ {
327
+ "epoch": 14.25914445133292,
328
+ "grad_norm": 0.5377441048622131,
329
+ "learning_rate": 7.220545320190874e-05,
330
+ "loss": 0.9245,
331
+ "step": 23000
332
+ },
333
+ {
334
+ "epoch": 14.569125852448852,
335
+ "grad_norm": 0.5402495265007019,
336
+ "learning_rate": 7.157922422754656e-05,
337
+ "loss": 0.9196,
338
+ "step": 23500
339
+ },
340
+ {
341
+ "epoch": 14.879107253564786,
342
+ "grad_norm": 0.5610705018043518,
343
+ "learning_rate": 7.095299525318437e-05,
344
+ "loss": 0.9146,
345
+ "step": 24000
346
+ },
347
+ {
348
+ "epoch": 15.189088654680718,
349
+ "grad_norm": 0.5305636525154114,
350
+ "learning_rate": 7.032676627882219e-05,
351
+ "loss": 0.9071,
352
+ "step": 24500
353
+ },
354
+ {
355
+ "epoch": 15.499070055796652,
356
+ "grad_norm": 0.5398979187011719,
357
+ "learning_rate": 6.970053730446e-05,
358
+ "loss": 0.9037,
359
+ "step": 25000
360
+ },
361
+ {
362
+ "epoch": 15.809051456912584,
363
+ "grad_norm": 0.5490283370018005,
364
+ "learning_rate": 6.907556078804655e-05,
365
+ "loss": 0.8982,
366
+ "step": 25500
367
+ },
368
+ {
369
+ "epoch": 16.11903285802852,
370
+ "grad_norm": 0.5505014061927795,
371
+ "learning_rate": 6.844933181368435e-05,
372
+ "loss": 0.8933,
373
+ "step": 26000
374
+ },
375
+ {
376
+ "epoch": 16.429014259144452,
377
+ "grad_norm": 0.5260488390922546,
378
+ "learning_rate": 6.782310283932217e-05,
379
+ "loss": 0.8865,
380
+ "step": 26500
381
+ },
382
+ {
383
+ "epoch": 16.738995660260386,
384
+ "grad_norm": 0.5459970235824585,
385
+ "learning_rate": 6.719687386495999e-05,
386
+ "loss": 0.8837,
387
+ "step": 27000
388
+ },
389
+ {
390
+ "epoch": 17.048977061376316,
391
+ "grad_norm": 0.5260828733444214,
392
+ "learning_rate": 6.657189734854653e-05,
393
+ "loss": 0.8812,
394
+ "step": 27500
395
+ },
396
+ {
397
+ "epoch": 17.35895846249225,
398
+ "grad_norm": 0.531878650188446,
399
+ "learning_rate": 6.594566837418435e-05,
400
+ "loss": 0.874,
401
+ "step": 28000
402
+ },
403
+ {
404
+ "epoch": 17.668939863608184,
405
+ "grad_norm": 0.5373751521110535,
406
+ "learning_rate": 6.531943939982215e-05,
407
+ "loss": 0.8703,
408
+ "step": 28500
409
+ },
410
+ {
411
+ "epoch": 17.978921264724118,
412
+ "grad_norm": 0.5685413479804993,
413
+ "learning_rate": 6.469321042545997e-05,
414
+ "loss": 0.8674,
415
+ "step": 29000
416
+ },
417
+ {
418
+ "epoch": 18.288902665840048,
419
+ "grad_norm": 0.5405117273330688,
420
+ "learning_rate": 6.406698145109778e-05,
421
+ "loss": 0.8618,
422
+ "step": 29500
423
+ },
424
+ {
425
+ "epoch": 18.598884066955982,
426
+ "grad_norm": 0.5303318500518799,
427
+ "learning_rate": 6.344325739263305e-05,
428
+ "loss": 0.8572,
429
+ "step": 30000
430
+ },
431
+ {
432
+ "epoch": 18.908865468071916,
433
