Hugging Face's logo

16arpi
/
stackoverflow

TensorBoard
Safetensors
bert
Model card Files
xet
 Metrics Community
16arpi commited on
Commit
7b06d07
·
verified ·
1 Parent(s): aa84a18

Upload folder using huggingface_hub

Browse files
Files changed (23) hide show
  1. checkpoint-1000/config.json +270 -0
  2. checkpoint-1000/model.safetensors +3 -0
  3. checkpoint-1000/optimizer.pt +3 -0
  4. checkpoint-1000/rng_state.pth +3 -0
  5. checkpoint-1000/scheduler.pt +3 -0
  6. checkpoint-1000/trainer_state.json +102 -0
  7. checkpoint-1000/training_args.bin +3 -0
  8. checkpoint-1490/config.json +270 -0
  9. checkpoint-1490/model.safetensors +3 -0
  10. checkpoint-1490/optimizer.pt +3 -0
  11. checkpoint-1490/rng_state.pth +3 -0
  12. checkpoint-1490/scheduler.pt +3 -0
  13. checkpoint-1490/trainer_state.json +129 -0
  14. checkpoint-1490/training_args.bin +3 -0
  15. checkpoint-500/config.json +270 -0
  16. checkpoint-500/model.safetensors +3 -0
  17. checkpoint-500/optimizer.pt +3 -0
  18. checkpoint-500/rng_state.pth +3 -0
  19. checkpoint-500/scheduler.pt +3 -0
  20. checkpoint-500/trainer_state.json +68 -0
  21. checkpoint-500/training_args.bin +3 -0
  22. model.zip +3 -0
  23. runs/May12_12-00-02_cs-01jv249ybzat2v926zrdde5asg/events.out.tfevents.1747051203.cs-01jv249ybzat2v926zrdde5asg.28951.0 +3 -0
checkpoint-1000/config.json ADDED
@@ -0,0 +1,270 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "gradient_checkpointing": false,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
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
+ },
133
+ "initializer_range": 0.02,
134
+ "intermediate_size": 3072,
135
+ "label2id": {
136
+ "LABEL_0": 0,
137
+ "LABEL_1": 1,
138
+ "LABEL_10": 10,
139
+ "LABEL_100": 100,
140
+ "LABEL_101": 101,
141
+ "LABEL_102": 102,
142
+ "LABEL_103": 103,
143
+ "LABEL_104": 104,
144
+ "LABEL_105": 105,
145
+ "LABEL_106": 106,
146
+ "LABEL_107": 107,
147
+ "LABEL_108": 108,
148
+ "LABEL_109": 109,
149
+ "LABEL_11": 11,
150
+ "LABEL_110": 110,
151
+ "LABEL_111": 111,
152
+ "LABEL_112": 112,
153
+ "LABEL_113": 113,
154
+ "LABEL_114": 114,
155
+ "LABEL_115": 115,
156
+ "LABEL_116": 116,
157
+ "LABEL_117": 117,
158
+ "LABEL_118": 118,
159
+ "LABEL_119": 119,
160
+ "LABEL_12": 12,
161
+ "LABEL_13": 13,
162
+ "LABEL_14": 14,
163
+ "LABEL_15": 15,
164
+ "LABEL_16": 16,
165
+ "LABEL_17": 17,
166
+ "LABEL_18": 18,
167
+ "LABEL_19": 19,
168
+ "LABEL_2": 2,
169
+ "LABEL_20": 20,
170
+ "LABEL_21": 21,
171
+ "LABEL_22": 22,
172
+ "LABEL_23": 23,
173
+ "LABEL_24": 24,
174
+ "LABEL_25": 25,
175
+ "LABEL_26": 26,
176
+ "LABEL_27": 27,
177
+ "LABEL_28": 28,
178
+ "LABEL_29": 29,
179
+ "LABEL_3": 3,
180
+ "LABEL_30": 30,
