cuongtran commited on
Commit
194706b
1 Parent(s): b290a01

Add model Vietnamese Text Summarization based Roberta Pretraining

Browse files
Files changed (8) hide show
  1. config.json +170 -0
  2. optimizer.pt +3 -0
  3. pytorch_model.bin +3 -0
  4. rng_state.pth +3 -0
  5. scaler.pt +3 -0
  6. scheduler.pt +3 -0
  7. trainer_state.json +916 -0
  8. training_args.bin +3 -0
config.json ADDED
@@ -0,0 +1,170 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "EncoderDecoderModel"
4
+ ],
5
+ "decoder": {
6
+ "_name_or_path": "vinai/phobert-base",
7
+ "add_cross_attention": true,
8
+ "architectures": [
9
+ "RobertaForMaskedLM"
10
+ ],
11
+ "attention_probs_dropout_prob": 0.1,
12
+ "bad_words_ids": null,
13
+ "bos_token_id": 0,
14
+ "chunk_size_feed_forward": 0,
15
+ "classifier_dropout": null,
16
+ "decoder_start_token_id": null,
17
+ "diversity_penalty": 0.0,
18
+ "do_sample": false,
19
+ "early_stopping": false,
20
+ "encoder_no_repeat_ngram_size": 0,
21
+ "eos_token_id": 2,
22
+ "finetuning_task": null,
23
+ "forced_bos_token_id": null,
24
+ "forced_eos_token_id": null,
25
+ "gradient_checkpointing": false,
26
+ "hidden_act": "gelu",
27
+ "hidden_dropout_prob": 0.1,
28
+ "hidden_size": 768,
29
+ "id2label": {
30
+ "0": "LABEL_0",
31
+ "1": "LABEL_1"
32
+ },
33
+ "initializer_range": 0.02,
34
+ "intermediate_size": 3072,
35
+ "is_decoder": true,
36
+ "is_encoder_decoder": false,
37
+ "label2id": {
38
+ "LABEL_0": 0,
39
+ "LABEL_1": 1
40
+ },
41
+ "layer_norm_eps": 1e-05,
42
+ "length_penalty": 1.0,
43
+ "max_length": 20,
44
+ "max_position_embeddings": 258,
45
+ "min_length": 0,
46
+ "model_type": "roberta",
47
+ "no_repeat_ngram_size": 0,
48
+ "num_attention_heads": 12,
49
+ "num_beam_groups": 1,
50
+ "num_beams": 1,
51
+ "num_hidden_layers": 12,
52
+ "num_return_sequences": 1,
53
+ "output_attentions": false,
54
+ "output_hidden_states": false,
55
+ "output_scores": false,
56
+ "pad_token_id": 1,
57
+ "position_embedding_type": "absolute",
58
+ "prefix": null,
59
+ "problem_type": null,
60
+ "pruned_heads": {},
61
+ "remove_invalid_values": false,
62
+ "repetition_penalty": 1.0,
63
+ "return_dict": true,
64
+ "return_dict_in_generate": false,
65
+ "sep_token_id": null,
66
+ "task_specific_params": null,
67
+ "temperature": 1.0,
68
+ "tie_encoder_decoder": false,
69
+ "tie_word_embeddings": true,
70
+ "tokenizer_class": "PhobertTokenizer",
71
+ "top_k": 50,
72
+ "top_p": 1.0,
73
+ "torch_dtype": null,
74
+ "torchscript": false,
75
+ "transformers_version": "4.11.0",
76
+ "type_vocab_size": 1,
77
+ "use_bfloat16": false,
78
+ "use_cache": true,
79
+ "vocab_size": 64001
80
+ },
81
+ "decoder_start_token_id": 0,
82
+ "early_stopping": true,
83
+ "encoder": {
84
+ "_name_or_path": "vinai/phobert-base",
85
+ "add_cross_attention": false,
86
+ "architectures": [
87
+ "RobertaForMaskedLM"
88
+ ],
89
+ "attention_probs_dropout_prob": 0.1,
90
+ "bad_words_ids": null,
91
+ "bos_token_id": 0,
92
+ "chunk_size_feed_forward": 0,
93
+ "classifier_dropout": null,
94
+ "decoder_start_token_id": null,
95
+ "diversity_penalty": 0.0,
96
+ "do_sample": false,
97
+ "early_stopping": false,
98