+ "grad_norm": 0.5173208117485046,
434
+ "learning_rate": 6.281702841827086e-05,
435
+ "loss": 0.8552,
436
+ "step": 30500
437
+ },
438
+ {
439
+ "epoch": 19.21884686918785,
440
+ "grad_norm": 0.5334449410438538,
441
+ "learning_rate": 6.219079944390868e-05,
442
+ "loss": 0.8494,
443
+ "step": 31000
444
+ },
445
+ {
446
+ "epoch": 19.52882827030378,
447
+ "grad_norm": 0.5522080659866333,
448
+ "learning_rate": 6.156457046954649e-05,
449
+ "loss": 0.8464,
450
+ "step": 31500
451
+ },
452
+ {
453
+ "epoch": 19.838809671419714,
454
+ "grad_norm": 0.5295758247375488,
455
+ "learning_rate": 6.09383414951843e-05,
456
+ "loss": 0.845,
457
+ "step": 32000
458
+ },
459
+ {
460
+ "epoch": 20.148791072535648,
461
+ "grad_norm": 0.5164583325386047,
462
+ "learning_rate": 6.0312112520822115e-05,
463
+ "loss": 0.8395,
464
+ "step": 32500
465
+ },
466
+ {
467
+ "epoch": 20.45877247365158,
468
+ "grad_norm": 0.5620171427726746,
469
+ "learning_rate": 5.968713600440865e-05,
470
+ "loss": 0.8354,
471
+ "step": 33000
472
+ },
473
+ {
474
+ "epoch": 20.768753874767516,
475
+ "grad_norm": 0.5254458785057068,
476
+ "learning_rate": 5.906090703004646e-05,
477
+ "loss": 0.8336,
478
+ "step": 33500
479
+ },
480
+ {
481
+ "epoch": 21.078735275883446,
482
+ "grad_norm": 0.5437597632408142,
483
+ "learning_rate": 5.8434678055684276e-05,
484
+ "loss": 0.8304,
485
+ "step": 34000
486
+ },
487
+ {
488
+ "epoch": 21.38871667699938,
489
+ "grad_norm": 0.5438856482505798,
490
+ "learning_rate": 5.78084490813221e-05,
491
+ "loss": 0.8263,
492
+ "step": 34500
493
+ },
494
+ {
495
+ "epoch": 21.698698078115314,
496
+ "grad_norm": 0.5386750102043152,
497
+ "learning_rate": 5.7182220106959916e-05,
498
+ "loss": 0.8248,
499
+ "step": 35000
500
+ },
501
+ {
502
+ "epoch": 22.008679479231247,
503
+ "grad_norm": 0.5307642817497253,
504
+ "learning_rate": 5.655724359054645e-05,
505
+ "loss": 0.8223,
506
+ "step": 35500
507
+ },
508
+ {
509
+ "epoch": 22.318660880347178,
510
+ "grad_norm": 0.5404214859008789,
511
+ "learning_rate": 5.5931014616184264e-05,
512
+ "loss": 0.8176,
513
+ "step": 36000
514
+ },
515
+ {
516
+ "epoch": 22.62864228146311,
517
+ "grad_norm": 0.555665910243988,
518
+ "learning_rate": 5.530478564182208e-05,
519
+ "loss": 0.8164,
520
+ "step": 36500
521
+ },
522
+ {
523
+ "epoch": 22.938623682579045,
524
+ "grad_norm": 0.5331476330757141,
525
+ "learning_rate": 5.467855666745989e-05,
526
+ "loss": 0.8135,
527
+ "step": 37000
528
+ },
529
+ {
530
+ "epoch": 23.24860508369498,
531
+ "grad_norm": 0.541491687297821,
532
+ "learning_rate": 5.405358015104644e-05,
533
+ "loss": 0.8097,
534
+ "step": 37500
535
+ },
536
+ {
537
+ "epoch": 23.55858648481091,
538
+ "grad_norm": 0.5554507374763489,
539