181
+ "LABEL_31": 31,
182
+ "LABEL_32": 32,
183
+ "LABEL_33": 33,
184
+ "LABEL_34": 34,
185
+ "LABEL_35": 35,
186
+ "LABEL_36": 36,
187
+ "LABEL_37": 37,
188
+ "LABEL_38": 38,
189
+ "LABEL_39": 39,
190
+ "LABEL_4": 4,
191
+ "LABEL_40": 40,
192
+ "LABEL_41": 41,
193
+ "LABEL_42": 42,
194
+ "LABEL_43": 43,
195
+ "LABEL_44": 44,
196
+ "LABEL_45": 45,
197
+ "LABEL_46": 46,
198
+ "LABEL_47": 47,
199
+ "LABEL_48": 48,
200
+ "LABEL_49": 49,
201
+ "LABEL_5": 5,
202
+ "LABEL_50": 50,
203
+ "LABEL_51": 51,
204
+ "LABEL_52": 52,
205
+ "LABEL_53": 53,
206
+ "LABEL_54": 54,
207
+ "LABEL_55": 55,
208
+ "LABEL_56": 56,
209
+ "LABEL_57": 57,
210
+ "LABEL_58": 58,
211
+ "LABEL_59": 59,
212
+ "LABEL_6": 6,
213
+ "LABEL_60": 60,
214
+ "LABEL_61": 61,
215
+ "LABEL_62": 62,
216
+ "LABEL_63": 63,
217
+ "LABEL_64": 64,
218
+ "LABEL_65": 65,
219
+ "LABEL_66": 66,
220
+ "LABEL_67": 67,
221
+ "LABEL_68": 68,
222
+ "LABEL_69": 69,
223
+ "LABEL_7": 7,
224
+ "LABEL_70": 70,
225
+ "LABEL_71": 71,
226
+ "LABEL_72": 72,
227
+ "LABEL_73": 73,
228
+ "LABEL_74": 74,
229
+ "LABEL_75": 75,
230
+ "LABEL_76": 76,
231
+ "LABEL_77": 77,
232
+ "LABEL_78": 78,
233
+ "LABEL_79": 79,
234
+ "LABEL_8": 8,
235
+ "LABEL_80": 80,
236
+ "LABEL_81": 81,
237
+ "LABEL_82": 82,
238
+ "LABEL_83": 83,
239
+ "LABEL_84": 84,
240
+ "LABEL_85": 85,
241
+ "LABEL_86": 86,
242
+ "LABEL_87": 87,
243
+ "LABEL_88": 88,
244
+ "LABEL_89": 89,
245
+ "LABEL_9": 9,
246
+ "LABEL_90": 90,
247
+ "LABEL_91": 91,
248
+ "LABEL_92": 92,
249
+ "LABEL_93": 93,
250
+ "LABEL_94": 94,
251
+ "LABEL_95": 95,
252
+ "LABEL_96": 96,
253
+ "LABEL_97": 97,
254
+ "LABEL_98": 98,
255
+ "LABEL_99": 99
256
+ },
257
+ "layer_norm_eps": 1e-12,
258
+ "max_position_embeddings": 512,
259
+ "model_type": "bert",
260
+ "num_attention_heads": 12,
261
+ "num_hidden_layers": 12,
262
+ "pad_token_id": 0,
263
+ "position_embedding_type": "absolute",
264
+ "problem_type": "single_label_classification",
265
+ "torch_dtype": "float32",
266
+ "transformers_version": "4.51.3",
267
+ "type_vocab_size": 2,
268
+ "use_cache": true,
269
+ "vocab_size": 28996
270
+ }
checkpoint-1000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:055d9f192e1fa685d1e4be62271d3f5f4545c5e319ac3bde7c39960e82954fb5
3
+ size 433633736
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ea60c0a9f1c61d51658d5ec172528f2dd6239d2f0331589f190fdae3d951d9d
3
+ size 867388939
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89267cfe1d44351a0db81091129bc70f292915f4f7cb67d4950ccb9ad396bb1e
3
+ size 14645
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c8bc86a9ceda8ab793daa11edd16333e706a7c719c7ebef19e2fe7f7b43b3c0
3
+ size 1465
