OskarMosberg commited on
Commit
915b108
·
verified ·
1 Parent(s): 8cae6b2

Training in progress, step 500

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. model.safetensors +1 -1
  2. run-4/checkpoint-1000/config.json +330 -0
  3. run-4/checkpoint-1000/model.safetensors +3 -0
  4. run-4/checkpoint-1000/optimizer.pt +3 -0
  5. run-4/checkpoint-1000/rng_state.pth +3 -0
  6. run-4/checkpoint-1000/scheduler.pt +3 -0
  7. run-4/checkpoint-1000/special_tokens_map.json +7 -0
  8. run-4/checkpoint-1000/tokenizer.json +0 -0
  9. run-4/checkpoint-1000/tokenizer_config.json +56 -0
  10. run-4/checkpoint-1000/trainer_state.json +79 -0
  11. run-4/checkpoint-1000/training_args.bin +3 -0
  12. run-4/checkpoint-1000/vocab.txt +0 -0
  13. run-4/checkpoint-1500/config.json +330 -0
  14. run-4/checkpoint-1500/model.safetensors +3 -0
  15. run-4/checkpoint-1500/optimizer.pt +3 -0
  16. run-4/checkpoint-1500/rng_state.pth +3 -0
  17. run-4/checkpoint-1500/scheduler.pt +3 -0
  18. run-4/checkpoint-1500/special_tokens_map.json +7 -0
  19. run-4/checkpoint-1500/tokenizer.json +0 -0
  20. run-4/checkpoint-1500/tokenizer_config.json +56 -0
  21. run-4/checkpoint-1500/trainer_state.json +95 -0
  22. run-4/checkpoint-1500/training_args.bin +3 -0
  23. run-4/checkpoint-1500/vocab.txt +0 -0
  24. run-4/checkpoint-2000/config.json +330 -0
  25. run-4/checkpoint-2000/model.safetensors +3 -0
  26. run-4/checkpoint-2000/optimizer.pt +3 -0
  27. run-4/checkpoint-2000/rng_state.pth +3 -0
  28. run-4/checkpoint-2000/scheduler.pt +3 -0
  29. run-4/checkpoint-2000/special_tokens_map.json +7 -0
  30. run-4/checkpoint-2000/tokenizer.json +0 -0
  31. run-4/checkpoint-2000/tokenizer_config.json +56 -0
  32. run-4/checkpoint-2000/trainer_state.json +120 -0
  33. run-4/checkpoint-2000/training_args.bin +3 -0
  34. run-4/checkpoint-2000/vocab.txt +0 -0
  35. run-4/checkpoint-2500/config.json +330 -0
  36. run-4/checkpoint-2500/model.safetensors +3 -0
  37. run-4/checkpoint-2500/optimizer.pt +3 -0
  38. run-4/checkpoint-2500/rng_state.pth +3 -0
  39. run-4/checkpoint-2500/scheduler.pt +3 -0
  40. run-4/checkpoint-2500/special_tokens_map.json +7 -0
  41. run-4/checkpoint-2500/tokenizer.json +0 -0
  42. run-4/checkpoint-2500/tokenizer_config.json +56 -0
  43. run-4/checkpoint-2500/trainer_state.json +136 -0
  44. run-4/checkpoint-2500/training_args.bin +3 -0
  45. run-4/checkpoint-2500/vocab.txt +0 -0
  46. run-4/checkpoint-2862/config.json +330 -0
  47. run-4/checkpoint-2862/model.safetensors +3 -0
  48. run-4/checkpoint-2862/optimizer.pt +3 -0
  49. run-4/checkpoint-2862/rng_state.pth +3 -0
  50. run-4/checkpoint-2862/scheduler.pt +3 -0
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c9e6034e2ed68de68978bd8c39776451d696e721aba8d51e615c0df4d829cded
3
  size 268290900
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58e009548f1ab29f6c08a893fefdf744acf38310979eddaa144c9d2071da6eb2
3
  size 268290900
run-4/checkpoint-1000/config.json ADDED
@@ -0,0 +1,330 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "gelu",
3
+ "architectures": [
4
+ "DistilBertForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.1,
7
+ "dim": 768,
8
+ "dropout": 0.1,
9
+ "dtype": "float32",
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "LABEL_0",
13
+ "1": "LABEL_1",
14
+ "2": "LABEL_2",
15
+ "3": "LABEL_3",
16
+ "4": "LABEL_4",
17
+ "5": "LABEL_5",
18
+ "6": "LABEL_6",
19
+ "7": "LABEL_7",
20
+ "8": "LABEL_8",
21
+ "9": "LABEL_9",
22
+ "10": "LABEL_10",
23
+ "11": "LABEL_11",
24
+ "12": "LABEL_12",
25
+ "13": "LABEL_13",
26
+ "14": "LABEL_14",
27
+ "15": "LABEL_15",
28
+ "16": "LABEL_16",
29
+ "17": "LABEL_17",
30
+ "18": "LABEL_18",
31
+ "19": "LABEL_19",
32
+ "20": "LABEL_20",
33
+ "21": "LABEL_21",
34
+ "22": "LABEL_22",
35
+ "23": "LABEL_23",
36
+ "24": "LABEL_24",
37
+ "25": "LABEL_25",
38
+ "26": "LABEL_26",
39
+ "27": "LABEL_27",
40
+ "28": "LABEL_28",
41
+ "29": "LABEL_29",
42
+ "30": "LABEL_30",
43
+ "31": "LABEL_31",
44
+ "32": "LABEL_32",
45
+ "33": "LABEL_33",
46
+ "34": "LABEL_34",
47
+ "35": "LABEL_35",
48
+ "36": "LABEL_36",
49
+ "37": "LABEL_37",
50
+ "38": "LABEL_38",
51
+ "39": "LABEL_39",
52
+ "40": "LABEL_40",
53
+ "41": "LABEL_41",
54
+ "42": "LABEL_42",
55
+ "43": "LABEL_43",
56
+ "44": "LABEL_44",
57
+ "45": "LABEL_45",
58
+ "46": "LABEL_46",
59
+ "47": "LABEL_47",
60
+ "48": "LABEL_48",
61
+ "49": "LABEL_49",
62
+ "50": "LABEL_50",
63
+ "51": "LABEL_51",
64
+ "52": "LABEL_52",
65
+ "53": "LABEL_53",
66
+ "54": "LABEL_54",
67
+ "55": "LABEL_55",
68
+ "56": "LABEL_56",
69
+ "57": "LABEL_57",
70
+ "58": "LABEL_58",
71
+ "59": "LABEL_59",
72
+ "60": "LABEL_60",
73
+ "61": "LABEL_61",
74
+ "62": "LABEL_62",
75
+ "63": "LABEL_63",
76
+ "64": "LABEL_64",
77
+ "65": "LABEL_65",
78
+ "66": "LABEL_66",
79
+ "67": "LABEL_67",
80
+ "68": "LABEL_68",
81
+ "69": "LABEL_69",
82
+ "70": "LABEL_70",
83
+ "71": "LABEL_71",
84
+ "72": "LABEL_72",
85
+ "73": "LABEL_73",
86
+ "74": "LABEL_74",
87
+ "75": "LABEL_75",
88
+ "76": "LABEL_76",
89
+ "77": "LABEL_77",
90
+ "78": "LABEL_78",
91
+ "79": "LABEL_79",
92
+ "80": "LABEL_80",
93
+ "81": "LABEL_81",
94
+ "82": "LABEL_82",
95
+ "83": "LABEL_83",
96
+ "84": "LABEL_84",
97
+ "85": "LABEL_85",
98
+ "86": "LABEL_86",
99
+ "87": "LABEL_87",
100
+ "88": "LABEL_88",
101
+ "89": "LABEL_89",
102
+ "90": "LABEL_90",
103
+ "91": "LABEL_91",
104
+ "92": "LABEL_92",
105
+ "93": "LABEL_93",
106
+ "94": "LABEL_94",
107
+ "95": "LABEL_95",
108
+ "96": "LABEL_96",
109
+ "97": "LABEL_97",
110
+ "98": "LABEL_98",
111
+ "99": "LABEL_99",
112
+ "100": "LABEL_100",
113
+ "101": "LABEL_101",
114
+ "102": "LABEL_102",
115
+ "103": "LABEL_103",
116
+ "104": "LABEL_104",
117
+ "105": "LABEL_105",
118
+ "106": "LABEL_106",
119
+ "107": "LABEL_107",
120
+ "108": "LABEL_108",
121
+ "109": "LABEL_109",
122
+ "110": "LABEL_110",
123
+ "111": "LABEL_111",
124
+ "112": "LABEL_112",
125
+ "113": "LABEL_113",
126
+ "114": "LABEL_114",
127
+ "115": "LABEL_115",
128
+ "116": "LABEL_116",
129
+ "117": "LABEL_117",
130
+ "118": "LABEL_118",
131
+ "119": "LABEL_119",
132
+ "120": "LABEL_120",
133
+ "121": "LABEL_121",
134
+ "122": "LABEL_122",
135
+ "123": "LABEL_123",
136
+ "124": "LABEL_124",
137
+ "125": "LABEL_125",
138
+ "126": "LABEL_126",
139
+ "127": "LABEL_127",
140
+ "128": "LABEL_128",
141
+ "129": "LABEL_129",
142
+ "130": "LABEL_130",
143
+ "131": "LABEL_131",
144
+ "132": "LABEL_132",
145
+ "133": "LABEL_133",
146
+ "134": "LABEL_134",
147
+ "135": "LABEL_135",
148
+ "136": "LABEL_136",
149
+ "137": "LABEL_137",
150
+ "138": "LABEL_138",
151
+ "139": "LABEL_139",
152
+ "140": "LABEL_140",
153
+ "141": "LABEL_141",
154
+ "142": "LABEL_142",
155
+ "143": "LABEL_143",
156
+ "144": "LABEL_144",
157
+ "145": "LABEL_145",
158
+ "146": "LABEL_146",
159
+ "147": "LABEL_147",
160
+ "148": "LABEL_148",
161
+ "149": "LABEL_149",
162
+ "150": "LABEL_150"
163
+ },
164
+ "initializer_range": 0.02,
165
+ "label2id": {
166
+ "LABEL_0": 0,
167
+ "LABEL_1": 1,
168
+ "LABEL_10": 10,
169
+ "LABEL_100": 100,
170
+ "LABEL_101": 101,
171
+ "LABEL_102": 102,
172
+ "LABEL_103": 103,
173
+ "LABEL_104": 104,
174
+ "LABEL_105": 105,
175
+ "LABEL_106": 106,
176
+ "LABEL_107": 107,
177
+ "LABEL_108": 108,
178
+ "LABEL_109": 109,
179
+ "LABEL_11": 11,
180
+ "LABEL_110": 110,
181
+ "LABEL_111": 111,
182
+ "LABEL_112": 112,
183
+ "LABEL_113": 113,