+ "encoder_no_repeat_ngram_size": 0,
99
+ "eos_token_id": 2,
100
+ "finetuning_task": null,
101
+ "forced_bos_token_id": null,
102
+ "forced_eos_token_id": null,
103
+ "gradient_checkpointing": false,
104
+ "hidden_act": "gelu",
105
+ "hidden_dropout_prob": 0.1,
106
+ "hidden_size": 768,
107
+ "id2label": {
108
+ "0": "LABEL_0",
109
+ "1": "LABEL_1"
110
+ },
111
+ "initializer_range": 0.02,
112
+ "intermediate_size": 3072,
113
+ "is_decoder": false,
114
+ "is_encoder_decoder": false,
115
+ "label2id": {
116
+ "LABEL_0": 0,
117
+ "LABEL_1": 1
118
+ },
119
+ "layer_norm_eps": 1e-05,
120
+ "length_penalty": 1.0,
121
+ "max_length": 20,
122
+ "max_position_embeddings": 258,
123
+ "min_length": 0,
124
+ "model_type": "roberta",
125
+ "no_repeat_ngram_size": 0,
126
+ "num_attention_heads": 12,
127
+ "num_beam_groups": 1,
128
+ "num_beams": 1,
129
+ "num_hidden_layers": 12,
130
+ "num_return_sequences": 1,
131
+ "output_attentions": false,
132
+ "output_hidden_states": false,
133
+ "output_scores": false,
134
+ "pad_token_id": 1,
135
+ "position_embedding_type": "absolute",
136
+ "prefix": null,
137
+ "problem_type": null,
138
+ "pruned_heads": {},
139
+ "remove_invalid_values": false,
140
+ "repetition_penalty": 1.0,
141
+ "return_dict": true,
142
+ "return_dict_in_generate": false,
143
+ "sep_token_id": null,
144
+ "task_specific_params": null,
145
+ "temperature": 1.0,
146
+ "tie_encoder_decoder": false,
147
+ "tie_word_embeddings": true,
148
+ "tokenizer_class": "PhobertTokenizer",
149
+ "top_k": 50,
150
+ "top_p": 1.0,
151
+ "torch_dtype": null,
152
+ "torchscript": false,
153
+ "transformers_version": "4.11.0",
154
+ "type_vocab_size": 1,
155
+ "use_bfloat16": false,
156
+ "use_cache": true,
157
+ "vocab_size": 64001
158
+ },
159
+ "eos_token_id": 2,
160
+ "is_encoder_decoder": true,
161
+ "length_penalty": 2.0,
162
+ "max_length": 64,
163
+ "model_type": "encoder-decoder",
164
+ "no_repeat_ngram_size": 3,
165
+ "num_beams": 4,
166
+ "tie_encoder_decoder": true,
167
+ "torch_dtype": "float32",
168
+ "transformers_version": null,
169
+ "vocab_size": 64001
170
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0dbd897bd2fe2cd29026fed447d8bb50cc8b8b53ce67ab4f8190926432d9c73
3
+ size 1307632505
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83764a4e3989e7ddddaed019888b8994a10db7fde4477fcb77f7fedfeee22ff9
3
+ size 656269045
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:23f9ec8ed83def271ea94c9daedf1ee5a0049986c434a91e7a1328334f7b264b
3
+ size 14503
scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7a276583024e3b843d2859571fb964022ec0fafd3bcede635b026c2af946d38
3
+ size 559
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce6c81dc99680b99e856baccbb52daa8d07486d7e603195409c08e84ba937775
3
+ size 623
trainer_state.json ADDED
@@ -0,0 +1,916 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 4.5530429503718315,
5
+ "global_step": 30000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.03,
12
+ "learning_rate": 3.2000000000000003e-06,
13
+ "loss": 9.4799,
14
+ "step": 200
15
+ },
16
+ {
17
+ "epoch": 0.06,
18
+ "learning_rate": 6.533333333333333e-06,
19
+ "loss": 7.4409,
20