+ "learning_rate": 5.342735117668425e-05,
540
+ "loss": 0.8074,
541
+ "step": 38000
542
+ },
543
+ {
544
+ "epoch": 23.868567885926844,
545
+ "grad_norm": 0.5485785007476807,
546
+ "learning_rate": 5.2801122202322065e-05,
547
+ "loss": 0.8054,
548
+ "step": 38500
549
+ },
550
+ {
551
+ "epoch": 24.178549287042777,
552
+ "grad_norm": 0.5320767164230347,
553
+ "learning_rate": 5.217489322795988e-05,
554
+ "loss": 0.8018,
555
+ "step": 39000
556
+ },
557
+ {
558
+ "epoch": 24.48853068815871,
559
+ "grad_norm": 0.5248667001724243,
560
+ "learning_rate": 5.154866425359769e-05,
561
+ "loss": 0.8008,
562
+ "step": 39500
563
+ },
564
+ {
565
+ "epoch": 24.798512089274645,
566
+ "grad_norm": 0.5368346571922302,
567
+ "learning_rate": 5.0922435279235505e-05,
568
+ "loss": 0.7975,
569
+ "step": 40000
570
+ },
571
+ {
572
+ "epoch": 25.108493490390575,
573
+ "grad_norm": 0.53144371509552,
574
+ "learning_rate": 5.029620630487332e-05,
575
+ "loss": 0.7947,
576
+ "step": 40500
577
+ },
578
+ {
579
+ "epoch": 25.41847489150651,
580
+ "grad_norm": 0.5482547879219055,
581
+ "learning_rate": 4.966997733051113e-05,
582
+ "loss": 0.793,
583
+ "step": 41000
584
+ },
585
+ {
586
+ "epoch": 25.728456292622443,
587
+ "grad_norm": 0.5446964502334595,
588
+ "learning_rate": 4.9043748356148946e-05,
589
+ "loss": 0.7905,
590
+ "step": 41500
591
+ },
592
+ {
593
+ "epoch": 26.038437693738377,
594
+ "grad_norm": 0.5257270932197571,
595
+ "learning_rate": 4.841751938178676e-05,
596
+ "loss": 0.7892,
597
+ "step": 42000
598
+ },
599
+ {
600
+ "epoch": 26.348419094854307,
601
+ "grad_norm": 0.5478941202163696,
602
+ "learning_rate": 4.779129040742457e-05,
603
+ "loss": 0.7856,
604
+ "step": 42500
605
+ },
606
+ {
607
+ "epoch": 26.65840049597024,
608
+ "grad_norm": 0.5381990671157837,
609
+ "learning_rate": 4.7165061433062386e-05,
610
+ "loss": 0.7863,
611
+ "step": 43000
612
+ },
613
+ {
614
+ "epoch": 26.968381897086175,
615
+ "grad_norm": 0.546461820602417,
616
+ "learning_rate": 4.65388324587002e-05,
617
+ "loss": 0.7826,
618
+ "step": 43500
619
+ },
620
+ {
621
+ "epoch": 27.27836329820211,
622
+ "grad_norm": 0.543404757976532,
623
+ "learning_rate": 4.591260348433802e-05,
624
+ "loss": 0.7796,
625
+ "step": 44000
626
+ },
627
+ {
628
+ "epoch": 27.58834469931804,
629
+ "grad_norm": 0.5448907613754272,
630
+ "learning_rate": 4.528637450997583e-05,
631
+ "loss": 0.7796,
632
+ "step": 44500
633
+ },
634
+ {
635
+ "epoch": 27.898326100433973,
636
+ "grad_norm": 0.5504478216171265,
637
+ "learning_rate": 4.466014553561365e-05,
638
+ "loss": 0.7761,
639
+ "step": 45000
640
+ },
641
+ {
642
+ "epoch": 28.208307501549907,
643
+ "grad_norm": 0.544154703617096,
644
+ "learning_rate": 4.403391656125146e-05,
645
+ "loss": 0.7753,
