metaphors commited on
Commit
4a3a727
·
verified ·
1 Parent(s): 4ad789d

Upload 7 files

Browse files
config.json ADDED
@@ -0,0 +1,58 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "../model/cino-small-v2",
3
+ "architectures": [
4
+ "XLMRobertaForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "classifier_dropout": null,
9
+ "eos_token_id": 2,
10
+ "gradient_checkpointing": false,
11
+ "hidden_act": "gelu",
12
+ "hidden_dropout_prob": 0.1,
13
+ "hidden_size": 768,
14
+ "id2label": {
15
+ "0": "Politics",
16
+ "1": "Economics",
17
+ "2": "Education",
18
+ "3": "Tourism",
19
+ "4": "Environment",
20
+ "5": "Language",
21
+ "6": "Literature",
22
+ "7": "Religion",
23
+ "8": "Arts",
24
+ "9": "Medicine",
25
+ "10": "Customs",
26
+ "11": "Instruments"
27
+ },
28
+ "initializer_range": 0.02,
29
+ "intermediate_size": 3072,
30
+ "label2id": {
31
+ "Arts": 8,
32
+ "Customs": 10,
33
+ "Economics": 1,
34
+ "Education": 2,
35
+ "Environment": 4,
36
+ "Instruments": 11,
37
+ "Language": 5,
38
+ "Literature": 6,
39
+ "Medicine": 9,
40
+ "Politics": 0,
41
+ "Religion": 7,
42
+ "Tourism": 3
43
+ },
44
+ "layer_norm_eps": 1e-05,
45
+ "max_position_embeddings": 514,
46
+ "model_type": "xlm-roberta",
47
+ "num_attention_heads": 12,
48
+ "num_hidden_layers": 6,
49
+ "output_past": true,
50
+ "pad_token_id": 1,
51
+ "position_embedding_type": "absolute",
52
+ "problem_type": "single_label_classification",
53
+ "torch_dtype": "float32",
54
+ "transformers_version": "4.36.2",
55
+ "type_vocab_size": 1,
56
+ "use_cache": true,
57
+ "vocab_size": 135359
58
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5522f112bf48c042eae8056e166a8ae16eafca66718487467d3eaf88d5d5d54
3
+ size 589953711
sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:abec8706178924453be115cd2da858ef32de70ba60d0c10300822a732a868cf7
3
+ size 2814898
special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
tokenizer_config.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "135358": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "bos_token": "<s>",
45
+ "clean_up_tokenization_spaces": true,
46
+ "cls_token": "<s>",
47
+ "eos_token": "</s>",
48
+ "mask_token": "<mask>",
49
+ "model_max_length": 1000000000000000019884624838656,
50
+ "pad_token": "<pad>",
51
+ "sep_token": "</s>",
52
+ "sp_model_kwargs": {},
53
+ "tokenizer_class": "XLMRobertaTokenizer",
54
+ "unk_token": "<unk>"
55
+ }
trainer_state.json ADDED
@@ -0,0 +1,729 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7314804685814131,
3
+ "best_model_checkpoint": "../saved_model/cino-small-v2_tncc-document_v3/checkpoint-1386",
4
+ "epoch": 40.0,
5
+ "eval_steps": 500,
6
+ "global_step": 9240,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "eval_accuracy": 0.5728260869565217,
14
+ "eval_loss": 1.4385737180709839,
15
+ "eval_macro-f1": 0.3838604227373153,
16
+ "eval_macro-precision": 0.4180870918370973,
17
+ "eval_macro-recall": 0.401101362786273,
18
+ "eval_runtime": 3.7275,
19
+ "eval_samples_per_second": 246.815,
20
+ "eval_steps_per_second": 7.78,
21
+ "eval_weighted-f1": 0.5305606310926694,
22
+ "eval_weighted-precision": 0.5297428752450334,
23
+ "eval_weighted-recall": 0.5728260869565217,
24
+ "step": 231
25
+ },
26
+ {
27
+ "epoch": 2.0,
28
+ "eval_accuracy": 0.6804347826086956,
29
+ "eval_loss": 0.9704039692878723,
30