+ "step": 400
21
+ },
22
+ {
23
+ "epoch": 0.09,
24
+ "learning_rate": 9.866666666666667e-06,
25
+ "loss": 6.6304,
26
+ "step": 600
27
+ },
28
+ {
29
+ "epoch": 0.12,
30
+ "learning_rate": 1.3183333333333333e-05,
31
+ "loss": 5.8163,
32
+ "step": 800
33
+ },
34
+ {
35
+ "epoch": 0.15,
36
+ "learning_rate": 1.6516666666666667e-05,
37
+ "loss": 5.2109,
38
+ "step": 1000
39
+ },
40
+ {
41
+ "epoch": 0.18,
42
+ "learning_rate": 1.985e-05,
43
+ "loss": 4.7702,
44
+ "step": 1200
45
+ },
46
+ {
47
+ "epoch": 0.21,
48
+ "learning_rate": 2.3183333333333336e-05,
49
+ "loss": 4.4959,
50
+ "step": 1400
51
+ },
52
+ {
53
+ "epoch": 0.24,
54
+ "learning_rate": 2.6516666666666666e-05,
55
+ "loss": 4.3014,
56
+ "step": 1600
57
+ },
58
+ {
59
+ "epoch": 0.27,
60
+ "learning_rate": 2.985e-05,
61
+ "loss": 4.1589,
62
+ "step": 1800
63
+ },
64
+ {
65
+ "epoch": 0.3,
66
+ "learning_rate": 3.318333333333333e-05,
67
+ "loss": 4.0593,
68
+ "step": 2000
69
+ },
70
+ {
71
+ "epoch": 0.33,
72
+ "learning_rate": 3.651666666666667e-05,
73
+ "loss": 3.9601,
74
+ "step": 2200
75
+ },
76
+ {
77
+ "epoch": 0.36,
78
+ "learning_rate": 3.9850000000000006e-05,
79
+ "loss": 3.8843,
80
+ "step": 2400
81
+ },
82
+ {
83
+ "epoch": 0.39,
84
+ "learning_rate": 4.318333333333334e-05,
85
+ "loss": 3.8441,
86
+ "step": 2600
87
+ },
88
+ {
89
+ "epoch": 0.42,
90
+ "learning_rate": 4.651666666666667e-05,
91
+ "loss": 3.8172,
92
+ "step": 2800
93
+ },
94
+ {
95
+ "epoch": 0.46,
96
+ "learning_rate": 4.9850000000000006e-05,
97
+ "loss": 3.7464,
98
+ "step": 3000
99
+ },
100
+ {
101
+ "epoch": 0.49,
102
+ "learning_rate": 4.968108198363667e-05,
103
+ "loss": 3.7161,
104
+ "step": 3200
105
+ },
106
+ {
107
+ "epoch": 0.52,
108
+ "learning_rate": 4.934713641676407e-05,
109
+ "loss": 3.6457,
110
+ "step": 3400
111
+ },
112
+ {
113
+ "epoch": 0.55,
114
+ "learning_rate": 4.901319084989147e-05,
115
+ "loss": 3.61,
116
+ "step": 3600
117
+ },
118
+ {
119
+ "epoch": 0.58,
120
+ "learning_rate": 4.867924528301887e-05,
121
+ "loss": 3.579,
122
+ "step": 3800
123
+ },
124
+ {
125
+ "epoch": 0.61,
126
+ "learning_rate": 4.834529971614627e-05,
127
+ "loss": 3.5302,
128
+ "step": 4000
129
+ },
130
+ {
131
+ "epoch": 0.64,
132
+ "learning_rate": 4.8011354149273667e-05,
133
+ "loss": 3.5092,
134
+ "step": 4200
135
+ },
136
+ {
137
+ "epoch": 0.67,
138
+ "learning_rate": 4.767740858240107e-05,
139
+ "loss": 3.485,
140
+ "step": 4400
141
+ },
142
+ {
143
+ "epoch": 0.7,
144
+ "learning_rate": 4.734346301552847e-05,
145
+ "loss": 3.4616,
146
+ "step": 4600
147
+ },
148
+ {
149
+ "epoch": 0.73,
150
+ "learning_rate": 4.7009517448655874e-05,
151
+ "loss": 3.4466,
152
+ "step": 4800
153
+ },
154
+ {
155
+ "epoch": 0.76,
156
+ "learning_rate": 4.667557188178327e-05,
157
+ "loss": 3.4108,
158
+ "step": 5000
159
+ },
160
+ {
161
+ "epoch": 0.79,
162
+ "learning_rate": 4.6341626314910676e-05,
163
+ "loss": 3.3687,
164
+ "step": 5200
165
+ },
166
+ {
167
+ "epoch": 0.82,
168
+ "learning_rate": 4.6007680748038074e-05,
169
+ "loss": 3.3308,
170
+ "step": 5400
171
+ },
172
+ {
173
+ "epoch": 0.85,
174
+ "learning_rate": 4.567373518116547e-05,
175
+ "loss": 3.3314,