646
+ "step": 45500
647
+ },
648
+ {
649
+ "epoch": 28.51828890266584,
650
+ "grad_norm": 0.542306125164032,
651
+ "learning_rate": 4.3407687586889274e-05,
652
+ "loss": 0.7735,
653
+ "step": 46000
654
+ },
655
+ {
656
+ "epoch": 28.828270303781775,
657
+ "grad_norm": 0.5549866557121277,
658
+ "learning_rate": 4.278145861252709e-05,
659
+ "loss": 0.7707,
660
+ "step": 46500
661
+ },
662
+ {
663
+ "epoch": 29.138251704897705,
664
+ "grad_norm": 0.538090169429779,
665
+ "learning_rate": 4.21552296381649e-05,
666
+ "loss": 0.7697,
667
+ "step": 47000
668
+ },
669
+ {
670
+ "epoch": 29.44823310601364,
671
+ "grad_norm": 0.5609955191612244,
672
+ "learning_rate": 4.1529000663802714e-05,
673
+ "loss": 0.7682,
674
+ "step": 47500
675
+ },
676
+ {
677
+ "epoch": 29.758214507129573,
678
+ "grad_norm": 0.5595529675483704,
679
+ "learning_rate": 4.090277168944053e-05,
680
+ "loss": 0.7659,
681
+ "step": 48000
682
+ },
683
+ {
684
+ "epoch": 30.068195908245507,
685
+ "grad_norm": 0.5461651086807251,
686
+ "learning_rate": 4.027654271507834e-05,
687
+ "loss": 0.7656,
688
+ "step": 48500
689
+ },
690
+ {
691
+ "epoch": 30.378177309361437,
692
+ "grad_norm": 0.5438820719718933,
693
+ "learning_rate": 3.9650313740716154e-05,
694
+ "loss": 0.7625,
695
+ "step": 49000
696
+ },
697
+ {
698
+ "epoch": 30.68815871047737,
699
+ "grad_norm": 0.5458811521530151,
700
+ "learning_rate": 3.902408476635397e-05,
701
+ "loss": 0.762,
702
+ "step": 49500
703
+ },
704
+ {
705
+ "epoch": 30.998140111593305,
706
+ "grad_norm": 0.535521388053894,
707
+ "learning_rate": 3.839785579199179e-05,
708
+ "loss": 0.7589,
709
+ "step": 50000
710
+ },
711
+ {
712
+ "epoch": 31.30812151270924,
713
+ "grad_norm": 0.5407618284225464,
714
+ "learning_rate": 3.77716268176296e-05,
715
+ "loss": 0.7576,
716
+ "step": 50500
717
+ },
718
+ {
719
+ "epoch": 31.61810291382517,
720
+ "grad_norm": 0.5259741544723511,
721
+ "learning_rate": 3.7145397843267415e-05,
722
+ "loss": 0.7571,
723
+ "step": 51000
724
+ },
725
+ {
726
+ "epoch": 31.928084314941103,
727
+ "grad_norm": 0.5338233709335327,
728
+ "learning_rate": 3.651916886890523e-05,
729
+ "loss": 0.7561,
730
+ "step": 51500
731
+ },
732
+ {
733
+ "epoch": 32.23806571605704,
734
+ "grad_norm": 0.5369750261306763,
735
+ "learning_rate": 3.589293989454304e-05,
736
+ "loss": 0.7541,
737
+ "step": 52000
738
+ },
739
+ {
740
+ "epoch": 32.54804711717297,
741
+ "grad_norm": 0.5418145656585693,
742
+ "learning_rate": 3.5266710920180856e-05,
743
+ "loss": 0.7521,
744
+ "step": 52500
745
+ },
746
+ {
747
+ "epoch": 32.858028518288904,
748
+ "grad_norm": 0.533149242401123,
749
+ "learning_rate": 3.464048194581867e-05,
750
+ "loss": 0.7519,
751
+ "step": 53000
752
+ },
753
+ {
754