+ "eval_macro-f1": 0.5670979693101942,
31
+ "eval_macro-precision": 0.6248049410073685,
32
+ "eval_macro-recall": 0.5626229599238465,
33
+ "eval_runtime": 3.649,
34
+ "eval_samples_per_second": 252.121,
35
+ "eval_steps_per_second": 7.947,
36
+ "eval_weighted-f1": 0.659306294880126,
37
+ "eval_weighted-precision": 0.6612128534838714,
38
+ "eval_weighted-recall": 0.6804347826086956,
39
+ "step": 462
40
+ },
41
+ {
42
+ "epoch": 2.16,
43
+ "learning_rate": 2.7056277056277056e-05,
44
+ "loss": 1.5683,
45
+ "step": 500
46
+ },
47
+ {
48
+ "epoch": 3.0,
49
+ "eval_accuracy": 0.7228260869565217,
50
+ "eval_loss": 0.8554874658584595,
51
+ "eval_macro-f1": 0.6439019323461226,
52
+ "eval_macro-precision": 0.7591831945218918,
53
+ "eval_macro-recall": 0.647150694134015,
54
+ "eval_runtime": 3.6619,
55
+ "eval_samples_per_second": 251.235,
56
+ "eval_steps_per_second": 7.919,
57
+ "eval_weighted-f1": 0.7074240890463663,
58
+ "eval_weighted-precision": 0.7364652520925159,
59
+ "eval_weighted-recall": 0.7228260869565217,
60
+ "step": 693
61
+ },
62
+ {
63
+ "epoch": 4.0,
64
+ "eval_accuracy": 0.7217391304347827,
65
+ "eval_loss": 0.8149855136871338,
66
+ "eval_macro-f1": 0.692605745410065,
67
+ "eval_macro-precision": 0.7440547514586789,
68
+ "eval_macro-recall": 0.6752749193421123,
69
+ "eval_runtime": 3.6631,
70
+ "eval_samples_per_second": 251.153,
71
+ "eval_steps_per_second": 7.917,
72
+ "eval_weighted-f1": 0.7209306678998675,
73
+ "eval_weighted-precision": 0.7414800435950141,
74
+ "eval_weighted-recall": 0.7217391304347827,
75
+ "step": 924
76
+ },
77
+ {
78
+ "epoch": 4.33,
79
+ "learning_rate": 4.9543049543049543e-05,
80
+ "loss": 0.8117,
81
+ "step": 1000
82
+ },
83
+ {
84
+ "epoch": 5.0,
85
+ "eval_accuracy": 0.7380434782608696,
86
+ "eval_loss": 0.7939425110816956,
87
+ "eval_macro-f1": 0.6992459555314743,
88
+ "eval_macro-precision": 0.7591820184313646,
89
+ "eval_macro-recall": 0.6754292333806964,
90
+ "eval_runtime": 3.6533,
91
+ "eval_samples_per_second": 251.83,
92
+ "eval_steps_per_second": 7.938,
93
+ "eval_weighted-f1": 0.7380000414393602,
94
+ "eval_weighted-precision": 0.7564326290768578,
95
+ "eval_weighted-recall": 0.7380434782608696,
96
+ "step": 1155
97
+ },
98
+ {
99
+ "epoch": 6.0,
100
+ "eval_accuracy": 0.758695652173913,
101
+ "eval_loss": 0.7805317640304565,
102
+ "eval_macro-f1": 0.7314804685814131,
103
+ "eval_macro-precision": 0.7488860778713532,
104
+ "eval_macro-recall": 0.7209236372848382,
105
+ "eval_runtime": 3.6563,
106
+ "eval_samples_per_second": 251.62,
107
+ "eval_steps_per_second": 7.932,
108
+ "eval_weighted-f1": 0.7573646414610792,
109
+ "eval_weighted-precision": 0.7607162388274387,
110
+ "eval_weighted-recall": 0.758695652173913,
111
+ "step": 1386
112
+ },
113
+ {
114
+ "epoch": 6.49,
115
+ "learning_rate": 4.653679653679654e-05,
116
+ "loss": 0.5138,
117
+ "step": 1500
118
+ },
119
+ {
120
+ "epoch": 7.0,
121
+ "eval_accuracy": 0.7271739130434782,
122
+ "eval_loss": 0.9128503799438477,
123
+ "eval_macro-f1": 0.6932349275628739,
124
+ "eval_macro-precision": 0.709503845500428,
125
+ "eval_macro-recall": 0.6965007036577074,
126
+ "eval_runtime": 3.6549,
127
+ "eval_samples_per_second": 251.715,
128
+ "eval_steps_per_second": 7.934,
129
+ "eval_weighted-f1": 0.7248736247795886,
130
+ "eval_weighted-precision": 0.7328289572198895,
131