176
+ "step": 5600
177
+ },
178
+ {
179
+ "epoch": 0.88,
180
+ "learning_rate": 4.533978961429287e-05,
181
+ "loss": 3.3101,
182
+ "step": 5800
183
+ },
184
+ {
185
+ "epoch": 0.91,
186
+ "learning_rate": 4.500584404742027e-05,
187
+ "loss": 3.3053,
188
+ "step": 6000
189
+ },
190
+ {
191
+ "epoch": 0.94,
192
+ "learning_rate": 4.467189848054767e-05,
193
+ "loss": 3.2995,
194
+ "step": 6200
195
+ },
196
+ {
197
+ "epoch": 0.97,
198
+ "learning_rate": 4.433795291367508e-05,
199
+ "loss": 3.2495,
200
+ "step": 6400
201
+ },
202
+ {
203
+ "epoch": 1.0,
204
+ "learning_rate": 4.4004007346802475e-05,
205
+ "loss": 3.2186,
206
+ "step": 6600
207
+ },
208
+ {
209
+ "epoch": 1.03,
210
+ "learning_rate": 4.367006177992987e-05,
211
+ "loss": 2.9888,
212
+ "step": 6800
213
+ },
214
+ {
215
+ "epoch": 1.06,
216
+ "learning_rate": 4.333611621305728e-05,
217
+ "loss": 2.98,
218
+ "step": 7000
219
+ },
220
+ {
221
+ "epoch": 1.09,
222
+ "learning_rate": 4.3002170646184675e-05,
223
+ "loss": 2.9622,
224
+ "step": 7200
225
+ },
226
+ {
227
+ "epoch": 1.12,
228
+ "learning_rate": 4.266822507931207e-05,
229
+ "loss": 2.9751,
230
+ "step": 7400
231
+ },
232
+ {
233
+ "epoch": 1.15,
234
+ "learning_rate": 4.233427951243947e-05,
235
+ "loss": 2.9455,
236
+ "step": 7600
237
+ },
238
+ {
239
+ "epoch": 1.18,
240
+ "learning_rate": 4.200033394556687e-05,
241
+ "loss": 2.9657,
242
+ "step": 7800
243
+ },
244
+ {
245
+ "epoch": 1.21,
246
+ "learning_rate": 4.1666388378694274e-05,
247
+ "loss": 2.9428,
248
+ "step": 8000
249
+ },
250
+ {
251
+ "epoch": 1.24,
252
+ "learning_rate": 4.133244281182168e-05,
253
+ "loss": 2.96,
254
+ "step": 8200
255
+ },
256
+ {
257
+ "epoch": 1.27,
258
+ "learning_rate": 4.0998497244949076e-05,
259
+ "loss": 2.935,
260
+ "step": 8400
261
+ },
262
+ {
263
+ "epoch": 1.31,
264
+ "learning_rate": 4.0664551678076474e-05,
265
+ "loss": 2.9543,
266
+ "step": 8600
267
+ },
268
+ {
269
+ "epoch": 1.34,
270
+ "learning_rate": 4.033060611120388e-05,
271
+ "loss": 2.9213,
272
+ "step": 8800
273
+ },
274
+ {
275
+ "epoch": 1.37,
276
+ "learning_rate": 3.9996660544331277e-05,
277
+ "loss": 2.9299,
278
+ "step": 9000
279
+ },
280
+ {
281
+ "epoch": 1.4,
282
+ "learning_rate": 3.9662714977458674e-05,
283
+ "loss": 2.9429,
284
+ "step": 9200
285
+ },
286
+ {
287
+ "epoch": 1.43,
288
+ "learning_rate": 3.932876941058607e-05,
289
+ "loss": 2.936,
290
+ "step": 9400
291
+ },
292
+ {
293
+ "epoch": 1.46,
294
+ "learning_rate": 3.899649357154784e-05,
295
+ "loss": 2.8932,
296
+ "step": 9600
297
+ },
298
+ {
299
+ "epoch": 1.49,
300
+ "learning_rate": 3.8662548004675245e-05,
301
+ "loss": 2.9069,
302
+ "step": 9800
303
+ },
304
+ {
305
+ "epoch": 1.52,
306
+ "learning_rate": 3.832860243780264e-05,
307
+ "loss": 2.8815,
308
+ "step": 10000
309
+ },
310
+ {
311
+ "epoch": 1.55,
312
+ "learning_rate": 3.799465687093004e-05,
313
+ "loss": 2.8844,
314
+ "step": 10200
315
+ },
316
+ {
317
+ "epoch": 1.58,
318
+ "learning_rate": 3.766071130405744e-05,
319
+ "loss": 2.8921,
320
+ "step": 10400
321
+ },
322
+ {
323
+ "epoch": 1.61,
324
+ "learning_rate": 3.73284354650192e-05,
325
+ "loss": 2.8557,
326