184
+ "LABEL_114": 114,
185
+ "LABEL_115": 115,
186
+ "LABEL_116": 116,
187
+ "LABEL_117": 117,
188
+ "LABEL_118": 118,
189
+ "LABEL_119": 119,
190
+ "LABEL_12": 12,
191
+ "LABEL_120": 120,
192
+ "LABEL_121": 121,
193
+ "LABEL_122": 122,
194
+ "LABEL_123": 123,
195
+ "LABEL_124": 124,
196
+ "LABEL_125": 125,
197
+ "LABEL_126": 126,
198
+ "LABEL_127": 127,
199
+ "LABEL_128": 128,
200
+ "LABEL_129": 129,
201
+ "LABEL_13": 13,
202
+ "LABEL_130": 130,
203
+ "LABEL_131": 131,
204
+ "LABEL_132": 132,
205
+ "LABEL_133": 133,
206
+ "LABEL_134": 134,
207
+ "LABEL_135": 135,
208
+ "LABEL_136": 136,
209
+ "LABEL_137": 137,
210
+ "LABEL_138": 138,
211
+ "LABEL_139": 139,
212
+ "LABEL_14": 14,
213
+ "LABEL_140": 140,
214
+ "LABEL_141": 141,
215
+ "LABEL_142": 142,
216
+ "LABEL_143": 143,
217
+ "LABEL_144": 144,
218
+ "LABEL_145": 145,
219
+ "LABEL_146": 146,
220
+ "LABEL_147": 147,
221
+ "LABEL_148": 148,
222
+ "LABEL_149": 149,
223
+ "LABEL_15": 15,
224
+ "LABEL_150": 150,
225
+ "LABEL_16": 16,
226
+ "LABEL_17": 17,
227
+ "LABEL_18": 18,
228
+ "LABEL_19": 19,
229
+ "LABEL_2": 2,
230
+ "LABEL_20": 20,
231
+ "LABEL_21": 21,
232
+ "LABEL_22": 22,
233
+ "LABEL_23": 23,
234
+ "LABEL_24": 24,
235
+ "LABEL_25": 25,
236
+ "LABEL_26": 26,
237
+ "LABEL_27": 27,
238
+ "LABEL_28": 28,
239
+ "LABEL_29": 29,
240
+ "LABEL_3": 3,
241
+ "LABEL_30": 30,
242
+ "LABEL_31": 31,
243
+ "LABEL_32": 32,
244
+ "LABEL_33": 33,
245
+ "LABEL_34": 34,
246
+ "LABEL_35": 35,
247
+ "LABEL_36": 36,
248
+ "LABEL_37": 37,
249
+ "LABEL_38": 38,
250
+ "LABEL_39": 39,
251
+ "LABEL_4": 4,
252
+ "LABEL_40": 40,
253
+ "LABEL_41": 41,
254
+ "LABEL_42": 42,
255
+ "LABEL_43": 43,
256
+ "LABEL_44": 44,
257
+ "LABEL_45": 45,
258
+ "LABEL_46": 46,
259
+ "LABEL_47": 47,
260
+ "LABEL_48": 48,
261
+ "LABEL_49": 49,
262
+ "LABEL_5": 5,
263
+ "LABEL_50": 50,
264
+ "LABEL_51": 51,
265
+ "LABEL_52": 52,
266
+ "LABEL_53": 53,
267
+ "LABEL_54": 54,
268
+ "LABEL_55": 55,
269
+ "LABEL_56": 56,
270
+ "LABEL_57": 57,
271
+ "LABEL_58": 58,
272
+ "LABEL_59": 59,
273
+ "LABEL_6": 6,
274
+ "LABEL_60": 60,
275
+ "LABEL_61": 61,
276
+ "LABEL_62": 62,
277
+ "LABEL_63": 63,
278
+ "LABEL_64": 64,
279
+ "LABEL_65": 65,
280
+ "LABEL_66": 66,
281
+ "LABEL_67": 67,
282
+ "LABEL_68": 68,
283
+ "LABEL_69": 69,
284
+ "LABEL_7": 7,
285
+ "LABEL_70": 70,
286
+ "LABEL_71": 71,
287
+ "LABEL_72": 72,
288
+ "LABEL_73": 73,
289
+ "LABEL_74": 74,
290
+ "LABEL_75": 75,
291
+ "LABEL_76": 76,
292
+ "LABEL_77": 77,
293
+ "LABEL_78": 78,
294
+ "LABEL_79": 79,
295
+ "LABEL_8": 8,
296
+ "LABEL_80": 80,
297
+ "LABEL_81": 81,
298
+ "LABEL_82": 82,
299
+ "LABEL_83": 83,
300
+ "LABEL_84": 84,
301
+ "LABEL_85": 85,
302
+ "LABEL_86": 86,
303
+ "LABEL_87": 87,
304
+ "LABEL_88": 88,
305
+ "LABEL_89": 89,
306
+ "LABEL_9": 9,
307
+ "LABEL_90": 90,
308
+ "LABEL_91": 91,
309
+ "LABEL_92": 92,
310
+ "LABEL_93": 93,
311
+ "LABEL_94": 94,
312
+ "LABEL_95": 95,
313
+ "LABEL_96": 96,
314
+ "LABEL_97": 97,
315
+ "LABEL_98": 98,
316
+ "LABEL_99": 99
317
+ },
318
+ "max_position_embeddings": 512,
319
+ "model_type": "distilbert",
320
+ "n_heads": 12,
321
+ "n_layers": 6,
322
+ "pad_token_id": 0,
323
+ "problem_type": "single_label_classification",
324
+ "qa_dropout": 0.1,
325
+ "seq_classif_dropout": 0.2,
326
+ "sinusoidal_pos_embds": false,
327
+ "tie_weights_": true,
328
+ "transformers_version": "4.57.0",
329
+ "vocab_size": 30522
330
+ }
run-4/checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:293068e257753089809ceab96d1e355def1286161a40548a31b6affe14be12ea
3
+ size 268290900
run-4/checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a0fbc861c2d01c16248eb96c9d3459a19ceff038563239a0e04d4723deed91f
3
+ size 536645835
run-4/checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0838252fc80b9ba9f2d9bc9f278258ad33973ffebd2b8e8f779e46819e8f8ff
3
+ size 14645
run-4/checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ed93fe6dda194b0c86a69e163cf484e0f83e3e30ca32af1674a99843a717a36
3
+ size 1465
run-4/checkpoint-1000/special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
run-4/checkpoint-1000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-4/checkpoint-1000/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "pad_token": "[PAD]",
51
+ "sep_token": "[SEP]",
52
+ "strip_accents": null,
53
+ "tokenize_chinese_chars": true,
54
+ "tokenizer_class": "DistilBertTokenizer",
55
+ "unk_token": "[UNK]"
56
+ }
run-4/checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,79 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.1446540880503147,
6
+ "eval_steps": 500,
7
+ "global_step": 1000,
8
+ "is_hyper_param_search": true,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "eval_accuracy": 0.6845161290322581,
15
+ "eval_loss": 0.10785406082868576,
16
+ "eval_runtime": 1.0937,
17
+ "eval_samples_per_second": 2834.299,
18
+ "eval_steps_per_second": 59.429,
19
+ "step": 318
20
+ },
21
+ {
22
+ "epoch": 1.5723270440251573,
23
+ "grad_norm": 0.4275546073913574,
24
+ "learning_rate": 1.6512928022361987e-05,
25
+ "loss": 0.2193,
26
+ "step": 500
27
+ },
28
+ {
29
+ "epoch": 2.0,
30
+ "eval_accuracy": 0.8438709677419355,
31
+ "eval_loss": 0.03479825332760811,
32
+ "eval_runtime": 1.097,
33
+ "eval_samples_per_second": 2825.922,
34
+ "eval_steps_per_second": 59.253,
35
+ "step": 636
36
+ },
37
+ {
38
+ "epoch": 3.0,
39
+ "eval_accuracy": 0.8703225806451613,
40
+ "eval_loss": 0.01689928211271763,
41
+ "eval_runtime": 1.0917,
42
+ "eval_samples_per_second": 2839.696,
43
+ "eval_steps_per_second": 59.542,
44
+ "step": 954
45
+ },
46
+ {
47
+ "epoch": 3.1446540880503147,
48
+ "grad_norm": 0.22454383969306946,
49
+ "learning_rate": 1.3018867924528303e-05,
50
+ "loss": 0.0599,
51
+ "step": 1000
52
+ }
53
+ ],
54
+ "logging_steps": 500,
55
+ "max_steps": 2862,
56
+ "num_input_tokens_seen": 0,
57
+ "num_train_epochs": 9,
58
+ "save_steps": 500,
59
+ "stateful_callbacks": {
60
+ "TrainerControl": {
61
+ "args": {
62
+ "should_epoch_stop": false,
63
+ "should_evaluate": false,
64
+ "should_log": false,
65
+ "should_save": true,
66
+ "should_training_stop": false
67
+ },
68
+ "attributes": {}
69
+ }
70
+ },
71
+ "total_flos": 260941334653608.0,
72
+ "train_batch_size": 48,
73
+ "trial_name": null,
74
+ "trial_params": {
75
+ "alpha": 0.9763664799887318,
76
+ "num_train_epochs": 9,
77
+ "temperature": 10
78
+ }
79
+ }
run-4/checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7cd8292610cea7a04a70d5b98c84050970fba3dabfd3ac7ee8ac97541992cb4
3
+ size 5841
run-4/checkpoint-1000/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
run-4/checkpoint-1500/config.json ADDED
@@ -0,0 +1,330 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "gelu",
3
+ "architectures": [
4
+ "DistilBertForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.1,
7
+ "dim": 768,
8
+ "dropout": 0.1,
9
+ "dtype": "float32",
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "LABEL_0",
13
+ "1": "LABEL_1",
14
+ "2": "LABEL_2",
15
+ "3": "LABEL_3",
16
+ "4": "LABEL_4",
17
+ "5": "LABEL_5",
18
+ "6": "LABEL_6",
19
+ "7": "LABEL_7",
20
+ "8": "LABEL_8",
21
+ "9": "LABEL_9",
22
+ "10": "LABEL_10",
23
+ "11": "LABEL_11",
24
+ "12": "LABEL_12",
25
+ "13": "LABEL_13",
26
+ "14": "LABEL_14",
27
+ "15": "LABEL_15",
28
+ "16": "LABEL_16",
29
+ "17": "LABEL_17",
30
+ "18": "LABEL_18",
31
+ "19": "LABEL_19",
32
+ "20": "LABEL_20",
33
+ "21": "LABEL_21",
34