+ "eval_weighted-recall": 0.7271739130434782,
132
+ "step": 1617
133
+ },
134
+ {
135
+ "epoch": 8.0,
136
+ "eval_accuracy": 0.7217391304347827,
137
+ "eval_loss": 0.9858090877532959,
138
+ "eval_macro-f1": 0.689163447413493,
139
+ "eval_macro-precision": 0.7090289269200102,
140
+ "eval_macro-recall": 0.6776571823481318,
141
+ "eval_runtime": 3.6807,
142
+ "eval_samples_per_second": 249.952,
143
+ "eval_steps_per_second": 7.879,
144
+ "eval_weighted-f1": 0.7165836364208036,
145
+ "eval_weighted-precision": 0.7206730807638884,
146
+ "eval_weighted-recall": 0.7217391304347827,
147
+ "step": 1848
148
+ },
149
+ {
150
+ "epoch": 8.66,
151
+ "learning_rate": 4.3530543530543535e-05,
152
+ "loss": 0.2896,
153
+ "step": 2000
154
+ },
155
+ {
156
+ "epoch": 9.0,
157
+ "eval_accuracy": 0.7206521739130435,
158
+ "eval_loss": 1.0959097146987915,
159
+ "eval_macro-f1": 0.6881695092418435,
160
+ "eval_macro-precision": 0.7242135136847904,
161
+ "eval_macro-recall": 0.6739069406781741,
162
+ "eval_runtime": 3.6639,
163
+ "eval_samples_per_second": 251.1,
164
+ "eval_steps_per_second": 7.915,
165
+ "eval_weighted-f1": 0.721541258098875,
166
+ "eval_weighted-precision": 0.7354744485281628,
167
+ "eval_weighted-recall": 0.7206521739130435,
168
+ "step": 2079
169
+ },
170
+ {
171
+ "epoch": 10.0,
172
+ "eval_accuracy": 0.7402173913043478,
173
+ "eval_loss": 1.0843907594680786,
174
+ "eval_macro-f1": 0.7110726162631839,
175
+ "eval_macro-precision": 0.7142340249185155,
176
+ "eval_macro-recall": 0.7171490983289037,
177
+ "eval_runtime": 3.6707,
178
+ "eval_samples_per_second": 250.635,
179
+ "eval_steps_per_second": 7.9,
180
+ "eval_weighted-f1": 0.7389745804216191,
181
+ "eval_weighted-precision": 0.7454086875631871,
182
+ "eval_weighted-recall": 0.7402173913043478,
183
+ "step": 2310
184
+ },
185
+ {
186
+ "epoch": 10.82,
187
+ "learning_rate": 4.0524290524290525e-05,
188
+ "loss": 0.1697,
189
+ "step": 2500
190
+ },
191
+ {
192
+ "epoch": 11.0,
193
+ "eval_accuracy": 0.7445652173913043,
194
+ "eval_loss": 1.233216643333435,
195
+ "eval_macro-f1": 0.7174385595847431,
196
+ "eval_macro-precision": 0.7222220197958554,
197
+ "eval_macro-recall": 0.7223862957961207,
198
+ "eval_runtime": 3.6639,
199
+ "eval_samples_per_second": 251.1,
200
+ "eval_steps_per_second": 7.915,
201
+ "eval_weighted-f1": 0.7415516921744333,
202
+ "eval_weighted-precision": 0.7454222933021499,
203
+ "eval_weighted-recall": 0.7445652173913043,
204
+ "step": 2541
205
+ },
206
+ {
207
+ "epoch": 12.0,
208
+ "eval_accuracy": 0.725,
209
+ "eval_loss": 1.3816906213760376,
210
+ "eval_macro-f1": 0.7016354693909223,
211
+ "eval_macro-precision": 0.7189672090787159,
212
+ "eval_macro-recall": 0.6935770369995492,
213
+ "eval_runtime": 3.6765,
214
+ "eval_samples_per_second": 250.236,
215
+ "eval_steps_per_second": 7.888,
216
+ "eval_weighted-f1": 0.724330602886681,
217
+ "eval_weighted-precision": 0.7285735967072182,
218
+ "eval_weighted-recall": 0.725,
219
+ "step": 2772
220
+ },
221
+ {
222
+ "epoch": 12.99,
223
+ "learning_rate": 3.751803751803752e-05,
224
+ "loss": 0.1103,
225
+ "step": 3000
226
+ },
227
+ {
228
+ "epoch": 13.0,
229
+ "eval_accuracy": 0.7489130434782608,
230
+ "eval_loss": 1.3391927480697632,
231
+ "eval_macro-f1": 0.7238074657296613,
232
+ "eval_macro-precision": 0.7446967874522551,
233
+ "eval_macro-recall": 0.712629424508482,