+ "epoch": 33.16800991940484,
755
+ "grad_norm": 0.5384135246276855,
756
+ "learning_rate": 3.401425297145648e-05,
757
+ "loss": 0.7497,
758
+ "step": 53500
759
+ },
760
+ {
761
+ "epoch": 33.47799132052077,
762
+ "grad_norm": 0.5323925018310547,
763
+ "learning_rate": 3.3388023997094296e-05,
764
+ "loss": 0.7485,
765
+ "step": 54000
766
+ },
767
+ {
768
+ "epoch": 33.7879727216367,
769
+ "grad_norm": 0.535434901714325,
770
+ "learning_rate": 3.276179502273211e-05,
771
+ "loss": 0.7472,
772
+ "step": 54500
773
+ },
774
+ {
775
+ "epoch": 34.09795412275263,
776
+ "grad_norm": 0.5496259331703186,
777
+ "learning_rate": 3.213556604836992e-05,
778
+ "loss": 0.7454,
779
+ "step": 55000
780
+ },
781
+ {
782
+ "epoch": 34.40793552386857,
783
+ "grad_norm": 0.5429278016090393,
784
+ "learning_rate": 3.150933707400774e-05,
785
+ "loss": 0.7447,
786
+ "step": 55500
787
+ },
788
+ {
789
+ "epoch": 34.7179169249845,
790
+ "grad_norm": 0.5489596724510193,
791
+ "learning_rate": 3.088310809964556e-05,
792
+ "loss": 0.7438,
793
+ "step": 56000
794
+ },
795
+ {
796
+ "epoch": 35.027898326100434,
797
+ "grad_norm": 0.5510178208351135,
798
+ "learning_rate": 3.025687912528337e-05,
799
+ "loss": 0.7416,
800
+ "step": 56500
801
+ },
802
+ {
803
+ "epoch": 35.33787972721637,
804
+ "grad_norm": 0.5540343523025513,
805
+ "learning_rate": 2.9630650150921187e-05,
806
+ "loss": 0.7401,
807
+ "step": 57000
808
+ },
809
+ {
810
+ "epoch": 35.6478611283323,
811
+ "grad_norm": 0.551895260810852,
812
+ "learning_rate": 2.9004421176559e-05,
813
+ "loss": 0.7404,
814
+ "step": 57500
815
+ },
816
+ {
817
+ "epoch": 35.957842529448236,
818
+ "grad_norm": 0.5412101149559021,
819
+ "learning_rate": 2.8378192202196814e-05,
820
+ "loss": 0.74,
821
+ "step": 58000
822
+ },
823
+ {
824
+ "epoch": 36.26782393056417,
825
+ "grad_norm": 0.5450315475463867,
826
+ "learning_rate": 2.7751963227834627e-05,
827
+ "loss": 0.7386,
828
+ "step": 58500
829
+ },
830
+ {
831
+ "epoch": 36.577805331680096,
832
+ "grad_norm": 0.5550098419189453,
833
+ "learning_rate": 2.712573425347244e-05,
834
+ "loss": 0.7382,
835
+ "step": 59000
836
+ },
837
+ {
838
+ "epoch": 36.88778673279603,
839
+ "grad_norm": 0.5502198338508606,
840
+ "learning_rate": 2.6499505279110254e-05,
841
+ "loss": 0.7345,
842
+ "step": 59500
843
+ },
844
+ {
845
+ "epoch": 37.197768133911964,
846
+ "grad_norm": 0.5401105880737305,
847
+ "learning_rate": 2.587452876269679e-05,
848
+ "loss": 0.7355,
849
+ "step": 60000
850
+ },
851
+ {
852
+ "epoch": 37.5077495350279,
853
+ "grad_norm": 0.543369710445404,
854
+ "learning_rate": 2.5248299788334605e-05,
855
+ "loss": 0.7338,
856
+ "step": 60500
857
+ },
858
+ {
859
+ "epoch": 37.81773093614383,
860
+ "grad_norm": 0.5440373420715332,