+ "22": "LABEL_22",
35
+ "23": "LABEL_23",
36
+ "24": "LABEL_24",
37
+ "25": "LABEL_25",
38
+ "26": "LABEL_26",
39
+ "27": "LABEL_27",
40
+ "28": "LABEL_28",
41
+ "29": "LABEL_29",
42
+ "30": "LABEL_30",
43
+ "31": "LABEL_31",
44
+ "32": "LABEL_32",
45
+ "33": "LABEL_33",
46
+ "34": "LABEL_34",
47
+ "35": "LABEL_35",
48
+ "36": "LABEL_36",
49
+ "37": "LABEL_37",
50
+ "38": "LABEL_38",
51
+ "39": "LABEL_39",
52
+ "40": "LABEL_40",
53
+ "41": "LABEL_41",
54
+ "42": "LABEL_42",
55
+ "43": "LABEL_43",
56
+ "44": "LABEL_44",
57
+ "45": "LABEL_45",
58
+ "46": "LABEL_46",
59
+ "47": "LABEL_47",
60
+ "48": "LABEL_48",
61
+ "49": "LABEL_49",
62
+ "50": "LABEL_50",
63
+ "51": "LABEL_51",
64
+ "52": "LABEL_52",
65
+ "53": "LABEL_53",
66
+ "54": "LABEL_54",
67
+ "55": "LABEL_55",
68
+ "56": "LABEL_56",
69
+ "57": "LABEL_57",
70
+ "58": "LABEL_58",
71
+ "59": "LABEL_59",
72
+ "60": "LABEL_60",
73
+ "61": "LABEL_61",
74
+ "62": "LABEL_62",
75
+ "63": "LABEL_63",
76
+ "64": "LABEL_64",
77
+ "65": "LABEL_65",
78
+ "66": "LABEL_66",
79
+ "67": "LABEL_67",
80
+ "68": "LABEL_68",
81
+ "69": "LABEL_69",
82
+ "70": "LABEL_70",
83
+ "71": "LABEL_71",
84
+ "72": "LABEL_72",
85
+ "73": "LABEL_73",
86
+ "74": "LABEL_74",
87
+ "75": "LABEL_75",
88
+ "76": "LABEL_76",
89
+ "77": "LABEL_77",
90
+ "78": "LABEL_78",
91
+ "79": "LABEL_79",
92
+ "80": "LABEL_80",
93
+ "81": "LABEL_81",
94
+ "82": "LABEL_82",
95
+ "83": "LABEL_83",
96
+ "84": "LABEL_84",
97
+ "85": "LABEL_85",
98
+ "86": "LABEL_86",
99
+ "87": "LABEL_87",
100
+ "88": "LABEL_88",
101
+ "89": "LABEL_89",
102
+ "90": "LABEL_90",
103
+ "91": "LABEL_91",
104
+ "92": "LABEL_92",
105
+ "93": "LABEL_93",
106
+ "94": "LABEL_94",
107
+ "95": "LABEL_95",
108
+ "96": "LABEL_96",
109
+ "97": "LABEL_97",
110
+ "98": "LABEL_98",
111
+ "99": "LABEL_99",
112
+ "100": "LABEL_100",
113
+ "101": "LABEL_101",
114
+ "102": "LABEL_102",
115
+ "103": "LABEL_103",
116
+ "104": "LABEL_104",
117
+ "105": "LABEL_105",
118
+ "106": "LABEL_106",
119
+ "107": "LABEL_107",
120
+ "108": "LABEL_108",
121
+ "109": "LABEL_109",
122
+ "110": "LABEL_110",
123
+ "111": "LABEL_111",
124
+ "112": "LABEL_112",
125
+ "113": "LABEL_113",
126
+ "114": "LABEL_114",
127
+ "115": "LABEL_115",
128
+ "116": "LABEL_116",
129
+ "117": "LABEL_117",
130
+ "118": "LABEL_118",
131
+ "119": "LABEL_119",
132
+ "120": "LABEL_120",
133
+ "121": "LABEL_121",
134
+ "122": "LABEL_122",
135
+ "123": "LABEL_123",
136
+ "124": "LABEL_124",
137
+ "125": "LABEL_125",
138
+ "126": "LABEL_126",
139
+ "127": "LABEL_127",
140
+ "128": "LABEL_128",
141
+ "129": "LABEL_129",
142
+ "130": "LABEL_130",
143
+ "131": "LABEL_131",
144
+ "132": "LABEL_132",
145
+ "133": "LABEL_133",
146
+ "134": "LABEL_134",
147
+ "135": "LABEL_135",
148
+ "136": "LABEL_136",
149
+ "137": "LABEL_137",
150
+ "138": "LABEL_138",
151
+ "139": "LABEL_139",
152
+ "140": "LABEL_140",
153
+ "141": "LABEL_141",
154
+ "142": "LABEL_142",
155
+ "143": "LABEL_143",
156
+ "144": "LABEL_144",
157
+ "145": "LABEL_145",
158
+ "146": "LABEL_146",
159
+ "147": "LABEL_147",
160
+ "148": "LABEL_148",
161
+ "149": "LABEL_149",
162
+ "150": "LABEL_150"
163
+ },
164
+ "initializer_range": 0.02,
165
+ "label2id": {
166
+ "LABEL_0": 0,
167
+ "LABEL_1": 1,
168
+ "LABEL_10": 10,
169
+ "LABEL_100": 100,
170
+ "LABEL_101": 101,
171
+ "LABEL_102": 102,
172
+ "LABEL_103": 103,
173
+ "LABEL_104": 104,
174
+ "LABEL_105": 105,
175
+ "LABEL_106": 106,
176
+ "LABEL_107": 107,
177
+ "LABEL_108": 108,
178
+ "LABEL_109": 109,
179
+ "LABEL_11": 11,
180
+ "LABEL_110": 110,
181
+ "LABEL_111": 111,
182
+ "LABEL_112": 112,
183
+ "LABEL_113": 113,
184
+ "LABEL_114": 114,
185
+ "LABEL_115": 115,
186
+ "LABEL_116": 116,
187
+ "LABEL_117": 117,
188
+ "LABEL_118": 118,
189
+ "LABEL_119": 119,
190
+ "LABEL_12": 12,
191
+ "LABEL_120": 120,
192
+ "LABEL_121": 121,
193
+ "LABEL_122": 122,
194
+ "LABEL_123": 123,
195
+ "LABEL_124": 124,
196
+ "LABEL_125": 125,
197
+ "LABEL_126": 126,
198
+ "LABEL_127": 127,
199
+ "LABEL_128": 128,
200
+ "LABEL_129": 129,
201
+ "LABEL_13": 13,
202
+ "LABEL_130": 130,
203
+ "LABEL_131": 131,
204
+ "LABEL_132": 132,
205
+ "LABEL_133": 133,
206
+ "LABEL_134": 134,
207
+ "LABEL_135": 135,
208
+ "LABEL_136": 136,
209
+ "LABEL_137": 137,
210
+ "LABEL_138": 138,
211
+ "LABEL_139": 139,
212
+ "LABEL_14": 14,
213
+ "LABEL_140": 140,
214
+ "LABEL_141": 141,
215
+ "LABEL_142": 142,
216
+ "LABEL_143": 143,
217
+ "LABEL_144": 144,
218
+ "LABEL_145": 145,
219
+ "LABEL_146": 146,
220
+ "LABEL_147": 147,
221
+ "LABEL_148": 148,
222
+ "LABEL_149": 149,
223
+ "LABEL_15": 15,
224
+ "LABEL_150": 150,
225
+ "LABEL_16": 16,
226
+ "LABEL_17": 17,
227
+ "LABEL_18": 18,
228
+ "LABEL_19": 19,
229
+ "LABEL_2": 2,
230
+ "LABEL_20": 20,
231
+ "LABEL_21": 21,
232
+ "LABEL_22": 22,
233
+ "LABEL_23": 23,
234
+ "LABEL_24": 24,
235
+ "LABEL_25": 25,
236
+ "LABEL_26": 26,
237
+ "LABEL_27": 27,
238
+ "LABEL_28": 28,
239
+ "LABEL_29": 29,
240
+ "LABEL_3": 3,
241
+ "LABEL_30": 30,
242
+ "LABEL_31": 31,
243
+ "LABEL_32": 32,
244
+ "LABEL_33": 33,
245
+ "LABEL_34": 34,
246
+ "LABEL_35": 35,
247
+ "LABEL_36": 36,
248
+ "LABEL_37": 37,
249
+ "LABEL_38": 38,
250
+ "LABEL_39": 39,
251
+ "LABEL_4": 4,
252
+ "LABEL_40": 40,
253
+ "LABEL_41": 41,
254
+ "LABEL_42": 42,
255
+ "LABEL_43": 43,
256
+ "LABEL_44": 44,
257
+ "LABEL_45": 45,
258
+ "LABEL_46": 46,
259
+ "LABEL_47": 47,
260
+ "LABEL_48": 48,
261
+ "LABEL_49": 49,
262
+ "LABEL_5": 5,
263
+ "LABEL_50": 50,
264
+ "LABEL_51": 51,
265
+ "LABEL_52": 52,
266
+ "LABEL_53": 53,
267
+ "LABEL_54": 54,
268
+ "LABEL_55": 55,
269
+ "LABEL_56": 56,
270
+ "LABEL_57": 57,
271
+ "LABEL_58": 58,
272
+ "LABEL_59": 59,
273
+ "LABEL_6": 6,
274
+ "LABEL_60": 60,
275
+ "LABEL_61": 61,
276
+ "LABEL_62": 62,
277
+ "LABEL_63": 63,
278
+ "LABEL_64": 64,
279
+ "LABEL_65": 65,
280
+ "LABEL_66": 66,
281
+ "LABEL_67": 67,
282
+ "LABEL_68": 68,
283
+ "LABEL_69": 69,
284
+ "LABEL_7": 7,
285
+ "LABEL_70": 70,
286
+ "LABEL_71": 71,
287
+ "LABEL_72": 72,
288
+ "LABEL_73": 73,
289
+ "LABEL_74": 74,
290
+ "LABEL_75": 75,
291
+ "LABEL_76": 76,
292
+ "LABEL_77": 77,
293
+ "LABEL_78": 78,
294
+ "LABEL_79": 79,
295
+ "LABEL_8": 8,
296
+ "LABEL_80": 80,
297
+ "LABEL_81": 81,
298
+ "LABEL_82": 82,
299
+ "LABEL_83": 83,
300
+ "LABEL_84": 84,
301
+ "LABEL_85": 85,
302
+ "LABEL_86": 86,
303
+ "LABEL_87": 87,
304
+ "LABEL_88": 88,
305
+ "LABEL_89": 89,
306
+ "LABEL_9": 9,
307
+ "LABEL_90": 90,
308
+ "LABEL_91": 91,
309
+ "LABEL_92": 92,
310
+ "LABEL_93": 93,
311
+ "LABEL_94": 94,
312
+ "LABEL_95": 95,
313
+ "LABEL_96": 96,
314
+ "LABEL_97": 97,
315
+ "LABEL_98": 98,
316
+ "LABEL_99": 99
317
+ },
318
+ "max_position_embeddings": 512,
319
+ "model_type": "distilbert",
320
+ "n_heads": 12,
321
+ "n_layers": 6,
322
+ "pad_token_id": 0,
323
+ "problem_type": "single_label_classification",
324
+ "qa_dropout": 0.1,
325
+ "seq_classif_dropout": 0.2,
326
+ "sinusoidal_pos_embds": false,
327
+ "tie_weights_": true,
328
+ "transformers_version": "4.57.0",
329
+ "vocab_size": 30522
330
+ }
run-4/checkpoint-1500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:faf4246b010c79ae974566541fa36a6b63f08cd8e719b66cde5819521fa72522
3
+ size 268290900
run-4/checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0888b2931c695bf133c3f5e7e922364b751318382f006ad9def53382a0213c50
3
+ size 536645835