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,102 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 6.7114093959731544,
6
+ "eval_steps": 500,
7
+ "global_step": 1000,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "eval_accuracy": 0.5050505050505051,
15
+ "eval_loss": 2.2999494075775146,
16
+ "eval_runtime": 8.1743,
17
+ "eval_samples_per_second": 36.333,
18
+ "eval_steps_per_second": 4.649,
19
+ "step": 149
20
+ },
21
+ {
22
+ "epoch": 2.0,
23
+ "eval_accuracy": 0.6195286195286195,
24
+ "eval_loss": 1.9136502742767334,
25
+ "eval_runtime": 8.1758,
26
+ "eval_samples_per_second": 36.327,
27
+ "eval_steps_per_second": 4.648,
28
+ "step": 298
29
+ },
30
+ {
31
+ "epoch": 3.0,
32
+ "eval_accuracy": 0.6599326599326599,
33
+ "eval_loss": 1.7068243026733398,
34
+ "eval_runtime": 8.1915,
35
+ "eval_samples_per_second": 36.257,
36
+ "eval_steps_per_second": 4.639,
37
+ "step": 447
38
+ },
39
+ {
40
+ "epoch": 3.3557046979865772,
41
+ "grad_norm": 15.295294761657715,
42
+ "learning_rate": 3.3255033557046986e-05,
43
+ "loss": 2.0179,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 4.0,
48
+ "eval_accuracy": 0.6767676767676768,
49
+ "eval_loss": 1.5854746103286743,
50
+ "eval_runtime": 8.1951,
51
+ "eval_samples_per_second": 36.241,
52
+ "eval_steps_per_second": 4.637,
53
+ "step": 596
54
+ },
55
+ {
56
+ "epoch": 5.0,
57
+ "eval_accuracy": 0.6868686868686869,
58
+ "eval_loss": 1.5601825714111328,
59
+ "eval_runtime": 8.1968,
60
+ "eval_samples_per_second": 36.233,
61
+ "eval_steps_per_second": 4.636,
62
+ "step": 745
63
+ },
64
+ {
65
+ "epoch": 6.0,
66
+ "eval_accuracy": 0.7037037037037037,
67
+ "eval_loss": 1.5054125785827637,
68
+ "eval_runtime": 8.1976,
69
+ "eval_samples_per_second": 36.23,
70
+ "eval_steps_per_second": 4.636,
71
+ "step": 894
72
+ },
73
+ {
74
+ "epoch": 6.7114093959731544,
75
+ "grad_norm": 7.224489212036133,
76
+ "learning_rate": 1.6476510067114094e-05,
77
+ "loss": 0.6927,
78
+ "step": 1000
79
+ }
80
+ ],
81
+ "logging_steps": 500,
82
+ "max_steps": 1490,
83
+ "num_input_tokens_seen": 0,
84
+ "num_train_epochs": 10,
85
+ "save_steps": 500,
86
+ "stateful_callbacks": {
87
+ "TrainerControl": {
88
+ "args": {
89
+ "should_epoch_stop": false,
90
+ "should_evaluate": false,
91
+ "should_log": false,
92
+ "should_save": true,
93
+ "should_training_stop": false
94
+ },
95
+ "attributes": {}
96
+ }
97
+ },
98
+ "total_flos": 2100797162717184.0,
99
+ "train_batch_size": 8,
100
+ "trial_name": null,
101
+ "trial_params": null
102
+ }
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef83674c87cf2d2222d7edf7ef7d46b26b92479e8b2f34efae7da644f1bfe77f
3
+ size 5713
checkpoint-1490/config.json ADDED