+ "step": 10600
327
+ },
328
+ {
329
+ "epoch": 1.64,
330
+ "learning_rate": 3.6994489898146604e-05,
331
+ "loss": 2.8768,
332
+ "step": 10800
333
+ },
334
+ {
335
+ "epoch": 1.67,
336
+ "learning_rate": 3.6660544331274e-05,
337
+ "loss": 2.8527,
338
+ "step": 11000
339
+ },
340
+ {
341
+ "epoch": 1.7,
342
+ "learning_rate": 3.632659876440141e-05,
343
+ "loss": 2.8839,
344
+ "step": 11200
345
+ },
346
+ {
347
+ "epoch": 1.73,
348
+ "learning_rate": 3.5992653197528805e-05,
349
+ "loss": 2.8512,
350
+ "step": 11400
351
+ },
352
+ {
353
+ "epoch": 1.76,
354
+ "learning_rate": 3.565870763065621e-05,
355
+ "loss": 2.8428,
356
+ "step": 11600
357
+ },
358
+ {
359
+ "epoch": 1.79,
360
+ "learning_rate": 3.532476206378361e-05,
361
+ "loss": 2.8329,
362
+ "step": 11800
363
+ },
364
+ {
365
+ "epoch": 1.82,
366
+ "learning_rate": 3.4990816496911005e-05,
367
+ "loss": 2.8266,
368
+ "step": 12000
369
+ },
370
+ {
371
+ "epoch": 1.85,
372
+ "learning_rate": 3.46568709300384e-05,
373
+ "loss": 2.8282,
374
+ "step": 12200
375
+ },
376
+ {
377
+ "epoch": 1.88,
378
+ "learning_rate": 3.4324595091000164e-05,
379
+ "loss": 2.8202,
380
+ "step": 12400
381
+ },
382
+ {
383
+ "epoch": 1.91,
384
+ "learning_rate": 3.399064952412757e-05,
385
+ "loss": 2.8035,
386
+ "step": 12600
387
+ },
388
+ {
389
+ "epoch": 1.94,
390
+ "learning_rate": 3.365670395725497e-05,
391
+ "loss": 2.8211,
392
+ "step": 12800
393
+ },
394
+ {
395
+ "epoch": 1.97,
396
+ "learning_rate": 3.332275839038237e-05,
397
+ "loss": 2.796,
398
+ "step": 13000
399
+ },
400
+ {
401
+ "epoch": 2.0,
402
+ "learning_rate": 3.298881282350977e-05,
403
+ "loss": 2.766,
404
+ "step": 13200
405
+ },
406
+ {
407
+ "epoch": 2.03,
408
+ "learning_rate": 3.265486725663717e-05,
409
+ "loss": 2.5018,
410
+ "step": 13400
411
+ },
412
+ {
413
+ "epoch": 2.06,
414
+ "learning_rate": 3.232092168976457e-05,
415
+ "loss": 2.5239,
416
+ "step": 13600
417
+ },
418
+ {
419
+ "epoch": 2.09,
420
+ "learning_rate": 3.198697612289197e-05,
421
+ "loss": 2.4989,
422
+ "step": 13800
423
+ },
424
+ {
425
+ "epoch": 2.12,
426
+ "learning_rate": 3.165303055601937e-05,
427
+ "loss": 2.5141,
428
+ "step": 14000
429
+ },
430
+ {
431
+ "epoch": 2.16,
432
+ "learning_rate": 3.1319084989146766e-05,
433
+ "loss": 2.4981,
434
+ "step": 14200
435
+ },
436
+ {
437
+ "epoch": 2.19,
438
+ "learning_rate": 3.098513942227417e-05,
439
+ "loss": 2.5149,
440
+ "step": 14400
441
+ },
442
+ {
443
+ "epoch": 2.22,
444
+ "learning_rate": 3.0651193855401575e-05,
445
+ "loss": 2.4817,
446
+ "step": 14600
447
+ },
448
+ {
449
+ "epoch": 2.25,
450
+ "learning_rate": 3.031724828852897e-05,
451
+ "loss": 2.5131,
452
+ "step": 14800
453
+ },
454
+ {
455
+ "epoch": 2.28,
456
+ "learning_rate": 2.998330272165637e-05,
457
+ "loss": 2.5058,
458
+ "step": 15000
459
+ },
460
+ {
461
+ "epoch": 2.31,
462
+ "learning_rate": 2.964935715478377e-05,
463
+ "loss": 2.5254,
464
+ "step": 15200
465
+ },
466
+ {
467
+ "epoch": 2.34,
468
+ "learning_rate": 2.9315411587911173e-05,
469
+ "loss": 2.5074,
470
+ "step": 15400
471
+ },
472
+ {
473
+ "epoch": 2.37,
474
+ "learning_rate": 2.898146602103857e-05,
475
+ "loss": 2.5147,