run-4/checkpoint-1500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b6c83f26643a77311ab57eab248e83731fa9e8d211e7359ac9b5df3296f760a
3
+ size 14645
run-4/checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70b70ddbefcfa3277288d512d07f3d8c83d3a9d932de6fb2c548390b26b3386d
3
+ size 1465
run-4/checkpoint-1500/special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
run-4/checkpoint-1500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-4/checkpoint-1500/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "pad_token": "[PAD]",
51
+ "sep_token": "[SEP]",
52
+ "strip_accents": null,
53
+ "tokenize_chinese_chars": true,
54
+ "tokenizer_class": "DistilBertTokenizer",
55
+ "unk_token": "[UNK]"
56
+ }
run-4/checkpoint-1500/trainer_state.json ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 4.716981132075472,
6
+ "eval_steps": 500,
7
+ "global_step": 1500,
8
+ "is_hyper_param_search": true,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "eval_accuracy": 0.6845161290322581,
15
+ "eval_loss": 0.10785406082868576,
16
+ "eval_runtime": 1.0937,
17
+ "eval_samples_per_second": 2834.299,
18
+ "eval_steps_per_second": 59.429,
19
+ "step": 318
20
+ },
21
+ {
22
+ "epoch": 1.5723270440251573,
23
+ "grad_norm": 0.4275546073913574,
24
+ "learning_rate": 1.6512928022361987e-05,
25
+ "loss": 0.2193,
26
+ "step": 500
27
+ },
28
+ {
29
+ "epoch": 2.0,
30
+ "eval_accuracy": 0.8438709677419355,
31
+ "eval_loss": 0.03479825332760811,
32
+ "eval_runtime": 1.097,
33
+ "eval_samples_per_second": 2825.922,
34
+ "eval_steps_per_second": 59.253,
35
+ "step": 636
36
+ },
37
+ {
38
+ "epoch": 3.0,
39
+ "eval_accuracy": 0.8703225806451613,
40
+ "eval_loss": 0.01689928211271763,
41
+ "eval_runtime": 1.0917,
42
+ "eval_samples_per_second": 2839.696,
43
+ "eval_steps_per_second": 59.542,
44
+ "step": 954
45
+ },
46
+ {
47
+ "epoch": 3.1446540880503147,
48
+ "grad_norm": 0.22454383969306946,
49
+ "learning_rate": 1.3018867924528303e-05,
50
+ "loss": 0.0599,
51
+ "step": 1000
52
+ },
53
+ {
54
+ "epoch": 4.0,
55
+ "eval_accuracy": 0.8829032258064516,
56
+ "eval_loss": 0.010705524124205112,
57
+ "eval_runtime": 1.0926,
58
+ "eval_samples_per_second": 2837.331,
59
+ "eval_steps_per_second": 59.492,
60
+ "step": 1272
61
+ },
62
+ {
63
+ "epoch": 4.716981132075472,
64
+ "grad_norm": 0.166969895362854,
65
+ "learning_rate": 9.52480782669462e-06,
66
+ "loss": 0.0389,
67
+ "step": 1500
68
+ }
69
+ ],
70
+ "logging_steps": 500,
71
+ "max_steps": 2862,
72
+ "num_input_tokens_seen": 0,
73
+ "num_train_epochs": 9,
74
+ "save_steps": 500,
75
+ "stateful_callbacks": {
76
+ "TrainerControl": {
77
+ "args": {
78
+ "should_epoch_stop": false,
79
+ "should_evaluate": false,
80
+ "should_log": false,
81
+ "should_save": true,
82
+ "should_training_stop": false
83
+ },
84
+ "attributes": {}
85
+ }
86
+ },
87
+ "total_flos": 391368939443328.0,
88
+ "train_batch_size": 48,
89
+ "trial_name": null,
90
+ "trial_params": {
91
+ "alpha": 0.9763664799887318,
92
+ "num_train_epochs": 9,
93
+ "temperature": 10
94
+ }
95
+ }
run-4/checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7cd8292610cea7a04a70d5b98c84050970fba3dabfd3ac7ee8ac97541992cb4
3
+ size 5841
run-4/checkpoint-1500/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
run-4/checkpoint-2000/config.json ADDED
@@ -0,0 +1,330 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "gelu",
3
+ "architectures": [
4
+ "DistilBertForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.1,
7
+ "dim": 768,
8
+ "dropout": 0.1,
9
+ "dtype": "float32",
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "LABEL_0",
13
+ "1": "LABEL_1",
14
+ "2": "LABEL_2",
15
+ "3": "LABEL_3",
16
+ "4": "LABEL_4",
17
+ "5": "LABEL_5",
18
+ "6": "LABEL_6",
19
+ "7": "LABEL_7",
20
+ "8": "LABEL_8",
21
+ "9": "LABEL_9",
22
+ "10": "LABEL_10",
23
+ "11": "LABEL_11",
24
+ "12": "LABEL_12",
25
+ "13": "LABEL_13",
26
+ "14": "LABEL_14",
27
+ "15": "LABEL_15",
28
+ "16": "LABEL_16",
29
+ "17": "LABEL_17",
30
+ "18": "LABEL_18",
31
+ "19": "LABEL_19",
32
+ "20": "LABEL_20",
33
+ "21": "LABEL_21",
34
+ "22": "LABEL_22",
35
+ "23": "LABEL_23",
36
+ "24": "LABEL_24",
37
+ "25": "LABEL_25",
38
+ "26": "LABEL_26",
39
+ "27": "LABEL_27",
40
+ "28": "LABEL_28",
41
+ "29": "LABEL_29",
42
+ "30": "LABEL_30",
43
+ "31": "LABEL_31",
44
+ "32": "LABEL_32",
45
+ "33": "LABEL_33",
46
+ "34": "LABEL_34",
47
+ "35": "LABEL_35",
48
+ "36": "LABEL_36",
49
+ "37": "LABEL_37",
50
+ "38": "LABEL_38",
51
+ "39": "LABEL_39",
52
+ "40": "LABEL_40",
53
+ "41": "LABEL_41",
54
+ "42": "LABEL_42",
55
+ "43": "LABEL_43",
56
+ "44": "LABEL_44",
57
+ "45": "LABEL_45",
58
+ "46": "LABEL_46",
59
+ "47": "LABEL_47",
60
+ "48": "LABEL_48",
61
+ "49": "LABEL_49",
62
+ "50": "LABEL_50",
63
+ "51": "LABEL_51",
64
+ "52": "LABEL_52",
65
+ "53": "LABEL_53",
66
+ "54": "LABEL_54",
67
+ "55": "LABEL_55",
68
+ "56": "LABEL_56",
69
+ "57": "LABEL_57",
70
+ "58": "LABEL_58",
71
+ "59": "LABEL_59",
72
+ "60": "LABEL_60",
73
+ "61": "LABEL_61",
74
+ "62": "LABEL_62",
75
+ "63": "LABEL_63",
76
+ "64": "LABEL_64",
77
+ "65": "LABEL_65",
78
+ "66": "LABEL_66",
79
+ "67": "LABEL_67",
80
+ "68": "LABEL_68",
81
+ "69": "LABEL_69",
82
+ "70": "LABEL_70",
83
+ "71": "LABEL_71",
84
+ "72": "LABEL_72",
85
+ "73": "LABEL_73",
86
+ "74": "LABEL_74",
87
+ "75": "LABEL_75",
88
+ "76": "LABEL_76",
89
+ "77": "LABEL_77",
90
+ "78": "LABEL_78",
91
+ "79": "LABEL_79",
92
+ "80": "LABEL_80",
93
+ "81": "LABEL_81",
94
+ "82": "LABEL_82",
95
+ "83": "LABEL_83",
96
+ "84": "LABEL_84",
97
+ "85": "LABEL_85",
98
+ "86": "LABEL_86",
99
+ "87": "LABEL_87",
100
+ "88": "LABEL_88",
101
+ "89": "LABEL_89",
102
+ "90": "LABEL_90",
103
+ "91": "LABEL_91",
104
+ "92": "LABEL_92",
105
+ "93": "LABEL_93",
106
+ "94": "LABEL_94",
107
+ "95": "LABEL_95",
108
+ "96": "LABEL_96",
109
+ "97": "LABEL_97",
110
+ "98": "LABEL_98",
111
+ "99": "LABEL_99",
112
+ "100": "LABEL_100",
113
+ "101": "LABEL_101",
114
+ "102": "LABEL_102",
115
+ "103": "LABEL_103",
116
+ "104": "LABEL_104",
117
+ "105": "LABEL_105",
118
+ "106": "LABEL_106",
119
+ "107": "LABEL_107",
120
+ "108": "LABEL_108",
121
+ "109": "LABEL_109",
122
+ "110": "LABEL_110",
123
+ "111": "LABEL_111",
124
+ "112": "LABEL_112",
125
+ "113": "LABEL_113",
126
+ "114": "LABEL_114",
127
+ "115": "LABEL_115",
128
+ "116": "LABEL_116",
129
+ "117": "LABEL_117",
130
+ "118": "LABEL_118",
131
+ "119": "LABEL_119",
132
+ "120": "LABEL_120",
133
+ "121": "LABEL_121",
134
+ "122": "LABEL_122",
135
+ "123": "LABEL_123",
136
+ "124": "LABEL_124",
137
+ "125": "LABEL_125",
138
+ "126": "LABEL_126",
139
+ "127": "LABEL_127",
140
+ "128": "LABEL_128",
141
+ "129": "LABEL_129",
142
+ "130": "LABEL_130",
143
+ "131": "LABEL_131",
144
+ "132": "LABEL_132",
145
+ "133": "LABEL_133",
146
+ "134": "LABEL_134",
147
+ "135": "LABEL_135",
148
+ "136": "LABEL_136",
149
+ "137": "LABEL_137",
150
+ "138": "LABEL_138",
151
+ "139": "LABEL_139",
152
+ "140": "LABEL_140",
153
+ "141": "LABEL_141",
154
+ "142": "LABEL_142",
155
+ "143": "LABEL_143",
156
+ "144": "LABEL_144",
157
+ "145": "LABEL_145",
158
+ "146": "LABEL_146",
159
+ "147": "LABEL_147",
160
+ "148": "LABEL_148",
161
+ "149": "LABEL_149",
162
+ "150": "LABEL_150"
163
+ },
164
+ "initializer_range": 0.02,
165
+ "label2id": {
166
+ "LABEL_0": 0,
167
+ "LABEL_1": 1,
168
+ "LABEL_10": 10,
169
+ "LABEL_100": 100,
170
+ "LABEL_101": 101,
171
+ "LABEL_102": 102,