234
+ "eval_runtime": 3.6667,
235
+ "eval_samples_per_second": 250.906,
236
+ "eval_steps_per_second": 7.909,
237
+ "eval_weighted-f1": 0.7501963340512701,
238
+ "eval_weighted-precision": 0.7590165879471408,
239
+ "eval_weighted-recall": 0.7489130434782608,
240
+ "step": 3003
241
+ },
242
+ {
243
+ "epoch": 14.0,
244
+ "eval_accuracy": 0.7402173913043478,
245
+ "eval_loss": 1.5429991483688354,
246
+ "eval_macro-f1": 0.7240452372603429,
247
+ "eval_macro-precision": 0.7331291046121945,
248
+ "eval_macro-recall": 0.7243088646296482,
249
+ "eval_runtime": 3.6687,
250
+ "eval_samples_per_second": 250.767,
251
+ "eval_steps_per_second": 7.905,
252
+ "eval_weighted-f1": 0.7380284868051685,
253
+ "eval_weighted-precision": 0.7450380621081162,
254
+ "eval_weighted-recall": 0.7402173913043478,
255
+ "step": 3234
256
+ },
257
+ {
258
+ "epoch": 15.0,
259
+ "eval_accuracy": 0.741304347826087,
260
+ "eval_loss": 1.5235261917114258,
261
+ "eval_macro-f1": 0.7183262530521685,
262
+ "eval_macro-precision": 0.7374055253765689,
263
+ "eval_macro-recall": 0.710378384840802,
264
+ "eval_runtime": 3.6757,
265
+ "eval_samples_per_second": 250.291,
266
+ "eval_steps_per_second": 7.89,
267
+ "eval_weighted-f1": 0.7413765884828194,
268
+ "eval_weighted-precision": 0.7457027535340536,
269
+ "eval_weighted-recall": 0.741304347826087,
270
+ "step": 3465
271
+ },
272
+ {
273
+ "epoch": 15.15,
274
+ "learning_rate": 3.451178451178451e-05,
275
+ "loss": 0.076,
276
+ "step": 3500
277
+ },
278
+ {
279
+ "epoch": 16.0,
280
+ "eval_accuracy": 0.7369565217391304,
281
+ "eval_loss": 1.601302981376648,
282
+ "eval_macro-f1": 0.7146795994306041,
283
+ "eval_macro-precision": 0.731834113865428,
284
+ "eval_macro-recall": 0.7046419450264447,
285
+ "eval_runtime": 3.6629,
286
+ "eval_samples_per_second": 251.17,
287
+ "eval_steps_per_second": 7.917,
288
+ "eval_weighted-f1": 0.7361455920467442,
289
+ "eval_weighted-precision": 0.7414209945981161,
290
+ "eval_weighted-recall": 0.7369565217391304,
291
+ "step": 3696
292
+ },
293
+ {
294
+ "epoch": 17.0,
295
+ "eval_accuracy": 0.7521739130434782,
296
+ "eval_loss": 1.6069446802139282,
297
+ "eval_macro-f1": 0.72890524872289,
298
+ "eval_macro-precision": 0.7550074105527117,
299
+ "eval_macro-recall": 0.7143027398745371,
300
+ "eval_runtime": 3.6875,
301
+ "eval_samples_per_second": 249.492,
302
+ "eval_steps_per_second": 7.864,
303
+ "eval_weighted-f1": 0.750626634711916,
304
+ "eval_weighted-precision": 0.7542096149679086,
305
+ "eval_weighted-recall": 0.7521739130434782,
306
+ "step": 3927
307
+ },
308
+ {
309
+ "epoch": 17.32,
310
+ "learning_rate": 3.1505531505531506e-05,
311
+ "loss": 0.0599,
312
+ "step": 4000
313
+ },
314
+ {
315
+ "epoch": 18.0,
316
+ "eval_accuracy": 0.7380434782608696,
317
+ "eval_loss": 1.6708563566207886,
318
+ "eval_macro-f1": 0.7113858264976506,
319
+ "eval_macro-precision": 0.721360196572633,
320
+ "eval_macro-recall": 0.7084908417098105,
321
+ "eval_runtime": 3.6673,
322
+ "eval_samples_per_second": 250.863,
323
+ "eval_steps_per_second": 7.908,
324
+ "eval_weighted-f1": 0.7362916788435598,
325
+ "eval_weighted-precision": 0.7410606549056714,
326
+ "eval_weighted-recall": 0.7380434782608696,
327
+ "step": 4158
328
+ },
329
+ {
330
+ "epoch": 19.0,
331
+ "eval_accuracy": 0.7304347826086957,
332
+ "eval_loss": 1.7036793231964111,