861
+ "learning_rate": 2.4622070813972422e-05,
862
+ "loss": 0.7326,
863
+ "step": 61000
864
+ },
865
+ {
866
+ "epoch": 38.127712337259766,
867
+ "grad_norm": 0.5450806021690369,
868
+ "learning_rate": 2.3995841839610235e-05,
869
+ "loss": 0.7315,
870
+ "step": 61500
871
+ },
872
+ {
873
+ "epoch": 38.4376937383757,
874
+ "grad_norm": 0.5412734746932983,
875
+ "learning_rate": 2.336961286524805e-05,
876
+ "loss": 0.7301,
877
+ "step": 62000
878
+ },
879
+ {
880
+ "epoch": 38.74767513949163,
881
+ "grad_norm": 0.5553017854690552,
882
+ "learning_rate": 2.274463634883459e-05,
883
+ "loss": 0.732,
884
+ "step": 62500
885
+ },
886
+ {
887
+ "epoch": 39.05765654060756,
888
+ "grad_norm": 0.5467730164527893,
889
+ "learning_rate": 2.2118407374472403e-05,
890
+ "loss": 0.7289,
891
+ "step": 63000
892
+ },
893
+ {
894
+ "epoch": 39.367637941723494,
895
+ "grad_norm": 0.551267683506012,
896
+ "learning_rate": 2.1492178400110216e-05,
897
+ "loss": 0.728,
898
+ "step": 63500
899
+ },
900
+ {
901
+ "epoch": 39.67761934283943,
902
+ "grad_norm": 0.5391538739204407,
903
+ "learning_rate": 2.0865949425748033e-05,
904
+ "loss": 0.7276,
905
+ "step": 64000
906
+ },
907
+ {
908
+ "epoch": 39.98760074395536,
909
+ "grad_norm": 0.5523350238800049,
910
+ "learning_rate": 2.0239720451385847e-05,
911
+ "loss": 0.7272,
912
+ "step": 64500
913
+ },
914
+ {
915
+ "epoch": 40.297582145071296,
916
+ "grad_norm": 0.5367141366004944,
917
+ "learning_rate": 1.961349147702366e-05,
918
+ "loss": 0.726,
919
+ "step": 65000
920
+ },
921
+ {
922
+ "epoch": 40.60756354618723,
923
+ "grad_norm": 0.5538766980171204,
924
+ "learning_rate": 1.8987262502661473e-05,
925
+ "loss": 0.7238,
926
+ "step": 65500
927
+ },
928
+ {
929
+ "epoch": 40.91754494730316,
930
+ "grad_norm": 0.5274632573127747,
931
+ "learning_rate": 1.8361033528299287e-05,
932
+ "loss": 0.725,
933
+ "step": 66000
934
+ },
935
+ {
936
+ "epoch": 41.2275263484191,
937
+ "grad_norm": 0.521597146987915,
938
+ "learning_rate": 1.7736057011885827e-05,
939
+ "loss": 0.7233,
940
+ "step": 66500
941
+ },
942
+ {
943
+ "epoch": 41.53750774953503,
944
+ "grad_norm": 0.5390001535415649,
945
+ "learning_rate": 1.710982803752364e-05,
946
+ "loss": 0.7225,
947
+ "step": 67000
948
+ },
949
+ {
950
+ "epoch": 41.84748915065096,
951
+ "grad_norm": 0.5474331378936768,
952
+ "learning_rate": 1.6483599063161458e-05,
953
+ "loss": 0.7218,
954
+ "step": 67500
955
+ },
956
+ {
957
+ "epoch": 42.15747055176689,
958
+ "grad_norm": 0.5352886915206909,
959
+ "learning_rate": 1.5858622546747995e-05,
960
+ "loss": 0.7213,
961
+ "step": 68000
962
+ },
963
+ {
964
+ "epoch": 42.467451952882826,
965
+ "grad_norm": 0.540053129196167,
966
+ "learning_rate": 1.5232393572385808e-05,