172
+ "LABEL_103": 103,
173
+ "LABEL_104": 104,
174
+ "LABEL_105": 105,
175
+ "LABEL_106": 106,
176
+ "LABEL_107": 107,
177
+ "LABEL_108": 108,
178
+ "LABEL_109": 109,
179
+ "LABEL_11": 11,
180
+ "LABEL_110": 110,
181
+ "LABEL_111": 111,
182
+ "LABEL_112": 112,
183
+ "LABEL_113": 113,
184
+ "LABEL_114": 114,
185
+ "LABEL_115": 115,
186
+ "LABEL_116": 116,
187
+ "LABEL_117": 117,
188
+ "LABEL_118": 118,
189
+ "LABEL_119": 119,
190
+ "LABEL_12": 12,
191
+ "LABEL_120": 120,
192
+ "LABEL_121": 121,
193
+ "LABEL_122": 122,
194
+ "LABEL_123": 123,
195
+ "LABEL_124": 124,
196
+ "LABEL_125": 125,
197
+ "LABEL_126": 126,
198
+ "LABEL_127": 127,
199
+ "LABEL_128": 128,
200
+ "LABEL_129": 129,
201
+ "LABEL_13": 13,
202
+ "LABEL_130": 130,
203
+ "LABEL_131": 131,
204
+ "LABEL_132": 132,
205
+ "LABEL_133": 133,
206
+ "LABEL_134": 134,
207
+ "LABEL_135": 135,
208
+ "LABEL_136": 136,
209
+ "LABEL_137": 137,
210
+ "LABEL_138": 138,
211
+ "LABEL_139": 139,
212
+ "LABEL_14": 14,
213
+ "LABEL_140": 140,
214
+ "LABEL_141": 141,
215
+ "LABEL_142": 142,
216
+ "LABEL_143": 143,
217
+ "LABEL_144": 144,
218
+ "LABEL_145": 145,
219
+ "LABEL_146": 146,
220
+ "LABEL_147": 147,
221
+ "LABEL_148": 148,
222
+ "LABEL_149": 149,
223
+ "LABEL_15": 15,
224
+ "LABEL_150": 150,
225
+ "LABEL_16": 16,
226
+ "LABEL_17": 17,
227
+ "LABEL_18": 18,
228
+ "LABEL_19": 19,
229
+ "LABEL_2": 2,
230
+ "LABEL_20": 20,
231
+ "LABEL_21": 21,
232
+ "LABEL_22": 22,
233
+ "LABEL_23": 23,
234
+ "LABEL_24": 24,
235
+ "LABEL_25": 25,
236
+ "LABEL_26": 26,
237
+ "LABEL_27": 27,
238
+ "LABEL_28": 28,
239
+ "LABEL_29": 29,
240
+ "LABEL_3": 3,
241
+ "LABEL_30": 30,
242
+ "LABEL_31": 31,
243
+ "LABEL_32": 32,
244
+ "LABEL_33": 33,
245
+ "LABEL_34": 34,
246
+ "LABEL_35": 35,
247
+ "LABEL_36": 36,
248
+ "LABEL_37": 37,
249
+ "LABEL_38": 38,
250
+ "LABEL_39": 39,
251
+ "LABEL_4": 4,
252
+ "LABEL_40": 40,
253
+ "LABEL_41": 41,
254
+ "LABEL_42": 42,
255
+ "LABEL_43": 43,
256
+ "LABEL_44": 44,
257
+ "LABEL_45": 45,
258
+ "LABEL_46": 46,
259
+ "LABEL_47": 47,
260
+ "LABEL_48": 48,
261
+ "LABEL_49": 49,
262
+ "LABEL_5": 5,
263
+ "LABEL_50": 50,
264
+ "LABEL_51": 51,
265
+ "LABEL_52": 52,
266
+ "LABEL_53": 53,
267
+ "LABEL_54": 54,
268
+ "LABEL_55": 55,
269
+ "LABEL_56": 56,
270
+ "LABEL_57": 57,
271
+ "LABEL_58": 58,
272
+ "LABEL_59": 59,
273
+ "LABEL_6": 6,
274
+ "LABEL_60": 60,
275
+ "LABEL_61": 61,
276
+ "LABEL_62": 62,
277
+ "LABEL_63": 63,
278
+ "LABEL_64": 64,
279
+ "LABEL_65": 65,
280
+ "LABEL_66": 66,
281
+ "LABEL_67": 67,
282
+ "LABEL_68": 68,
283
+ "LABEL_69": 69,
284
+ "LABEL_7": 7,
285
+ "LABEL_70": 70,
286
+ "LABEL_71": 71,
287
+ "LABEL_72": 72,
288
+ "LABEL_73": 73,
289
+ "LABEL_74": 74,
290
+ "LABEL_75": 75,
291
+ "LABEL_76": 76,
292
+ "LABEL_77": 77,
293
+ "LABEL_78": 78,
294
+ "LABEL_79": 79,
295
+ "LABEL_8": 8,
296
+ "LABEL_80": 80,
297
+ "LABEL_81": 81,
298
+ "LABEL_82": 82,
299
+ "LABEL_83": 83,
300
+ "LABEL_84": 84,
301
+ "LABEL_85": 85,
302
+ "LABEL_86": 86,
303
+ "LABEL_87": 87,
304
+ "LABEL_88": 88,
305
+ "LABEL_89": 89,
306
+ "LABEL_9": 9,
307
+ "LABEL_90": 90,
308
+ "LABEL_91": 91,
309
+ "LABEL_92": 92,
310
+ "LABEL_93": 93,
311
+ "LABEL_94": 94,
312
+ "LABEL_95": 95,
313
+ "LABEL_96": 96,
314
+ "LABEL_97": 97,
315
+ "LABEL_98": 98,
316
+ "LABEL_99": 99
317
+ },
318
+ "max_position_embeddings": 512,
319
+ "model_type": "distilbert",
320
+ "n_heads": 12,
321
+ "n_layers": 6,
322
+ "pad_token_id": 0,
323
+ "problem_type": "single_label_classification",
324
+ "qa_dropout": 0.1,
325
+ "seq_classif_dropout": 0.2,
326
+ "sinusoidal_pos_embds": false,
327
+ "tie_weights_": true,
328
+ "transformers_version": "4.57.0",
329
+ "vocab_size": 30522
330
+ }
run-4/checkpoint-2000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:187ceea69b00ba222029a50cb15d9cb819032882e520ff00277195d7f85472c2
3
+ size 268290900
run-4/checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c94da2a4af3494b5f39bae6854275b93c7c4cea2de3a2fb9c3aab8070294a52
3
+ size 536645835
run-4/checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c8b5ffe3d30fdc8b63b5ecf062b7473e81dd47473350776eeab61ee54625fe12
3
+ size 14645
run-4/checkpoint-2000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ee83880cade73641ded87dd7e6193f29df5476dae4a4bb1ed8ac3581afb387cc
3
+ size 1465
run-4/checkpoint-2000/special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
run-4/checkpoint-2000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-4/checkpoint-2000/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "pad_token": "[PAD]",
51
+ "sep_token": "[SEP]",
52
+ "strip_accents": null,
53
+ "tokenize_chinese_chars": true,
54
+ "tokenizer_class": "DistilBertTokenizer",
55
+ "unk_token": "[UNK]"
56
+ }
run-4/checkpoint-2000/trainer_state.json ADDED
@@ -0,0 +1,120 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 6.289308176100629,
6
+ "eval_steps": 500,
7
+ "global_step": 2000,
8
+ "is_hyper_param_search": true,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "eval_accuracy": 0.6845161290322581,
15
+ "eval_loss": 0.10785406082868576,
16
+ "eval_runtime": 1.0937,
17
+ "eval_samples_per_second": 2834.299,
18
+ "eval_steps_per_second": 59.429,
19
+ "step": 318
20
+ },
21
+ {
22
+ "epoch": 1.5723270440251573,
23
+ "grad_norm": 0.4275546073913574,
24
+ "learning_rate": 1.6512928022361987e-05,
25
+ "loss": 0.2193,
26
+ "step": 500
27
+ },
28
+ {
29
+ "epoch": 2.0,
30
+ "eval_accuracy": 0.8438709677419355,
31
+ "eval_loss": 0.03479825332760811,
32
+ "eval_runtime": 1.097,
33
+ "eval_samples_per_second": 2825.922,
34
+ "eval_steps_per_second": 59.253,
35
+ "step": 636
36
+ },
37
+ {
38
+ "epoch": 3.0,
39
+ "eval_accuracy": 0.8703225806451613,
40
+ "eval_loss": 0.01689928211271763,
41
+ "eval_runtime": 1.0917,
42
+ "eval_samples_per_second": 2839.696,
43
+ "eval_steps_per_second": 59.542,
44
+ "step": 954
45
+ },
46
+ {
47
+ "epoch": 3.1446540880503147,
48
+ "grad_norm": 0.22454383969306946,
49
+ "learning_rate": 1.3018867924528303e-05,
50
+ "loss": 0.0599,
51
+ "step": 1000
52
+ },
53
+ {
54
+ "epoch": 4.0,
55
+ "eval_accuracy": 0.8829032258064516,
56
+ "eval_loss": 0.010705524124205112,
57
+ "eval_runtime": 1.0926,
58
+ "eval_samples_per_second": 2837.331,
59
+ "eval_steps_per_second": 59.492,
60
+ "step": 1272
61
+ },
62
+ {
63
+ "epoch": 4.716981132075472,
64
+ "grad_norm": 0.166969895362854,
65
+ "learning_rate": 9.52480782669462e-06,
66
+ "loss": 0.0389,
67
+ "step": 1500
68
+ },
69
+ {
70
+ "epoch": 5.0,
71
+ "eval_accuracy": 0.8932258064516129,
72
+ "eval_loss": 0.008237027563154697,
73
+ "eval_runtime": 1.0908,
74
+ "eval_samples_per_second": 2841.932,
75
+ "eval_steps_per_second": 59.589,
76
+ "step": 1590
77
+ },
78
+ {
79
+ "epoch": 6.0,
80
+ "eval_accuracy": 0.895483870967742,
81
+ "eval_loss": 0.006902290508151054,
82
+ "eval_runtime": 1.2987,
83
+ "eval_samples_per_second": 2387.094,
84
+ "eval_steps_per_second": 50.052,
85
+ "step": 1908
86
+ },
87
+ {
88
+ "epoch": 6.289308176100629,
89
+ "grad_norm": 0.14763705432415009,
90
+ "learning_rate": 6.030747728860937e-06,
91
+ "loss": 0.0326,
92
+ "step": 2000
93
+ }
94
+ ],
95
+ "logging_steps": 500,
96
+ "max_steps": 2862,
97
+ "num_input_tokens_seen": 0,
98
+ "num_train_epochs": 9,
99
+ "save_steps": 500,
100
+ "stateful_callbacks": {
101
+ "TrainerControl": {
102
+ "args": {
103
+ "should_epoch_stop": false,
104