@@ -0,0 +1,270 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "gradient_checkpointing": false,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
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
+ },
133
+ "initializer_range": 0.02,
134
+ "intermediate_size": 3072,
135
+ "label2id": {
136
+ "LABEL_0": 0,
137
+ "LABEL_1": 1,
138
+ "LABEL_10": 10,
139
+ "LABEL_100": 100,
140
+ "LABEL_101": 101,
141
+ "LABEL_102": 102,
142
+ "LABEL_103": 103,
143
+ "LABEL_104": 104,
144
+ "LABEL_105": 105,
145
+ "LABEL_106": 106,
146
+ "LABEL_107": 107,
147
+ "LABEL_108": 108,
148
+ "LABEL_109": 109,
149
+ "LABEL_11": 11,
150
+ "LABEL_110": 110,
151
+ "LABEL_111": 111,
152
+ "LABEL_112": 112,
153
+ "LABEL_113": 113,
154
+ "LABEL_114": 114,
155
+ "LABEL_115": 115,
156
+ "LABEL_116": 116,
157
+ "LABEL_117": 117,
158
+ "LABEL_118": 118,
159
+ "LABEL_119": 119,
160
+ "LABEL_12": 12,
161
+ "LABEL_13": 13,
162
+ "LABEL_14": 14,
163
+ "LABEL_15": 15,
164
+ "LABEL_16": 16,
165
+ "LABEL_17": 17,
166
+ "LABEL_18": 18,
167
+ "LABEL_19": 19,
168
+ "LABEL_2": 2,
169
+ "LABEL_20": 20,
170
+ "LABEL_21": 21,
171
+ "LABEL_22": 22,
172
+ "LABEL_23": 23,
173
+ "LABEL_24": 24,
174
+ "LABEL_25": 25,
175
+ "LABEL_26": 26,
176
+ "LABEL_27": 27,
177
+ "LABEL_28": 28,
178
+ "LABEL_29": 29,
179
+ "LABEL_3": 3,
180
+ "LABEL_30": 30,
181
+ "LABEL_31": 31,
182
+ "LABEL_32": 32,
183
+ "LABEL_33": 33,
184
+ "LABEL_34": 34,
185
+ "LABEL_35": 35,
186
+ "LABEL_36": 36,
187
+ "LABEL_37": 37,
188
+ "LABEL_38": 38,
189
+ "LABEL_39": 39,
190
+ "LABEL_4": 4,
191
+ "LABEL_40": 40,
192
+ "LABEL_41": 41,
193
+ "LABEL_42": 42,
194
+ "LABEL_43": 43,
195
+ "LABEL_44": 44,
196
+ "LABEL_45": 45,
197
+ "LABEL_46": 46,
198
+ "LABEL_47": 47,
199
+ "LABEL_48": 48,
200
+ "LABEL_49": 49,
201
+ "LABEL_5": 5,
202
+ "LABEL_50": 50,
203
+ "LABEL_51": 51,
204
+ "LABEL_52": 52,
205
+ "LABEL_53": 53,
206
+ "LABEL_54": 54,
207
+ "LABEL_55": 55,
208
+ "LABEL_56": 56,
209
+ "LABEL_57": 57,
210
+ "LABEL_58": 58,
211
+ "LABEL_59": 59,
212
+ "LABEL_6": 6,
213
+ "LABEL_60": 60,
214
+ "LABEL_61": 61,
215
+ "LABEL_62": 62,
216
+ "LABEL_63": 63,
217
+ "LABEL_64": 64,
218
+ "LABEL_65": 65,
219
+ "LABEL_66": 66,
220
+ "LABEL_67": 67,
221
+ "LABEL_68": 68,
222
+ "LABEL_69": 69,
223
+ "LABEL_7": 7,
224
+ "LABEL_70": 70,
225
+ "LABEL_71": 71,
226
+ "LABEL_72": 72,
227
+ "LABEL_73": 73,
228
+ "LABEL_74": 74,
229
+ "LABEL_75": 75,
230
+ "LABEL_76": 76,
231
+ "LABEL_77": 77,
232
+ "LABEL_78": 78,
233
+ "LABEL_79": 79,
234
+ "LABEL_8": 8,
235
+ "LABEL_80": 80,
236
+ "LABEL_81": 81,
237
+ "LABEL_82": 82,
238
+ "LABEL_83": 83,
239
+ "LABEL_84": 84,
240
+ "LABEL_85": 85,
241
+ "LABEL_86": 86,
242
+ "LABEL_87": 87,
243
+ "LABEL_88": 88,