333
+ "eval_macro-f1": 0.706200706354292,
334
+ "eval_macro-precision": 0.725013063692726,
335
+ "eval_macro-recall": 0.6960154772814597,
336
+ "eval_runtime": 3.6618,
337
+ "eval_samples_per_second": 251.244,
338
+ "eval_steps_per_second": 7.92,
339
+ "eval_weighted-f1": 0.7272025757986477,
340
+ "eval_weighted-precision": 0.7299415345713408,
341
+ "eval_weighted-recall": 0.7304347826086957,
342
+ "step": 4389
343
+ },
344
+ {
345
+ "epoch": 19.48,
346
+ "learning_rate": 2.8499278499278502e-05,
347
+ "loss": 0.0527,
348
+ "step": 4500
349
+ },
350
+ {
351
+ "epoch": 20.0,
352
+ "eval_accuracy": 0.7456521739130435,
353
+ "eval_loss": 1.6668744087219238,
354
+ "eval_macro-f1": 0.7224498874030757,
355
+ "eval_macro-precision": 0.7310054057580726,
356
+ "eval_macro-recall": 0.7165591792061506,
357
+ "eval_runtime": 3.6616,
358
+ "eval_samples_per_second": 251.257,
359
+ "eval_steps_per_second": 7.92,
360
+ "eval_weighted-f1": 0.7462188604522836,
361
+ "eval_weighted-precision": 0.748502249382744,
362
+ "eval_weighted-recall": 0.7456521739130435,
363
+ "step": 4620
364
+ },
365
+ {
366
+ "epoch": 21.0,
367
+ "eval_accuracy": 0.7369565217391304,
368
+ "eval_loss": 1.714359998703003,
369
+ "eval_macro-f1": 0.7010216405318593,
370
+ "eval_macro-precision": 0.7089052674512984,
371
+ "eval_macro-recall": 0.698099718977543,
372
+ "eval_runtime": 3.6583,
373
+ "eval_samples_per_second": 251.485,
374
+ "eval_steps_per_second": 7.927,
375
+ "eval_weighted-f1": 0.7359926399774661,
376
+ "eval_weighted-precision": 0.7389162747277125,
377
+ "eval_weighted-recall": 0.7369565217391304,
378
+ "step": 4851
379
+ },
380
+ {
381
+ "epoch": 21.65,
382
+ "learning_rate": 2.5493025493025498e-05,
383
+ "loss": 0.041,
384
+ "step": 5000
385
+ },
386
+ {
387
+ "epoch": 22.0,
388
+ "eval_accuracy": 0.741304347826087,
389
+ "eval_loss": 1.8112448453903198,
390
+ "eval_macro-f1": 0.7129367870323943,
391
+ "eval_macro-precision": 0.7109062719547,
392
+ "eval_macro-recall": 0.7246584013705878,
393
+ "eval_runtime": 3.6643,
394
+ "eval_samples_per_second": 251.071,
395
+ "eval_steps_per_second": 7.914,
396
+ "eval_weighted-f1": 0.7373824826939279,
397
+ "eval_weighted-precision": 0.742304135787633,
398
+ "eval_weighted-recall": 0.741304347826087,
399
+ "step": 5082
400
+ },
401
+ {
402
+ "epoch": 23.0,
403
+ "eval_accuracy": 0.7358695652173913,
404
+ "eval_loss": 1.776945948600769,
405
+ "eval_macro-f1": 0.7100717867057288,
406
+ "eval_macro-precision": 0.7175705473737001,
407
+ "eval_macro-recall": 0.7091991070542707,
408
+ "eval_runtime": 3.6635,
409
+ "eval_samples_per_second": 251.123,
410
+ "eval_steps_per_second": 7.916,
411
+ "eval_weighted-f1": 0.7345006810309644,
412
+ "eval_weighted-precision": 0.7390098845189911,
413
+ "eval_weighted-recall": 0.7358695652173913,
414
+ "step": 5313
415
+ },
416
+ {
417
+ "epoch": 23.81,
418
+ "learning_rate": 2.2486772486772487e-05,
419
+ "loss": 0.0412,
420
+ "step": 5500
421
+ },
422
+ {
423
+ "epoch": 24.0,
424
+ "eval_accuracy": 0.75,
425
+ "eval_loss": 1.747084140777588,
426
+ "eval_macro-f1": 0.7227708998896428,
427
+ "eval_macro-precision": 0.7279946870184277,
428
+ "eval_macro-recall": 0.7269702794261114,
429
+ "eval_runtime": 3.6637,
430
+ "eval_samples_per_second": 251.112,
431
+ "eval_steps_per_second": 7.915,
432