+ "should_evaluate": false,
105
+ "should_log": false,
106
+ "should_save": true,
107
+ "should_training_stop": false
108
+ },
109
+ "attributes": {}
110
+ }
111
+ },
112
+ "total_flos": 520991326672152.0,
113
+ "train_batch_size": 48,
114
+ "trial_name": null,
115
+ "trial_params": {
116
+ "alpha": 0.9763664799887318,
117
+ "num_train_epochs": 9,
118
+ "temperature": 10
119
+ }
120
+ }
run-4/checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7cd8292610cea7a04a70d5b98c84050970fba3dabfd3ac7ee8ac97541992cb4
3
+ size 5841
run-4/checkpoint-2000/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
run-4/checkpoint-2500/config.json ADDED
@@ -0,0 +1,330 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "gelu",
3
+ "architectures": [
4
+ "DistilBertForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.1,
7
+ "dim": 768,
8
+ "dropout": 0.1,
9
+ "dtype": "float32",
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "LABEL_0",
13
+ "1": "LABEL_1",
14
+ "2": "LABEL_2",
15
+ "3": "LABEL_3",
16
+ "4": "LABEL_4",
17
+ "5": "LABEL_5",
18
+ "6": "LABEL_6",
19
+ "7": "LABEL_7",
20
+ "8": "LABEL_8",
21
+ "9": "LABEL_9",
22
+ "10": "LABEL_10",
23
+ "11": "LABEL_11",
24
+ "12": "LABEL_12",
25
+ "13": "LABEL_13",
26
+ "14": "LABEL_14",
27
+ "15": "LABEL_15",
28
+ "16": "LABEL_16",
29
+ "17": "LABEL_17",
30
+ "18": "LABEL_18",
31
+ "19": "LABEL_19",
32
+ "20": "LABEL_20",
33
+ "21": "LABEL_21",
34
+ "22": "LABEL_22",
35
+ "23": "LABEL_23",
36
+ "24": "LABEL_24",
37
+ "25": "LABEL_25",
38
+ "26": "LABEL_26",
39
+ "27": "LABEL_27",
40
+ "28": "LABEL_28",
41
+ "29": "LABEL_29",
42
+ "30": "LABEL_30",
43
+ "31": "LABEL_31",
44
+ "32": "LABEL_32",
45
+ "33": "LABEL_33",
46
+ "34": "LABEL_34",
47
+ "35": "LABEL_35",
48
+ "36": "LABEL_36",
49
+ "37": "LABEL_37",
50
+ "38": "LABEL_38",
51
+ "39": "LABEL_39",
52
+ "40": "LABEL_40",
53
+ "41": "LABEL_41",
54
+ "42": "LABEL_42",
55
+ "43": "LABEL_43",
56
+ "44": "LABEL_44",
57
+ "45": "LABEL_45",
58
+ "46": "LABEL_46",
59
+ "47": "LABEL_47",
60
+ "48": "LABEL_48",
61
+ "49": "LABEL_49",
62
+ "50": "LABEL_50",
63
+ "51": "LABEL_51",
64
+ "52": "LABEL_52",
65
+ "53": "LABEL_53",
66
+ "54": "LABEL_54",
67
+ "55": "LABEL_55",
68
+ "56": "LABEL_56",
69
+ "57": "LABEL_57",
70
+ "58": "LABEL_58",
71
+ "59": "LABEL_59",
72
+ "60": "LABEL_60",
73
+ "61": "LABEL_61",
74
+ "62": "LABEL_62",
75
+ "63": "LABEL_63",
76
+ "64": "LABEL_64",
77
+ "65": "LABEL_65",
78
+ "66": "LABEL_66",
79
+ "67": "LABEL_67",
80
+ "68": "LABEL_68",
81
+ "69": "LABEL_69",
82
+ "70": "LABEL_70",
83
+ "71": "LABEL_71",
84
+ "72": "LABEL_72",
85
+ "73": "LABEL_73",
86
+ "74": "LABEL_74",
87
+ "75": "LABEL_75",
88
+ "76": "LABEL_76",
89
+ "77": "LABEL_77",
90
+ "78": "LABEL_78",
91
+ "79": "LABEL_79",
92
+ "80": "LABEL_80",
93
+ "81": "LABEL_81",
94
+ "82": "LABEL_82",
95
+ "83": "LABEL_83",
96
+ "84": "LABEL_84",
97
+ "85": "LABEL_85",
98
+ "86": "LABEL_86",
99
+ "87": "LABEL_87",
100
+ "88": "LABEL_88",
101
+ "89": "LABEL_89",
102
+ "90": "LABEL_90",
103
+ "91": "LABEL_91",
104
+ "92": "LABEL_92",
105
+ "93": "LABEL_93",
106
+ "94": "LABEL_94",
107
+ "95": "LABEL_95",
108
+ "96": "LABEL_96",
109
+ "97": "LABEL_97",
110
+ "98": "LABEL_98",
111
+ "99": "LABEL_99",
112
+ "100": "LABEL_100",
113
+ "101": "LABEL_101",
114
+ "102": "LABEL_102",
115
+ "103": "LABEL_103",
116
+ "104": "LABEL_104",
117
+ "105": "LABEL_105",
118
+ "106": "LABEL_106",
119
+ "107": "LABEL_107",
120
+ "108": "LABEL_108",
121
+ "109": "LABEL_109",
122
+ "110": "LABEL_110",
123
+ "111": "LABEL_111",
124
+ "112": "LABEL_112",
125
+ "113": "LABEL_113",
126
+ "114": "LABEL_114",
127
+ "115": "LABEL_115",
128
+ "116": "LABEL_116",
129
+ "117": "LABEL_117",
130
+ "118": "LABEL_118",
131
+ "119": "LABEL_119",
132
+ "120": "LABEL_120",
133
+ "121": "LABEL_121",
134
+ "122": "LABEL_122",
135
+ "123": "LABEL_123",
136
+ "124": "LABEL_124",
137
+ "125": "LABEL_125",
138
+ "126": "LABEL_126",
139
+ "127": "LABEL_127",
140
+ "128": "LABEL_128",
141
+ "129": "LABEL_129",
142
+ "130": "LABEL_130",
143
+ "131": "LABEL_131",
144
+ "132": "LABEL_132",
145
+ "133": "LABEL_133",
146
+ "134": "LABEL_134",
147
+ "135": "LABEL_135",
148
+ "136": "LABEL_136",
149
+ "137": "LABEL_137",
150
+ "138": "LABEL_138",
151
+ "139": "LABEL_139",
152
+ "140": "LABEL_140",
153
+ "141": "LABEL_141",
154
+ "142": "LABEL_142",
155
+ "143": "LABEL_143",
156
+ "144": "LABEL_144",
157
+ "145": "LABEL_145",
158
+ "146": "LABEL_146",
159
+ "147": "LABEL_147",
160
+ "148": "LABEL_148",
161
+ "149": "LABEL_149",
162
+ "150": "LABEL_150"
163
+ },
164
+ "initializer_range": 0.02,
165
+ "label2id": {
166
+ "LABEL_0": 0,
167
+ "LABEL_1": 1,
168
+ "LABEL_10": 10,
169
+ "LABEL_100": 100,
170
+ "LABEL_101": 101,
171
+ "LABEL_102": 102,
172
+ "LABEL_103": 103,
173
+ "LABEL_104": 104,
174
+ "LABEL_105": 105,
175
+ "LABEL_106": 106,
176
+ "LABEL_107": 107,
177
+ "LABEL_108": 108,
178
+ "LABEL_109": 109,
179
+ "LABEL_11": 11,
180
+ "LABEL_110": 110,
181
+ "LABEL_111": 111,
182
+ "LABEL_112": 112,
183
+ "LABEL_113": 113,
184
+ "LABEL_114": 114,
185
+ "LABEL_115": 115,
186
+ "LABEL_116": 116,
187
+ "LABEL_117": 117,
188
+ "LABEL_118": 118,
189
+ "LABEL_119": 119,
190
+ "LABEL_12": 12,
191
+ "LABEL_120": 120,
192
+ "LABEL_121": 121,
193
+ "LABEL_122": 122,
194
+ "LABEL_123": 123,
195
+ "LABEL_124": 124,
196
+ "LABEL_125": 125,
197
+ "LABEL_126": 126,
198
+ "LABEL_127": 127,
199
+ "LABEL_128": 128,
200
+ "LABEL_129": 129,
201
+ "LABEL_13": 13,
202
+ "LABEL_130": 130,
203
+ "LABEL_131": 131,
204
+ "LABEL_132": 132,
205
+ "LABEL_133": 133,
206
+ "LABEL_134": 134,
207
+ "LABEL_135": 135,
208
+ "LABEL_136": 136,
209
+ "LABEL_137": 137,
210
+ "LABEL_138": 138,
211
+ "LABEL_139": 139,
212
+ "LABEL_14": 14,
213
+ "LABEL_140": 140,
214
+ "LABEL_141": 141,
215
+ "LABEL_142": 142,
216
+ "LABEL_143": 143,
217
+ "LABEL_144": 144,
218
+ "LABEL_145": 145,
219
+ "LABEL_146": 146,
220
+ "LABEL_147": 147,
221
+ "LABEL_148": 148,
222
+ "LABEL_149": 149,
223
+ "LABEL_15": 15,
224
+ "LABEL_150": 150,
225
+ "LABEL_16": 16,
226
+ "LABEL_17": 17,
227
+ "LABEL_18": 18,
228
+ "LABEL_19": 19,
229
+ "LABEL_2": 2,
230
+ "LABEL_20": 20,
231
+ "LABEL_21": 21,
232
+ "LABEL_22": 22,
233
+ "LABEL_23": 23,
234
+ "LABEL_24": 24,
235
+ "LABEL_25": 25,
236
+ "LABEL_26": 26,
237
+ "LABEL_27": 27,
238
+ "LABEL_28": 28,
239
+ "LABEL_29": 29,
240
+ "LABEL_3": 3,
241
+ "LABEL_30": 30,
242
+ "LABEL_31": 31,
243
+ "LABEL_32": 32,
244
+ "LABEL_33": 33,
245
+ "LABEL_34": 34,
246
+ "LABEL_35": 35,
247
+ "LABEL_36": 36,
248
+ "LABEL_37": 37,
249
+ "LABEL_38": 38,
250
+ "LABEL_39": 39,
251
+ "LABEL_4": 4,
252
+ "LABEL_40": 40,
253
+ "LABEL_41": 41,
254
+ "LABEL_42": 42,
255
+ "LABEL_43": 43,
256
+ "LABEL_44": 44,
257
+ "LABEL_45": 45,
258
+ "LABEL_46": 46,
259
+ "LABEL_47": 47,
260
+ "LABEL_48": 48,
261
+ "LABEL_49": 49,
262
+ "LABEL_5": 5,
263
+ "LABEL_50": 50,
264
+ "LABEL_51": 51,
265
+ "LABEL_52": 52,
266
+ "LABEL_53": 53,
267
+ "LABEL_54": 54,
268
+ "LABEL_55": 55,
269
+ "LABEL_56": 56,
270
+ "LABEL_57": 57,
271
+ "LABEL_58": 58,
272
+ "LABEL_59": 59,
273
+ "LABEL_6": 6,
274
+ "LABEL_60": 60,
275
+ "LABEL_61": 61,
276
+ "LABEL_62": 62,
277
+ "LABEL_63": 63,
278
+ "LABEL_64": 64,
279
+ "LABEL_65": 65,
280
+ "LABEL_66": 66,
281
+ "LABEL_67": 67,
282
+ "LABEL_68": 68,
283
+ "LABEL_69": 69,
284
+ "LABEL_7": 7,
285
+ "LABEL_70": 70,
286
+ "LABEL_71": 71,