244
+ "LABEL_89": 89,
245
+ "LABEL_9": 9,
246
+ "LABEL_90": 90,
247
+ "LABEL_91": 91,
248
+ "LABEL_92": 92,
249
+ "LABEL_93": 93,
250
+ "LABEL_94": 94,
251
+ "LABEL_95": 95,
252
+ "LABEL_96": 96,
253
+ "LABEL_97": 97,
254
+ "LABEL_98": 98,
255
+ "LABEL_99": 99
256
+ },
257
+ "layer_norm_eps": 1e-12,
258
+ "max_position_embeddings": 512,
259
+ "model_type": "bert",
260
+ "num_attention_heads": 12,
261
+ "num_hidden_layers": 12,
262
+ "pad_token_id": 0,
263
+ "position_embedding_type": "absolute",
264
+ "problem_type": "single_label_classification",
265
+ "torch_dtype": "float32",
266
+ "transformers_version": "4.51.3",
267
+ "type_vocab_size": 2,
268
+ "use_cache": true,
269
+ "vocab_size": 28996
270
+ }
checkpoint-1490/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b67cb98f52568a37c451f89bf7039ceb8c8d3601b70825ebb8ef90dd1e4080b8
3
+ size 433633736
checkpoint-1490/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62390149fcf8c7ea00ef16427cad03369eeaed863197e2dd4e3f187ce65e01bd
3
+ size 867388939
checkpoint-1490/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:782108381daf8f21ccde237392caf61b71b423b64cd36e01b5c7152cf359eaeb
3
+ size 14645
checkpoint-1490/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:454aeecc6ffe2dc61a5ee34e1c45ddc2a6fe12373249aecd7fc5201f77ef66e1
3
+ size 1465
checkpoint-1490/trainer_state.json ADDED
@@ -0,0 +1,129 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 10.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1490,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "eval_accuracy": 0.5050505050505051,
15
+ "eval_loss": 2.2999494075775146,
16
+ "eval_runtime": 8.1743,
17
+ "eval_samples_per_second": 36.333,
18
+ "eval_steps_per_second": 4.649,
19
+ "step": 149
20
+ },
21
+ {
22
+ "epoch": 2.0,
23
+ "eval_accuracy": 0.6195286195286195,
24
+ "eval_loss": 1.9136502742767334,
25
+ "eval_runtime": 8.1758,
26
+ "eval_samples_per_second": 36.327,
27
+ "eval_steps_per_second": 4.648,
28
+ "step": 298
29
+ },
30
+ {
31
+ "epoch": 3.0,
32
+ "eval_accuracy": 0.6599326599326599,
33
+ "eval_loss": 1.7068243026733398,
34
+ "eval_runtime": 8.1915,
35
+ "eval_samples_per_second": 36.257,
36
+ "eval_steps_per_second": 4.639,
37
+ "step": 447
38
+ },
39
+ {
40
+ "epoch": 3.3557046979865772,
41
+ "grad_norm": 15.295294761657715,
42
+ "learning_rate": 3.3255033557046986e-05,
43
+ "loss": 2.0179,
44
+ "step": 500
45
+ },
46
+ {
47
+ "epoch": 4.0,
48
+ "eval_accuracy": 0.6767676767676768,
49
+ "eval_loss": 1.5854746103286743,
50
+ "eval_runtime": 8.1951,
51
+ "eval_samples_per_second": 36.241,
52
+ "eval_steps_per_second": 4.637,
53
+ "step": 596
54
+ },
55
+ {
56
+ "epoch": 5.0,
57