+ "eval_weighted-f1": 0.7458470196881466,
433
+ "eval_weighted-precision": 0.7496528420569815,
434
+ "eval_weighted-recall": 0.75,
435
+ "step": 5544
436
+ },
437
+ {
438
+ "epoch": 25.0,
439
+ "eval_accuracy": 0.7532608695652174,
440
+ "eval_loss": 1.7346147298812866,
441
+ "eval_macro-f1": 0.7300620988998959,
442
+ "eval_macro-precision": 0.7254323915116926,
443
+ "eval_macro-recall": 0.7396215021682684,
444
+ "eval_runtime": 3.6654,
445
+ "eval_samples_per_second": 250.999,
446
+ "eval_steps_per_second": 7.912,
447
+ "eval_weighted-f1": 0.7504535503896388,
448
+ "eval_weighted-precision": 0.7526891843637886,
449
+ "eval_weighted-recall": 0.7532608695652174,
450
+ "step": 5775
451
+ },
452
+ {
453
+ "epoch": 25.97,
454
+ "learning_rate": 1.9480519480519483e-05,
455
+ "loss": 0.0326,
456
+ "step": 6000
457
+ },
458
+ {
459
+ "epoch": 26.0,
460
+ "eval_accuracy": 0.7456521739130435,
461
+ "eval_loss": 1.773017168045044,
462
+ "eval_macro-f1": 0.7210970682285488,
463
+ "eval_macro-precision": 0.7283775906058515,
464
+ "eval_macro-recall": 0.719270945602435,
465
+ "eval_runtime": 3.6601,
466
+ "eval_samples_per_second": 251.357,
467
+ "eval_steps_per_second": 7.923,
468
+ "eval_weighted-f1": 0.7454697580769286,
469
+ "eval_weighted-precision": 0.7489157944224945,
470
+ "eval_weighted-recall": 0.7456521739130435,
471
+ "step": 6006
472
+ },
473
+ {
474
+ "epoch": 27.0,
475
+ "eval_accuracy": 0.741304347826087,
476
+ "eval_loss": 1.7493728399276733,
477
+ "eval_macro-f1": 0.7025880602354205,
478
+ "eval_macro-precision": 0.7130648727750518,
479
+ "eval_macro-recall": 0.6993768549946884,
480
+ "eval_runtime": 3.6688,
481
+ "eval_samples_per_second": 250.762,
482
+ "eval_steps_per_second": 7.904,
483
+ "eval_weighted-f1": 0.7414220307733658,
484
+ "eval_weighted-precision": 0.7452401075563395,
485
+ "eval_weighted-recall": 0.741304347826087,
486
+ "step": 6237
487
+ },
488
+ {
489
+ "epoch": 28.0,
490
+ "eval_accuracy": 0.7380434782608696,
491
+ "eval_loss": 1.7450823783874512,
492
+ "eval_macro-f1": 0.7026259912227975,
493
+ "eval_macro-precision": 0.7214901347374018,
494
+ "eval_macro-recall": 0.6911837077117138,
495
+ "eval_runtime": 3.7325,
496
+ "eval_samples_per_second": 246.486,
497
+ "eval_steps_per_second": 7.77,
498
+ "eval_weighted-f1": 0.736513945546438,
499
+ "eval_weighted-precision": 0.7385934880642558,
500
+ "eval_weighted-recall": 0.7380434782608696,
501
+ "step": 6468
502
+ },
503
+ {
504
+ "epoch": 28.14,
505
+ "learning_rate": 1.6474266474266476e-05,
506
+ "loss": 0.0257,
507
+ "step": 6500
508
+ },
509
+ {
510
+ "epoch": 29.0,
511
+ "eval_accuracy": 0.7402173913043478,
512
+ "eval_loss": 1.7633217573165894,
513
+ "eval_macro-f1": 0.7134642108516674,
514
+ "eval_macro-precision": 0.7190433251313625,
515
+ "eval_macro-recall": 0.710258338303699,
516
+ "eval_runtime": 3.6817,
517
+ "eval_samples_per_second": 249.884,
518
+ "eval_steps_per_second": 7.877,
519
+ "eval_weighted-f1": 0.7398481546483754,
520
+ "eval_weighted-precision": 0.7415064106206414,
521
+ "eval_weighted-recall": 0.7402173913043478,
522
+ "step": 6699
523
+ },
524
+ {
525
+ "epoch": 30.0,
526
+ "eval_accuracy": 0.7347826086956522,
527
+ "eval_loss": 1.820716381072998,
528
+ "eval_macro-f1": 0.7078683429969447,
529
+ "eval_macro-precision": 0.7372603591067731,
530
+ "eval_macro-recall": 0.6893626457978533,