476
+ "step": 15600
477
+ },
478
+ {
479
+ "epoch": 2.4,
480
+ "learning_rate": 2.8647520454165972e-05,
481
+ "loss": 2.4908,
482
+ "step": 15800
483
+ },
484
+ {
485
+ "epoch": 2.43,
486
+ "learning_rate": 2.8315244615127734e-05,
487
+ "loss": 2.478,
488
+ "step": 16000
489
+ },
490
+ {
491
+ "epoch": 2.46,
492
+ "learning_rate": 2.7981299048255138e-05,
493
+ "loss": 2.5051,
494
+ "step": 16200
495
+ },
496
+ {
497
+ "epoch": 2.49,
498
+ "learning_rate": 2.7647353481382536e-05,
499
+ "loss": 2.5101,
500
+ "step": 16400
501
+ },
502
+ {
503
+ "epoch": 2.52,
504
+ "learning_rate": 2.7313407914509937e-05,
505
+ "loss": 2.502,
506
+ "step": 16600
507
+ },
508
+ {
509
+ "epoch": 2.55,
510
+ "learning_rate": 2.6979462347637335e-05,
511
+ "loss": 2.5053,
512
+ "step": 16800
513
+ },
514
+ {
515
+ "epoch": 2.58,
516
+ "learning_rate": 2.6645516780764733e-05,
517
+ "loss": 2.4969,
518
+ "step": 17000
519
+ },
520
+ {
521
+ "epoch": 2.61,
522
+ "learning_rate": 2.6311571213892138e-05,
523
+ "loss": 2.4882,
524
+ "step": 17200
525
+ },
526
+ {
527
+ "epoch": 2.64,
528
+ "learning_rate": 2.597762564701954e-05,
529
+ "loss": 2.4755,
530
+ "step": 17400
531
+ },
532
+ {
533
+ "epoch": 2.67,
534
+ "learning_rate": 2.5643680080146937e-05,
535
+ "loss": 2.4798,
536
+ "step": 17600
537
+ },
538
+ {
539
+ "epoch": 2.7,
540
+ "learning_rate": 2.5309734513274335e-05,
541
+ "loss": 2.4795,
542
+ "step": 17800
543
+ },
544
+ {
545
+ "epoch": 2.73,
546
+ "learning_rate": 2.4975788946401736e-05,
547
+ "loss": 2.4913,
548
+ "step": 18000
549
+ },
550
+ {
551
+ "epoch": 2.76,
552
+ "learning_rate": 2.464184337952914e-05,
553
+ "loss": 2.4894,
554
+ "step": 18200
555
+ },
556
+ {
557
+ "epoch": 2.79,
558
+ "learning_rate": 2.430789781265654e-05,
559
+ "loss": 2.4678,
560
+ "step": 18400
561
+ },
562
+ {
563
+ "epoch": 2.82,
564
+ "learning_rate": 2.3973952245783936e-05,
565
+ "loss": 2.4418,
566
+ "step": 18600
567
+ },
568
+ {
569
+ "epoch": 2.85,
570
+ "learning_rate": 2.3640006678911338e-05,
571
+ "loss": 2.4705,
572
+ "step": 18800
573
+ },
574
+ {
575
+ "epoch": 2.88,
576
+ "learning_rate": 2.330606111203874e-05,
577
+ "loss": 2.4659,
578
+ "step": 19000
579
+ },
580
+ {
581
+ "epoch": 2.91,
582
+ "learning_rate": 2.297211554516614e-05,
583
+ "loss": 2.4752,
584
+ "step": 19200
585
+ },
586
+ {
587
+ "epoch": 2.94,
588
+ "learning_rate": 2.2638169978293538e-05,
589
+ "loss": 2.4694,
590
+ "step": 19400
591
+ },
592
+ {
593
+ "epoch": 2.97,
594
+ "learning_rate": 2.230422441142094e-05,
595
+ "loss": 2.4442,
596
+ "step": 19600
597
+ },
598
+ {
599
+ "epoch": 3.01,
600
+ "learning_rate": 2.197027884454834e-05,
601
+ "loss": 2.4312,
602
+ "step": 19800
603
+ },
604
+ {
605
+ "epoch": 3.04,
606
+ "learning_rate": 2.1636333277675742e-05,
607
+ "loss": 2.2147,
608
+ "step": 20000
609
+ },
610
+ {
611
+ "epoch": 3.07,
612
+ "learning_rate": 2.130238771080314e-05,
613
+ "loss": 2.2146,
614
+ "step": 20200
615
+ },
616
+ {
617
+ "epoch": 3.1,
618
+ "learning_rate": 2.0968442143930538e-05,
619
+ "loss": 2.213,
620
+ "step": 20400
621
+ },
622
+ {
623
+ "epoch": 3.13,
624
+ "learning_rate": 2.0634496577057942e-05,