+ "eval_accuracy": 0.6868686868686869,
58
+ "eval_loss": 1.5601825714111328,
59
+ "eval_runtime": 8.1968,
60
+ "eval_samples_per_second": 36.233,
61
+ "eval_steps_per_second": 4.636,
62
+ "step": 745
63
+ },
64
+ {
65
+ "epoch": 6.0,
66
+ "eval_accuracy": 0.7037037037037037,
67
+ "eval_loss": 1.5054125785827637,
68
+ "eval_runtime": 8.1976,
69
+ "eval_samples_per_second": 36.23,
70
+ "eval_steps_per_second": 4.636,
71
+ "step": 894
72
+ },
73
+ {
74
+ "epoch": 6.7114093959731544,
75
+ "grad_norm": 7.224489212036133,
76
+ "learning_rate": 1.6476510067114094e-05,
77
+ "loss": 0.6927,
78
+ "step": 1000
79
+ },
80
+ {
81
+ "epoch": 7.0,
82
+ "eval_accuracy": 0.7070707070707071,
83
+ "eval_loss": 1.506157636642456,
84
+ "eval_runtime": 8.2414,
85
+ "eval_samples_per_second": 36.038,
86
+ "eval_steps_per_second": 4.611,
87
+ "step": 1043
88
+ },
89
+ {
90
+ "epoch": 8.0,
91
+ "eval_accuracy": 0.7306397306397306,
92
+ "eval_loss": 1.4550477266311646,
93
+ "eval_runtime": 8.2258,
94
+ "eval_samples_per_second": 36.106,
95
+ "eval_steps_per_second": 4.62,
96
+ "step": 1192
97
+ },
98
+ {
99
+ "epoch": 9.0,
100
+ "eval_accuracy": 0.7272727272727273,
101
+ "eval_loss": 1.4771537780761719,
102
+ "eval_runtime": 8.2112,
103
+ "eval_samples_per_second": 36.17,
104
+ "eval_steps_per_second": 4.628,
105
+ "step": 1341
106
+ }
107
+ ],
108
+ "logging_steps": 500,
109
+ "max_steps": 1490,
110
+ "num_input_tokens_seen": 0,
111
+ "num_train_epochs": 10,
112
+ "save_steps": 500,
113
+ "stateful_callbacks": {
114
+ "TrainerControl": {
115
+ "args": {
116
+ "should_epoch_stop": false,
117
+ "should_evaluate": false,
118
+ "should_log": false,
119
+ "should_save": true,
120
+ "should_training_stop": true
121
+ },
122
+ "attributes": {}
123
+ }
124
+ },
125
+ "total_flos": 3129070999633920.0,
126
+ "train_batch_size": 8,
127
+ "trial_name": null,
128
+ "trial_params": null
129
+ }
checkpoint-1490/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef83674c87cf2d2222d7edf7ef7d46b26b92479e8b2f34efae7da644f1bfe77f
3
+ size 5713
checkpoint-500/config.json ADDED
@@ -0,0 +1,270 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "BertForSequenceClassification"
4
+ ],
5
+ "attention_probs_dropout_prob": 0.1,
6
+ "classifier_dropout": null,
7
+ "gradient_checkpointing": false,
8
+ "hidden_act": "gelu",
9
+ "hidden_dropout_prob": 0.1,
10
+ "hidden_size": 768,
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
+ },
133
+ "initializer_range": 0.02,
134
+ "intermediate_size": 3072,
135
+ "label2id": {
136
+ "LABEL_0": 0,
137
+ "LABEL_1": 1,
138
+ "LABEL_10": 10,
139
+ "LABEL_100": 100,
140
+ "LABEL_101": 101,
141
+ "LABEL_102": 102,
142
+ "LABEL_103": 103,
143