531
+ "eval_runtime": 3.7068,
532
+ "eval_samples_per_second": 248.19,
533
+ "eval_steps_per_second": 7.823,
534
+ "eval_weighted-f1": 0.7322350095276792,
535
+ "eval_weighted-precision": 0.7372777947320607,
536
+ "eval_weighted-recall": 0.7347826086956522,
537
+ "step": 6930
538
+ },
539
+ {
540
+ "epoch": 30.3,
541
+ "learning_rate": 1.3468013468013468e-05,
542
+ "loss": 0.0262,
543
+ "step": 7000
544
+ },
545
+ {
546
+ "epoch": 31.0,
547
+ "eval_accuracy": 0.7510869565217392,
548
+ "eval_loss": 1.76492178440094,
549
+ "eval_macro-f1": 0.731348188671936,
550
+ "eval_macro-precision": 0.7382460346691814,
551
+ "eval_macro-recall": 0.7298883924645497,
552
+ "eval_runtime": 3.6813,
553
+ "eval_samples_per_second": 249.913,
554
+ "eval_steps_per_second": 7.878,
555
+ "eval_weighted-f1": 0.7508384750022834,
556
+ "eval_weighted-precision": 0.7539289140986796,
557
+ "eval_weighted-recall": 0.7510869565217392,
558
+ "step": 7161
559
+ },
560
+ {
561
+ "epoch": 32.0,
562
+ "eval_accuracy": 0.7358695652173913,
563
+ "eval_loss": 1.8048666715621948,
564
+ "eval_macro-f1": 0.7135206623010025,
565
+ "eval_macro-precision": 0.7206599735879954,
566
+ "eval_macro-recall": 0.713146029256634,
567
+ "eval_runtime": 3.6665,
568
+ "eval_samples_per_second": 250.92,
569
+ "eval_steps_per_second": 7.909,
570
+ "eval_weighted-f1": 0.7342617186640648,
571
+ "eval_weighted-precision": 0.7381810820989364,
572
+ "eval_weighted-recall": 0.7358695652173913,
573
+ "step": 7392
574
+ },
575
+ {
576
+ "epoch": 32.47,
577
+ "learning_rate": 1.0461760461760463e-05,
578
+ "loss": 0.0195,
579
+ "step": 7500
580
+ },
581
+ {
582
+ "epoch": 33.0,
583
+ "eval_accuracy": 0.7434782608695653,
584
+ "eval_loss": 1.7627431154251099,
585
+ "eval_macro-f1": 0.72002488158224,
586
+ "eval_macro-precision": 0.727980412915839,
587
+ "eval_macro-recall": 0.7171781195809488,
588
+ "eval_runtime": 3.6606,
589
+ "eval_samples_per_second": 251.327,
590
+ "eval_steps_per_second": 7.922,
591
+ "eval_weighted-f1": 0.7428966222335778,
592
+ "eval_weighted-precision": 0.7455932383430586,
593
+ "eval_weighted-recall": 0.7434782608695653,
594
+ "step": 7623
595
+ },
596
+ {
597
+ "epoch": 34.0,
598
+ "eval_accuracy": 0.7445652173913043,
599
+ "eval_loss": 1.8427157402038574,
600
+ "eval_macro-f1": 0.7238120517774087,
601
+ "eval_macro-precision": 0.7310564585025658,
602
+ "eval_macro-recall": 0.7232388744393482,
603
+ "eval_runtime": 3.6585,
604
+ "eval_samples_per_second": 251.468,
605
+ "eval_steps_per_second": 7.927,
606
+ "eval_weighted-f1": 0.7436940578232905,
607
+ "eval_weighted-precision": 0.7475437346184277,
608
+ "eval_weighted-recall": 0.7445652173913043,
609
+ "step": 7854
610
+ },
611
+ {
612
+ "epoch": 34.63,
613
+ "learning_rate": 7.455507455507456e-06,
614
+ "loss": 0.0188,
615
+ "step": 8000
616
+ },
617
+ {
618
+ "epoch": 35.0,
619
+ "eval_accuracy": 0.7445652173913043,
620
+ "eval_loss": 1.82598078250885,
621
+ "eval_macro-f1": 0.7226558273088113,
622
+ "eval_macro-precision": 0.7280808646863082,
623
+ "eval_macro-recall": 0.721750060320641,
624
+ "eval_runtime": 3.9179,
625
+ "eval_samples_per_second": 234.822,
626
+ "eval_steps_per_second": 7.402,
627
+ "eval_weighted-f1": 0.7433397528033767,
628
+ "eval_weighted-precision": 0.7455193887868321,
629
+ "eval_weighted-recall": 0.7445652173913043,
630
+ "step": 8085