625
+ "loss": 2.2176,
626
+ "step": 20600
627
+ },
628
+ {
629
+ "epoch": 3.16,
630
+ "learning_rate": 2.030055101018534e-05,
631
+ "loss": 2.2169,
632
+ "step": 20800
633
+ },
634
+ {
635
+ "epoch": 3.19,
636
+ "learning_rate": 1.9968275171147105e-05,
637
+ "loss": 2.223,
638
+ "step": 21000
639
+ },
640
+ {
641
+ "epoch": 3.22,
642
+ "learning_rate": 1.9634329604274503e-05,
643
+ "loss": 2.2279,
644
+ "step": 21200
645
+ },
646
+ {
647
+ "epoch": 3.25,
648
+ "learning_rate": 1.9300384037401907e-05,
649
+ "loss": 2.2322,
650
+ "step": 21400
651
+ },
652
+ {
653
+ "epoch": 3.28,
654
+ "learning_rate": 1.8966438470529305e-05,
655
+ "loss": 2.2428,
656
+ "step": 21600
657
+ },
658
+ {
659
+ "epoch": 3.31,
660
+ "learning_rate": 1.8632492903656703e-05,
661
+ "loss": 2.235,
662
+ "step": 21800
663
+ },
664
+ {
665
+ "epoch": 3.34,
666
+ "learning_rate": 1.8298547336784104e-05,
667
+ "loss": 2.2323,
668
+ "step": 22000
669
+ },
670
+ {
671
+ "epoch": 3.37,
672
+ "learning_rate": 1.7964601769911506e-05,
673
+ "loss": 2.1997,
674
+ "step": 22200
675
+ },
676
+ {
677
+ "epoch": 3.4,
678
+ "learning_rate": 1.7630656203038907e-05,
679
+ "loss": 2.2353,
680
+ "step": 22400
681
+ },
682
+ {
683
+ "epoch": 3.43,
684
+ "learning_rate": 1.7296710636166305e-05,
685
+ "loss": 2.2258,
686
+ "step": 22600
687
+ },
688
+ {
689
+ "epoch": 3.46,
690
+ "learning_rate": 1.6962765069293706e-05,
691
+ "loss": 2.2269,
692
+ "step": 22800
693
+ },
694
+ {
695
+ "epoch": 3.49,
696
+ "learning_rate": 1.6628819502421107e-05,
697
+ "loss": 2.2351,
698
+ "step": 23000
699
+ },
700
+ {
701
+ "epoch": 3.52,
702
+ "learning_rate": 1.629654366338287e-05,
703
+ "loss": 2.2387,
704
+ "step": 23200
705
+ },
706
+ {
707
+ "epoch": 3.55,
708
+ "learning_rate": 1.596259809651027e-05,
709
+ "loss": 2.2285,
710
+ "step": 23400
711
+ },
712
+ {
713
+ "epoch": 3.58,
714
+ "learning_rate": 1.5628652529637668e-05,
715
+ "loss": 2.2378,
716
+ "step": 23600
717
+ },
718
+ {
719
+ "epoch": 3.61,
720
+ "learning_rate": 1.5294706962765072e-05,
721
+ "loss": 2.2274,
722
+ "step": 23800
723
+ },
724
+ {
725
+ "epoch": 3.64,
726
+ "learning_rate": 1.496076139589247e-05,
727
+ "loss": 2.2081,
728
+ "step": 24000
729
+ },
730
+ {
731
+ "epoch": 3.67,
732
+ "learning_rate": 1.4626815829019871e-05,
733
+ "loss": 2.2183,
734
+ "step": 24200
735
+ },
736
+ {
737
+ "epoch": 3.7,
738
+ "learning_rate": 1.4292870262147271e-05,
739
+ "loss": 2.2316,
740
+ "step": 24400
741
+ },
742
+ {
743
+ "epoch": 3.73,
744
+ "learning_rate": 1.3958924695274672e-05,
745
+ "loss": 2.2066,
746
+ "step": 24600
747
+ },
748
+ {
749
+ "epoch": 3.76,
750
+ "learning_rate": 1.3624979128402072e-05,
751
+ "loss": 2.2214,
752
+ "step": 24800
753
+ },
754
+ {
755
+ "epoch": 3.79,
756
+ "learning_rate": 1.329103356152947e-05,
757
+ "loss": 2.2308,
758
+ "step": 25000
759
+ },
760
+ {
761
+ "epoch": 3.82,
762
+ "learning_rate": 1.2957087994656873e-05,
763
+ "loss": 2.2389,
764
+ "step": 25200
765
+ },
766
+ {
767
+ "epoch": 3.85,
768
+ "learning_rate": 1.262314242778427e-05,
769
+ "loss": 2.226,
770
+ "step": 25400
771
+ },
772
+ {
773
+ "epoch": 3.89,
774