287
+ "LABEL_72": 72,
288
+ "LABEL_73": 73,
289
+ "LABEL_74": 74,
290
+ "LABEL_75": 75,
291
+ "LABEL_76": 76,
292
+ "LABEL_77": 77,
293
+ "LABEL_78": 78,
294
+ "LABEL_79": 79,
295
+ "LABEL_8": 8,
296
+ "LABEL_80": 80,
297
+ "LABEL_81": 81,
298
+ "LABEL_82": 82,
299
+ "LABEL_83": 83,
300
+ "LABEL_84": 84,
301
+ "LABEL_85": 85,
302
+ "LABEL_86": 86,
303
+ "LABEL_87": 87,
304
+ "LABEL_88": 88,
305
+ "LABEL_89": 89,
306
+ "LABEL_9": 9,
307
+ "LABEL_90": 90,
308
+ "LABEL_91": 91,
309
+ "LABEL_92": 92,
310
+ "LABEL_93": 93,
311
+ "LABEL_94": 94,
312
+ "LABEL_95": 95,
313
+ "LABEL_96": 96,
314
+ "LABEL_97": 97,
315
+ "LABEL_98": 98,
316
+ "LABEL_99": 99
317
+ },
318
+ "max_position_embeddings": 512,
319
+ "model_type": "distilbert",
320
+ "n_heads": 12,
321
+ "n_layers": 6,
322
+ "pad_token_id": 0,
323
+ "problem_type": "single_label_classification",
324
+ "qa_dropout": 0.1,
325
+ "seq_classif_dropout": 0.2,
326
+ "sinusoidal_pos_embds": false,
327
+ "tie_weights_": true,
328
+ "transformers_version": "4.57.0",
329
+ "vocab_size": 30522
330
+ }
run-4/checkpoint-2500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:15ca40870a8f54728bb87b6467c3e67fe9e7a162ed0a54c2cece6cd49fa704ef
3
+ size 268290900
run-4/checkpoint-2500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f8b59e6fd8941d73a55116d4bffd97945538acbdd790c86010a82d8bc7da288
3
+ size 536645835
run-4/checkpoint-2500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a766bbb59b6c537b5f237f67f5b650c80553d4505ede2b6080d492a45f0bea8
3
+ size 14645
run-4/checkpoint-2500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f35d5a79d1e55ab45da1b3d70c8f3ded036b0315a6f37b852b01951f6d9d0c0
3
+ size 1465
run-4/checkpoint-2500/special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
run-4/checkpoint-2500/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
run-4/checkpoint-2500/tokenizer_config.json ADDED
@@ -0,0 +1,56 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "100": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "101": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "102": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "103": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": false,
45
+ "cls_token": "[CLS]",
46
+ "do_lower_case": true,
47
+ "extra_special_tokens": {},
48
+ "mask_token": "[MASK]",
49
+ "model_max_length": 512,
50
+ "pad_token": "[PAD]",
51
+ "sep_token": "[SEP]",
52
+ "strip_accents": null,
53
+ "tokenize_chinese_chars": true,
54
+ "tokenizer_class": "DistilBertTokenizer",
55
+ "unk_token": "[UNK]"
56
+ }
run-4/checkpoint-2500/trainer_state.json ADDED
@@ -0,0 +1,136 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 7.861635220125786,
6
+ "eval_steps": 500,
7
+ "global_step": 2500,
8
+ "is_hyper_param_search": true,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "eval_accuracy": 0.6845161290322581,
15
+ "eval_loss": 0.10785406082868576,
16
+ "eval_runtime": 1.0937,
17
+ "eval_samples_per_second": 2834.299,
18
+ "eval_steps_per_second": 59.429,
19
+ "step": 318
20
+ },
21
+ {
22
+ "epoch": 1.5723270440251573,
23
+ "grad_norm": 0.4275546073913574,
24
+ "learning_rate": 1.6512928022361987e-05,
25
+ "loss": 0.2193,
26
+ "step": 500
27
+ },
28
+ {
29
+ "epoch": 2.0,
30
+ "eval_accuracy": 0.8438709677419355,
31
+ "eval_loss": 0.03479825332760811,
32
+ "eval_runtime": 1.097,
33
+ "eval_samples_per_second": 2825.922,
34
+ "eval_steps_per_second": 59.253,
35
+ "step": 636
36
+ },
37
+ {
38
+ "epoch": 3.0,
39
+ "eval_accuracy": 0.8703225806451613,
40
+ "eval_loss": 0.01689928211271763,
41
+ "eval_runtime": 1.0917,
42
+ "eval_samples_per_second": 2839.696,
43
+ "eval_steps_per_second": 59.542,
44
+ "step": 954
45
+ },
46
+ {
47
+ "epoch": 3.1446540880503147,
48
+ "grad_norm": 0.22454383969306946,
49
+ "learning_rate": 1.3018867924528303e-05,
50
+ "loss": 0.0599,
51
+ "step": 1000
52
+ },
53
+ {
54
+ "epoch": 4.0,
55
+ "eval_accuracy": 0.8829032258064516,
56
+ "eval_loss": 0.010705524124205112,
57
+ "eval_runtime": 1.0926,
58
+ "eval_samples_per_second": 2837.331,
59
+ "eval_steps_per_second": 59.492,
60
+ "step": 1272
61
+ },
62
+ {
63
+ "epoch": 4.716981132075472,
64
+ "grad_norm": 0.166969895362854,
65
+ "learning_rate": 9.52480782669462e-06,
66
+ "loss": 0.0389,
67
+ "step": 1500
68
+ },
69
+ {
70
+ "epoch": 5.0,
71
+ "eval_accuracy": 0.8932258064516129,
72
+ "eval_loss": 0.008237027563154697,
73
+ "eval_runtime": 1.0908,
74
+ "eval_samples_per_second": 2841.932,
75
+ "eval_steps_per_second": 59.589,
76
+ "step": 1590
77
+ },
78
+ {
79
+ "epoch": 6.0,
80
+ "eval_accuracy": 0.895483870967742,
81
+ "eval_loss": 0.006902290508151054,
82
+ "eval_runtime": 1.2987,
83
+ "eval_samples_per_second": 2387.094,
84
+ "eval_steps_per_second": 50.052,
85
+ "step": 1908
86
+ },
87
+ {
88
+ "epoch": 6.289308176100629,
89
+ "grad_norm": 0.14763705432415009,
90
+ "learning_rate": 6.030747728860937e-06,
91
+ "loss": 0.0326,
92
+ "step": 2000
93
+ },
94
+ {
95
+ "epoch": 7.0,
96
+ "eval_accuracy": 0.9006451612903226,
97
+ "eval_loss": 0.006077259313315153,
98
+ "eval_runtime": 1.1676,
99
+ "eval_samples_per_second": 2655.017,
100
+ "eval_steps_per_second": 55.67,
101
+ "step": 2226
102
+ },
103
+ {
104
+ "epoch": 7.861635220125786,
105
+ "grad_norm": 0.13377271592617035,
106
+ "learning_rate": 2.5366876310272536e-06,
107
+ "loss": 0.03,
108
+ "step": 2500
109
+ }
110
+ ],
111
+ "logging_steps": 500,
112
+ "max_steps": 2862,
113
+ "num_input_tokens_seen": 0,
114
+ "num_train_epochs": 9,
115
+ "save_steps": 500,
116
+ "stateful_callbacks": {
117
+ "TrainerControl": {
118
+ "args": {
119
+ "should_epoch_stop": false,
120
+ "should_evaluate": false,
121
+ "should_log": false,
122
+ "should_save": true,
123
+ "should_training_stop": false
124
+ },
125
+ "attributes": {}
126
+ }
127
+ },
128
+ "total_flos": 651155886807636.0,
129
+ "train_batch_size": 48,
130
+ "trial_name": null,
131
+ "trial_params": {
132
+ "alpha": 0.9763664799887318,
133
+ "num_train_epochs": 9,
134
+ "temperature": 10
135
+ }
136
+ }
run-4/checkpoint-2500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7cd8292610cea7a04a70d5b98c84050970fba3dabfd3ac7ee8ac97541992cb4
3
+ size 5841
run-4/checkpoint-2500/vocab.txt ADDED
The diff for this file is too large to render. See raw diff
 
run-4/checkpoint-2862/config.json ADDED
@@ -0,0 +1,330 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation": "gelu",
3
+ "architectures": [
4
+ "DistilBertForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.1,
7
+ "dim": 768,
8
+ "dropout": 0.1,
9
+ "dtype": "float32",
10
+ "hidden_dim": 3072,
11
+ "id2label": {
12
+ "0": "LABEL_0",
13
+ "1": "LABEL_1",
14
+ "2": "LABEL_2",
15
+ "3": "LABEL_3",
16
+ "4": "LABEL_4",
17
+ "5": "LABEL_5",
18
+ "6": "LABEL_6",
19
+ "7": "LABEL_7",
20
+ "8": "LABEL_8",
21
+ "9": "LABEL_9",
22
+ "10": "LABEL_10",
23
+ "11": "LABEL_11",
24
+ "12": "LABEL_12",
25
+ "13": "LABEL_13",
26
+ "14": "LABEL_14",
27
+ "15": "LABEL_15",
28
+ "16": "LABEL_16",
29
+ "17": "LABEL_17",
30
+ "18": "LABEL_18",
31
+ "19": "LABEL_19",
32
+ "20": "LABEL_20",
33
+ "21": "LABEL_21",
34
+ "22": "LABEL_22",
35
+ "23": "LABEL_23",
36
+ "24": "LABEL_24",
37
+ "25": "LABEL_25",
38
+ "26": "LABEL_26",
39
+ "27": "LABEL_27",
40
+ "28": "LABEL_28",
41
+ "29": "LABEL_29",
42
+ "30": "LABEL_30",
43
+ "31": "LABEL_31",
44
+ "32": "LABEL_32",
45
+ "33": "LABEL_33",
46