+ "LABEL_104": 104,
144
+ "LABEL_105": 105,
145
+ "LABEL_106": 106,
146
+ "LABEL_107": 107,
147
+ "LABEL_108": 108,
148
+ "LABEL_109": 109,
149
+ "LABEL_11": 11,
150
+ "LABEL_110": 110,
151
+ "LABEL_111": 111,
152
+ "LABEL_112": 112,
153
+ "LABEL_113": 113,
154
+ "LABEL_114": 114,
155
+ "LABEL_115": 115,
156
+ "LABEL_116": 116,
157
+ "LABEL_117": 117,
158
+ "LABEL_118": 118,
159
+ "LABEL_119": 119,
160
+ "LABEL_12": 12,
161
+ "LABEL_13": 13,
162
+ "LABEL_14": 14,
163
+ "LABEL_15": 15,
164
+ "LABEL_16": 16,
165
+ "LABEL_17": 17,
166
+ "LABEL_18": 18,
167
+ "LABEL_19": 19,
168
+ "LABEL_2": 2,
169
+ "LABEL_20": 20,
170
+ "LABEL_21": 21,
171
+ "LABEL_22": 22,
172
+ "LABEL_23": 23,
173
+ "LABEL_24": 24,
174
+ "LABEL_25": 25,
175
+ "LABEL_26": 26,
176
+ "LABEL_27": 27,
177
+ "LABEL_28": 28,
178
+ "LABEL_29": 29,
179
+ "LABEL_3": 3,
180
+ "LABEL_30": 30,
181
+ "LABEL_31": 31,
182
+ "LABEL_32": 32,
183
+ "LABEL_33": 33,
184
+ "LABEL_34": 34,
185
+ "LABEL_35": 35,
186
+ "LABEL_36": 36,
187
+ "LABEL_37": 37,
188
+ "LABEL_38": 38,
189
+ "LABEL_39": 39,
190
+ "LABEL_4": 4,
191
+ "LABEL_40": 40,
192
+ "LABEL_41": 41,
193
+ "LABEL_42": 42,
194
+ "LABEL_43": 43,
195
+ "LABEL_44": 44,
196
+ "LABEL_45": 45,
197
+ "LABEL_46": 46,
198
+ "LABEL_47": 47,
199
+ "LABEL_48": 48,
200
+ "LABEL_49": 49,
201
+ "LABEL_5": 5,
202
+ "LABEL_50": 50,
203
+ "LABEL_51": 51,
204
+ "LABEL_52": 52,
205
+ "LABEL_53": 53,
206
+ "LABEL_54": 54,
207
+ "LABEL_55": 55,
208
+ "LABEL_56": 56,
209
+ "LABEL_57": 57,
210
+ "LABEL_58": 58,
211
+ "LABEL_59": 59,
212
+ "LABEL_6": 6,
213
+ "LABEL_60": 60,
214
+ "LABEL_61": 61,
215
+ "LABEL_62": 62,
216
+ "LABEL_63": 63,
217
+ "LABEL_64": 64,
218
+ "LABEL_65": 65,
219
+ "LABEL_66": 66,
220
+ "LABEL_67": 67,
221
+ "LABEL_68": 68,
222
+ "LABEL_69": 69,
223
+ "LABEL_7": 7,
224
+ "LABEL_70": 70,
225
+ "LABEL_71": 71,
226
+ "LABEL_72": 72,
227
+ "LABEL_73": 73,
228
+ "LABEL_74": 74,
229
+ "LABEL_75": 75,
230
+ "LABEL_76": 76,
231
+ "LABEL_77": 77,
232
+ "LABEL_78": 78,
233
+ "LABEL_79": 79,
234
+ "LABEL_8": 8,
235
+ "LABEL_80": 80,
236
+ "LABEL_81": 81,
237
+ "LABEL_82": 82,
238
+ "LABEL_83": 83,
239
+ "LABEL_84": 84,
240
+ "LABEL_85": 85,
241
+ "LABEL_86": 86,
242
+ "LABEL_87": 87,
243
+ "LABEL_88": 88,
244
+ "LABEL_89": 89,
245
+ "LABEL_9": 9,
246
+ "LABEL_90": 90,
247
+ "LABEL_91": 91,
248
+ "LABEL_92": 92,
249
+ "LABEL_93": 93,
250
+ "LABEL_94": 94,
251
+ "LABEL_95": 95,
252
+ "LABEL_96": 96,
253
+ "LABEL_97": 97,
254
+ "LABEL_98": 98,
255
+ "LABEL_99": 99
256
+ },
257
+ "layer_norm_eps": 1e-12,
258
+ "max_position_embeddings": 512,
259
+ "model_type": "bert",
260
+ "num_attention_heads": 12,
261
+ "num_hidden_layers": 12,
262