967
+ "loss": 0.7204,
968
+ "step": 68500
969
+ },
970
+ {
971
+ "epoch": 42.77743335399876,
972
+ "grad_norm": 0.5470998883247375,
973
+ "learning_rate": 1.4606164598023622e-05,
974
+ "loss": 0.721,
975
+ "step": 69000
976
+ },
977
+ {
978
+ "epoch": 43.08741475511469,
979
+ "grad_norm": 0.5613588094711304,
980
+ "learning_rate": 1.3979935623661435e-05,
981
+ "loss": 0.7194,
982
+ "step": 69500
983
+ },
984
+ {
985
+ "epoch": 43.39739615623063,
986
+ "grad_norm": 0.5471562743186951,
987
+ "learning_rate": 1.3354959107247974e-05,
988
+ "loss": 0.7178,
989
+ "step": 70000
990
+ },
991
+ {
992
+ "epoch": 43.70737755734656,
993
+ "grad_norm": 0.5386627912521362,
994
+ "learning_rate": 1.2728730132885787e-05,
995
+ "loss": 0.7184,
996
+ "step": 70500
997
+ },
998
+ {
999
+ "epoch": 44.017358958462495,
1000
+ "grad_norm": 0.5391978621482849,
1001
+ "learning_rate": 1.2102501158523603e-05,
1002
+ "loss": 0.7186,
1003
+ "step": 71000
1004
+ },
1005
+ {
1006
+ "epoch": 44.32734035957843,
1007
+ "grad_norm": 0.5381629467010498,
1008
+ "learning_rate": 1.1476272184161418e-05,
1009
+ "loss": 0.7168,
1010
+ "step": 71500
1011
+ },
1012
+ {
1013
+ "epoch": 44.637321760694356,
1014
+ "grad_norm": 0.5467249155044556,
1015
+ "learning_rate": 1.0850043209799233e-05,
1016
+ "loss": 0.7162,
1017
+ "step": 72000
1018
+ },
1019
+ {
1020
+ "epoch": 44.94730316181029,
1021
+ "grad_norm": 0.5548228025436401,
1022
+ "learning_rate": 1.0223814235437046e-05,
1023
+ "loss": 0.7146,
1024
+ "step": 72500
1025
+ },
1026
+ {
1027
+ "epoch": 45.25728456292622,
1028
+ "grad_norm": 0.5488151907920837,
1029
+ "learning_rate": 9.59758526107486e-06,
1030
+ "loss": 0.7152,
1031
+ "step": 73000
1032
+ },
1033
+ {
1034
+ "epoch": 45.56726596404216,
1035
+ "grad_norm": 0.5473387241363525,
1036
+ "learning_rate": 8.971356286712675e-06,
1037
+ "loss": 0.7142,
1038
+ "step": 73500
1039
+ },
1040
+ {
1041
+ "epoch": 45.87724736515809,
1042
+ "grad_norm": 0.5331913828849792,
1043
+ "learning_rate": 8.345127312350489e-06,
1044
+ "loss": 0.7155,
1045
+ "step": 74000
1046
+ },
1047
+ {
1048
+ "epoch": 46.187228766274025,
1049
+ "grad_norm": 0.5443392395973206,
1050
+ "learning_rate": 7.718898337988302e-06,
1051
+ "loss": 0.7136,
1052
+ "step": 74500
1053
+ },
1054
+ {
1055
+ "epoch": 46.49721016738996,
1056
+ "grad_norm": 0.5461409091949463,
1057
+ "learning_rate": 7.092669363626117e-06,
1058
+ "loss": 0.7148,
1059
+ "step": 75000
1060
+ },
1061
+ {
1062
+ "epoch": 46.80719156850589,
1063
+ "grad_norm": 0.5504785180091858,
1064
+ "learning_rate": 6.466440389263931e-06,
1065
+ "loss": 0.7133,
1066
+ "step": 75500
1067
+ },
1068
+ {
1069
+ "epoch": 47.11717296962182,
1070
+ "grad_norm": 0.5478015542030334,
1071