+ "learning_rate": 1.2289196860911672e-05,
775
+ "loss": 2.2305,
776
+ "step": 25600
777
+ },
778
+ {
779
+ "epoch": 3.92,
780
+ "learning_rate": 1.1955251294039071e-05,
781
+ "loss": 2.2206,
782
+ "step": 25800
783
+ },
784
+ {
785
+ "epoch": 3.95,
786
+ "learning_rate": 1.1621305727166473e-05,
787
+ "loss": 2.2303,
788
+ "step": 26000
789
+ },
790
+ {
791
+ "epoch": 3.98,
792
+ "learning_rate": 1.1287360160293872e-05,
793
+ "loss": 2.218,
794
+ "step": 26200
795
+ },
796
+ {
797
+ "epoch": 4.01,
798
+ "learning_rate": 1.0953414593421272e-05,
799
+ "loss": 2.1781,
800
+ "step": 26400
801
+ },
802
+ {
803
+ "epoch": 4.04,
804
+ "learning_rate": 1.0619469026548673e-05,
805
+ "loss": 2.0351,
806
+ "step": 26600
807
+ },
808
+ {
809
+ "epoch": 4.07,
810
+ "learning_rate": 1.0285523459676073e-05,
811
+ "loss": 2.0511,
812
+ "step": 26800
813
+ },
814
+ {
815
+ "epoch": 4.1,
816
+ "learning_rate": 9.951577892803474e-06,
817
+ "loss": 2.0529,
818
+ "step": 27000
819
+ },
820
+ {
821
+ "epoch": 4.13,
822
+ "learning_rate": 9.617632325930873e-06,
823
+ "loss": 2.0306,
824
+ "step": 27200
825
+ },
826
+ {
827
+ "epoch": 4.16,
828
+ "learning_rate": 9.283686759058275e-06,
829
+ "loss": 2.0486,
830
+ "step": 27400
831
+ },
832
+ {
833
+ "epoch": 4.19,
834
+ "learning_rate": 8.949741192185674e-06,
835
+ "loss": 2.0505,
836
+ "step": 27600
837
+ },
838
+ {
839
+ "epoch": 4.22,
840
+ "learning_rate": 8.615795625313076e-06,
841
+ "loss": 2.0395,
842
+ "step": 27800
843
+ },
844
+ {
845
+ "epoch": 4.25,
846
+ "learning_rate": 8.281850058440473e-06,
847
+ "loss": 2.0387,
848
+ "step": 28000
849
+ },
850
+ {
851
+ "epoch": 4.28,
852
+ "learning_rate": 7.947904491567875e-06,
853
+ "loss": 2.0537,
854
+ "step": 28200
855
+ },
856
+ {
857
+ "epoch": 4.31,
858
+ "learning_rate": 7.615628652529638e-06,
859
+ "loss": 2.0339,
860
+ "step": 28400
861
+ },
862
+ {
863
+ "epoch": 4.34,
864
+ "learning_rate": 7.281683085657039e-06,
865
+ "loss": 2.0676,
866
+ "step": 28600
867
+ },
868
+ {
869
+ "epoch": 4.37,
870
+ "learning_rate": 6.947737518784439e-06,
871
+ "loss": 2.0498,
872
+ "step": 28800
873
+ },
874
+ {
875
+ "epoch": 4.4,
876
+ "learning_rate": 6.61379195191184e-06,
877
+ "loss": 2.0341,
878
+ "step": 29000
879
+ },
880
+ {
881
+ "epoch": 4.43,
882
+ "learning_rate": 6.279846385039238e-06,
883
+ "loss": 2.0446,
884
+ "step": 29200
885
+ },
886
+ {
887
+ "epoch": 4.46,
888
+ "learning_rate": 5.94590081816664e-06,
889
+ "loss": 2.0435,
890
+ "step": 29400
891
+ },
892
+ {
893
+ "epoch": 4.49,
894
+ "learning_rate": 5.611955251294039e-06,
895
+ "loss": 2.0513,
896
+ "step": 29600
897
+ },
898
+ {
899
+ "epoch": 4.52,
900
+ "learning_rate": 5.27800968442144e-06,
901
+ "loss": 2.0378,
902
+ "step": 29800
903
+ },
904
+ {
905
+ "epoch": 4.55,
906
+ "learning_rate": 4.94406411754884e-06,
907
+ "loss": 2.0463,
908
+ "step": 30000
909
+ }
910
+ ],
911
+ "max_steps": 32945,
912
+ "num_train_epochs": 5,
913
+ "total_flos": 1.0567419081666048e+17,
914
+ "trial_name": null,
915
+ "trial_params": null
916
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5a513a5d83ba797279e54192c065f3206dbfc8dd98ea7d0788de591927d78b5
3
+ size 2991