631
+ },
632
+ {
633
+ "epoch": 36.0,
634
+ "eval_accuracy": 0.7521739130434782,
635
+ "eval_loss": 1.8167239427566528,
636
+ "eval_macro-f1": 0.7311172141769898,
637
+ "eval_macro-precision": 0.7387132761496166,
638
+ "eval_macro-recall": 0.7260195259385496,
639
+ "eval_runtime": 3.6724,
640
+ "eval_samples_per_second": 250.519,
641
+ "eval_steps_per_second": 7.897,
642
+ "eval_weighted-f1": 0.7512808559429369,
643
+ "eval_weighted-precision": 0.7518176004978223,
644
+ "eval_weighted-recall": 0.7521739130434782,
645
+ "step": 8316
646
+ },
647
+ {
648
+ "epoch": 36.8,
649
+ "learning_rate": 4.44925444925445e-06,
650
+ "loss": 0.0176,
651
+ "step": 8500
652
+ },
653
+ {
654
+ "epoch": 37.0,
655
+ "eval_accuracy": 0.75,
656
+ "eval_loss": 1.8308371305465698,
657
+ "eval_macro-f1": 0.7308495034985188,
658
+ "eval_macro-precision": 0.7403845771754249,
659
+ "eval_macro-recall": 0.7242967832014863,
660
+ "eval_runtime": 3.6615,
661
+ "eval_samples_per_second": 251.261,
662
+ "eval_steps_per_second": 7.92,
663
+ "eval_weighted-f1": 0.7478384491335044,
664
+ "eval_weighted-precision": 0.7485903968276678,
665
+ "eval_weighted-recall": 0.75,
666
+ "step": 8547
667
+ },
668
+ {
669
+ "epoch": 38.0,
670
+ "eval_accuracy": 0.7467391304347826,
671
+ "eval_loss": 1.807198405265808,
672
+ "eval_macro-f1": 0.7291086447475449,
673
+ "eval_macro-precision": 0.7329302540393156,
674
+ "eval_macro-recall": 0.7271188451475986,
675
+ "eval_runtime": 3.6625,
676
+ "eval_samples_per_second": 251.195,
677
+ "eval_steps_per_second": 7.918,
678
+ "eval_weighted-f1": 0.7458071950555136,
679
+ "eval_weighted-precision": 0.7463050676859623,
680
+ "eval_weighted-recall": 0.7467391304347826,
681
+ "step": 8778
682
+ },
683
+ {
684
+ "epoch": 38.96,
685
+ "learning_rate": 1.4430014430014432e-06,
686
+ "loss": 0.0164,
687
+ "step": 9000
688
+ },
689
+ {
690
+ "epoch": 39.0,
691
+ "eval_accuracy": 0.7456521739130435,
692
+ "eval_loss": 1.8073784112930298,
693
+ "eval_macro-f1": 0.7256401482067129,
694
+ "eval_macro-precision": 0.733018922323657,
695
+ "eval_macro-recall": 0.7204178880954032,
696
+ "eval_runtime": 3.673,
697
+ "eval_samples_per_second": 250.478,
698
+ "eval_steps_per_second": 7.896,
699
+ "eval_weighted-f1": 0.7445134806621682,
700
+ "eval_weighted-precision": 0.7450872143093246,
701
+ "eval_weighted-recall": 0.7456521739130435,
702
+ "step": 9009
703
+ },
704
+ {
705
+ "epoch": 40.0,
706
+ "eval_accuracy": 0.741304347826087,
707
+ "eval_loss": 1.818052887916565,
708
+ "eval_macro-f1": 0.7215539592969167,
709
+ "eval_macro-precision": 0.7317731195834322,
710
+ "eval_macro-recall": 0.7133028256075337,
711
+ "eval_runtime": 3.6643,
712
+ "eval_samples_per_second": 251.069,
713
+ "eval_steps_per_second": 7.914,
714
+ "eval_weighted-f1": 0.7401045633293094,
715
+ "eval_weighted-precision": 0.740661997113788,
716
+ "eval_weighted-recall": 0.741304347826087,
717
+ "step": 9240
718
+ }
719
+ ],
720
+ "logging_steps": 500,
721
+ "max_steps": 9240,
722
+ "num_input_tokens_seen": 0,
723
+ "num_train_epochs": 40,
724
+ "save_steps": 500,
725
+ "total_flos": 3.902655553929216e+16,
726
+ "train_batch_size": 32,
727
+ "trial_name": null,
728
+ "trial_params": null
729
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d25ebe045df09d09a09adfed7a087270b003ff43849571039aab6046873db6f
3
+ size 4335