+ "learning_rate": 5.840211414901745e-06,
1072
+ "loss": 0.7125,
1073
+ "step": 76000
1074
+ },
1075
+ {
1076
+ "epoch": 47.42715437073775,
1077
+ "grad_norm": 0.5464319586753845,
1078
+ "learning_rate": 5.2139824405395585e-06,
1079
+ "loss": 0.7125,
1080
+ "step": 76500
1081
+ },
1082
+ {
1083
+ "epoch": 47.73713577185369,
1084
+ "grad_norm": 0.5370163321495056,
1085
+ "learning_rate": 4.587753466177374e-06,
1086
+ "loss": 0.7117,
1087
+ "step": 77000
1088
+ },
1089
+ {
1090
+ "epoch": 48.04711717296962,
1091
+ "grad_norm": 0.5529221892356873,
1092
+ "learning_rate": 3.961524491815188e-06,
1093
+ "loss": 0.711,
1094
+ "step": 77500
1095
+ },
1096
+ {
1097
+ "epoch": 48.357098574085555,
1098
+ "grad_norm": 0.549679160118103,
1099
+ "learning_rate": 3.3352955174530015e-06,
1100
+ "loss": 0.7112,
1101
+ "step": 78000
1102
+ },
1103
+ {
1104
+ "epoch": 48.66707997520149,
1105
+ "grad_norm": 0.5416662096977234,
1106
+ "learning_rate": 2.709066543090816e-06,
1107
+ "loss": 0.7112,
1108
+ "step": 78500
1109
+ },
1110
+ {
1111
+ "epoch": 48.97706137631742,
1112
+ "grad_norm": 0.5428098440170288,
1113
+ "learning_rate": 2.08283756872863e-06,
1114
+ "loss": 0.7109,
1115
+ "step": 79000
1116
+ },
1117
+ {
1118
+ "epoch": 49.287042777433356,
1119
+ "grad_norm": 0.5247154235839844,
1120
+ "learning_rate": 1.4566085943664442e-06,
1121
+ "loss": 0.7106,
1122
+ "step": 79500
1123
+ },
1124
+ {
1125
+ "epoch": 49.59702417854929,
1126
+ "grad_norm": 0.5486724376678467,
1127
+ "learning_rate": 8.303796200042584e-07,
1128
+ "loss": 0.7097,
1129
+ "step": 80000
1130
+ },
1131
+ {
1132
+ "epoch": 49.90700557966522,
1133
+ "grad_norm": 0.5495786070823669,
1134
+ "learning_rate": 2.0415064564207257e-07,
1135
+ "loss": 0.7106,
1136
+ "step": 80500
1137
+ },
1138
+ {
1139
+ "epoch": 50.0,
1140
+ "step": 80650,
1141
+ "total_flos": 2.052104150815488e+18,
1142
+ "train_loss": 0.04098836247254364,
1143
+ "train_runtime": 10357.3823,
1144
+ "train_samples_per_second": 11959.61,
1145
+ "train_steps_per_second": 7.787
1146
+ }
1147
+ ],
1148
+ "logging_steps": 500,
1149
+ "max_steps": 80650,
1150
+ "num_input_tokens_seen": 0,
1151
+ "num_train_epochs": 50,
1152
+ "save_steps": 500,
1153
+ "stateful_callbacks": {
1154
+ "TrainerControl": {
1155
+ "args": {
1156
+ "should_epoch_stop": false,
1157
+ "should_evaluate": false,
1158
+ "should_log": false,
1159
+ "should_save": true,
1160
+ "should_training_stop": true
1161
+ },
1162
+ "attributes": {}
1163
+ }
1164
+ },
1165
+ "total_flos": 2.052104150815488e+18,
1166
+ "train_batch_size": 192,
1167
+ "trial_name": null,
1168
+ "trial_params": null
1169
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd30ff3a126c35a62842c1cfc50eb0b17ba4cb1c25b1391133e9d8cdeca49418
3
+ size 5112