+ "34": "LABEL_34",
47
+ "35": "LABEL_35",
48
+ "36": "LABEL_36",
49
+ "37": "LABEL_37",
50
+ "38": "LABEL_38",
51
+ "39": "LABEL_39",
52
+ "40": "LABEL_40",
53
+ "41": "LABEL_41",
54
+ "42": "LABEL_42",
55
+ "43": "LABEL_43",
56
+ "44": "LABEL_44",
57
+ "45": "LABEL_45",
58
+ "46": "LABEL_46",
59
+ "47": "LABEL_47",
60
+ "48": "LABEL_48",
61
+ "49": "LABEL_49",
62
+ "50": "LABEL_50",
63
+ "51": "LABEL_51",
64
+ "52": "LABEL_52",
65
+ "53": "LABEL_53",
66
+ "54": "LABEL_54",
67
+ "55": "LABEL_55",
68
+ "56": "LABEL_56",
69
+ "57": "LABEL_57",
70
+ "58": "LABEL_58",
71
+ "59": "LABEL_59",
72
+ "60": "LABEL_60",
73
+ "61": "LABEL_61",
74
+ "62": "LABEL_62",
75
+ "63": "LABEL_63",
76
+ "64": "LABEL_64",
77
+ "65": "LABEL_65",
78
+ "66": "LABEL_66",
79
+ "67": "LABEL_67",
80
+ "68": "LABEL_68",
81
+ "69": "LABEL_69",
82
+ "70": "LABEL_70",
83
+ "71": "LABEL_71",
84
+ "72": "LABEL_72",
85
+ "73": "LABEL_73",
86
+ "74": "LABEL_74",
87
+ "75": "LABEL_75",
88
+ "76": "LABEL_76",
89
+ "77": "LABEL_77",
90
+ "78": "LABEL_78",
91
+ "79": "LABEL_79",
92
+ "80": "LABEL_80",
93
+ "81": "LABEL_81",
94
+ "82": "LABEL_82",
95
+ "83": "LABEL_83",
96
+ "84": "LABEL_84",
97
+ "85": "LABEL_85",
98
+ "86": "LABEL_86",
99
+ "87": "LABEL_87",
100
+ "88": "LABEL_88",
101
+ "89": "LABEL_89",
102
+ "90": "LABEL_90",
103
+ "91": "LABEL_91",
104
+ "92": "LABEL_92",
105
+ "93": "LABEL_93",
106
+ "94": "LABEL_94",
107
+ "95": "LABEL_95",
108
+ "96": "LABEL_96",
109
+ "97": "LABEL_97",
110
+ "98": "LABEL_98",
111
+ "99": "LABEL_99",
112
+ "100": "LABEL_100",
113
+ "101": "LABEL_101",
114
+ "102": "LABEL_102",
115
+ "103": "LABEL_103",
116
+ "104": "LABEL_104",
117
+ "105": "LABEL_105",
118
+ "106": "LABEL_106",
119
+ "107": "LABEL_107",
120
+ "108": "LABEL_108",
121
+ "109": "LABEL_109",
122
+ "110": "LABEL_110",
123
+ "111": "LABEL_111",
124
+ "112": "LABEL_112",
125
+ "113": "LABEL_113",
126
+ "114": "LABEL_114",
127
+ "115": "LABEL_115",
128
+ "116": "LABEL_116",
129
+ "117": "LABEL_117",
130
+ "118": "LABEL_118",
131
+ "119": "LABEL_119",
132
+ "120": "LABEL_120",
133
+ "121": "LABEL_121",
134
+ "122": "LABEL_122",
135
+ "123": "LABEL_123",
136
+ "124": "LABEL_124",
137
+ "125": "LABEL_125",
138
+ "126": "LABEL_126",
139
+ "127": "LABEL_127",
140
+ "128": "LABEL_128",
141
+ "129": "LABEL_129",
142
+ "130": "LABEL_130",
143
+ "131": "LABEL_131",
144
+ "132": "LABEL_132",
145
+ "133": "LABEL_133",
146
+ "134": "LABEL_134",
147
+ "135": "LABEL_135",
148
+ "136": "LABEL_136",
149
+ "137": "LABEL_137",
150
+ "138": "LABEL_138",
151
+ "139": "LABEL_139",
152
+ "140": "LABEL_140",
153
+ "141": "LABEL_141",
154
+ "142": "LABEL_142",
155
+ "143": "LABEL_143",
156
+ "144": "LABEL_144",
157
+ "145": "LABEL_145",
158
+ "146": "LABEL_146",
159
+ "147": "LABEL_147",
160
+ "148": "LABEL_148",
161
+ "149": "LABEL_149",
162
+ "150": "LABEL_150"
163
+ },
164
+ "initializer_range": 0.02,
165
+ "label2id": {
166
+ "LABEL_0": 0,
167
+ "LABEL_1": 1,
168
+ "LABEL_10": 10,
169
+ "LABEL_100": 100,
170
+ "LABEL_101": 101,
171
+ "LABEL_102": 102,
172
+ "LABEL_103": 103,
173
+ "LABEL_104": 104,
174
+ "LABEL_105": 105,
175
+ "LABEL_106": 106,
176
+ "LABEL_107": 107,
177
+ "LABEL_108": 108,
178
+ "LABEL_109": 109,
179
+ "LABEL_11": 11,
180
+ "LABEL_110": 110,
181
+ "LABEL_111": 111,
182
+ "LABEL_112": 112,
183
+ "LABEL_113": 113,
184
+ "LABEL_114": 114,
185
+ "LABEL_115": 115,
186
+ "LABEL_116": 116,
187
+ "LABEL_117": 117,
188
+ "LABEL_118": 118,
189
+ "LABEL_119": 119,
190
+ "LABEL_12": 12,
191
+ "LABEL_120": 120,
192
+ "LABEL_121": 121,
193
+ "LABEL_122": 122,
194
+ "LABEL_123": 123,
195
+ "LABEL_124": 124,
196
+ "LABEL_125": 125,
197
+ "LABEL_126": 126,
198
+ "LABEL_127": 127,
199
+ "LABEL_128": 128,
200
+ "LABEL_129": 129,
201
+ "LABEL_13": 13,
202
+ "LABEL_130": 130,
203
+ "LABEL_131": 131,
204
+ "LABEL_132": 132,
205
+ "LABEL_133": 133,
206
+ "LABEL_134": 134,
207
+ "LABEL_135": 135,
208
+ "LABEL_136": 136,
209
+ "LABEL_137": 137,
210
+ "LABEL_138": 138,
211
+ "LABEL_139": 139,
212
+ "LABEL_14": 14,
213
+ "LABEL_140": 140,
214
+ "LABEL_141": 141,
215
+ "LABEL_142": 142,
216
+ "LABEL_143": 143,
217
+ "LABEL_144": 144,
218
+ "LABEL_145": 145,
219
+ "LABEL_146": 146,
220
+ "LABEL_147": 147,
221
+ "LABEL_148": 148,
222
+ "LABEL_149": 149,
223
+ "LABEL_15": 15,
224
+ "LABEL_150": 150,
225
+ "LABEL_16": 16,
226
+ "LABEL_17": 17,
227
+ "LABEL_18": 18,
228
+ "LABEL_19": 19,
229
+ "LABEL_2": 2,
230
+ "LABEL_20": 20,
231
+ "LABEL_21": 21,
232
+ "LABEL_22": 22,
233
+ "LABEL_23": 23,
234
+ "LABEL_24": 24,
235
+ "LABEL_25": 25,
236
+ "LABEL_26": 26,
237
+ "LABEL_27": 27,
238
+ "LABEL_28": 28,
239
+ "LABEL_29": 29,
240
+ "LABEL_3": 3,
241
+ "LABEL_30": 30,
242
+ "LABEL_31": 31,
243
+ "LABEL_32": 32,
244
+ "LABEL_33": 33,
245
+ "LABEL_34": 34,
246
+ "LABEL_35": 35,
247
+ "LABEL_36": 36,
248
+ "LABEL_37": 37,
249
+ "LABEL_38": 38,
250
+ "LABEL_39": 39,
251
+ "LABEL_4": 4,
252
+ "LABEL_40": 40,
253
+ "LABEL_41": 41,
254
+ "LABEL_42": 42,
255
+ "LABEL_43": 43,
256
+ "LABEL_44": 44,
257
+ "LABEL_45": 45,
258
+ "LABEL_46": 46,
259
+ "LABEL_47": 47,
260
+ "LABEL_48": 48,
261
+ "LABEL_49": 49,
262
+ "LABEL_5": 5,
263
+ "LABEL_50": 50,
264
+ "LABEL_51": 51,
265
+ "LABEL_52": 52,
266
+ "LABEL_53": 53,
267
+ "LABEL_54": 54,
268
+ "LABEL_55": 55,
269
+ "LABEL_56": 56,
270
+ "LABEL_57": 57,
271
+ "LABEL_58": 58,
272
+ "LABEL_59": 59,
273
+ "LABEL_6": 6,
274
+ "LABEL_60": 60,
275
+ "LABEL_61": 61,
276
+ "LABEL_62": 62,
277
+ "LABEL_63": 63,
278
+ "LABEL_64": 64,
279
+ "LABEL_65": 65,
280
+ "LABEL_66": 66,
281
+ "LABEL_67": 67,
282
+ "LABEL_68": 68,
283
+ "LABEL_69": 69,
284
+ "LABEL_7": 7,
285
+ "LABEL_70": 70,
286
+ "LABEL_71": 71,
287
+ "LABEL_72": 72,
288
+ "LABEL_73": 73,
289
+ "LABEL_74": 74,
290
+ "LABEL_75": 75,
291
+ "LABEL_76": 76,
292
+ "LABEL_77": 77,
293
+ "LABEL_78": 78,
294
+ "LABEL_79": 79,
295
+ "LABEL_8": 8,
296
+ "LABEL_80": 80,
297
+ "LABEL_81": 81,
298
+ "LABEL_82": 82,
299
+ "LABEL_83": 83,
300
+ "LABEL_84": 84,
301
+ "LABEL_85": 85,
302
+ "LABEL_86": 86,
303
+ "LABEL_87": 87,
304
+ "LABEL_88": 88,
305
+ "LABEL_89": 89,
306
+ "LABEL_9": 9,
307
+ "LABEL_90": 90,
308
+ "LABEL_91": 91,
309
+ "LABEL_92": 92,
310
+ "LABEL_93": 93,
311
+ "LABEL_94": 94,
312
+ "LABEL_95": 95,
313
+ "LABEL_96": 96,
314
+ "LABEL_97": 97,
315
+ "LABEL_98": 98,
316
+ "LABEL_99": 99
317
+ },
318
+ "max_position_embeddings": 512,
319
+ "model_type": "distilbert",
320
+ "n_heads": 12,
321
+ "n_layers": 6,
322
+ "pad_token_id": 0,
323
+ "problem_type": "single_label_classification",
324
+ "qa_dropout": 0.1,
325
+ "seq_classif_dropout": 0.2,
326
+ "sinusoidal_pos_embds": false,
327
+ "tie_weights_": true,
328
+ "transformers_version": "4.57.0",
329
+ "vocab_size": 30522
330
+ }
run-4/checkpoint-2862/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e445044ffc616c171dcab30d6b102a811ddc556b2429e1ed6b18c9584ae9bcd
3
+ size 268290900
run-4/checkpoint-2862/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dabf199f065d99aa9d4bed5ef6b64ca025471ab43ec785f7dbae60d8bd571a80
3
+ size 536645835
run-4/checkpoint-2862/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d5526722d16ec930596f11e111486fb6ce3342314ef139ef124194300a8c10bd
3
+ size 14645
run-4/checkpoint-2862/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01d096fac789ed022e5bee1c20f0a12bd1365801b6087b9ba94b777b004aa708
3
+ size 1465