+ "pad_token_id": 0,
263
+ "position_embedding_type": "absolute",
264
+ "problem_type": "single_label_classification",
265
+ "torch_dtype": "float32",
266
+ "transformers_version": "4.51.3",
267
+ "type_vocab_size": 2,
268
+ "use_cache": true,
269
+ "vocab_size": 28996
270
+ }
checkpoint-500/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1fd99681d983cff60268831e57463cc001e664004f2c7ea5ad7f7b876f680c95
3
+ size 433633736
checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b771e52e60367d431fab11d33c41c7cca038d92318100621de9e4421ea585448
3
+ size 867388939
checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e1327a39df0824f334c53ac5de142f83e7848832abd0f23ccde55b94e03f169a
3
+ size 14645
checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:34d9f5f95442001c974e6609391142d25efea68db946030395d1e33ae667f82f
3
+ size 1465
checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.3557046979865772,
6
+ "eval_steps": 500,
7
+ "global_step": 500,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "eval_accuracy": 0.5050505050505051,
15
+ "eval_loss": 2.2999494075775146,
16
+ "eval_runtime": 8.1743,
17
+ "eval_samples_per_second": 36.333,
18
+ "eval_steps_per_second": 4.649,
19
+ "step": 149
20
+ },
21
+ {
22
+ "epoch": 2.0,
23
+ "eval_accuracy": 0.6195286195286195,
24
+ "eval_loss": 1.9136502742767334,
25
+ "eval_runtime": 8.1758,
26
+ "eval_samples_per_second": 36.327,
27
+ "eval_steps_per_second": 4.648,
28
+ "step": 298
29
+ },
30
+ {
31
+ "epoch": 3.0,
32
+ "eval_accuracy": 0.6599326599326599,
33
+ "eval_loss": 1.7068243026733398,
34
+ "eval_runtime": 8.1915,
35
+ "eval_samples_per_second": 36.257,
36
+ "eval_steps_per_second": 4.639,
37
+ "step": 447
38
+ },
39
+ {
40
+ "epoch": 3.3557046979865772,
41
+ "grad_norm": 15.295294761657715,
42
+ "learning_rate": 3.3255033557046986e-05,
43
+ "loss": 2.0179,
44
+ "step": 500
45
+ }
46
+ ],
47
+ "logging_steps": 500,
48
+ "max_steps": 1490,
49
+ "num_input_tokens_seen": 0,
50
+ "num_train_epochs": 10,
51
+ "save_steps": 500,
52
+ "stateful_callbacks": {
53
+ "TrainerControl": {
54
+ "args": {
55
+ "should_epoch_stop": false,
56
+ "should_evaluate": false,
57
+ "should_log": false,
58
+ "should_save": true,
59
+ "should_training_stop": false
60
+ },
61
+ "attributes": {}
62
+ }
63
+ },
64
+ "total_flos": 1050398581358592.0,
65
+ "train_batch_size": 8,
66
+ "trial_name": null,
67
+ "trial_params": null
68
+ }
checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ef83674c87cf2d2222d7edf7ef7d46b26b92479e8b2f34efae7da644f1bfe77f
3
+ size 5713
model.zip ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0699fe5a19f01c70a7e4443d0536b79c3753ab143cfb64bd1cd122a3bc3c1349
3
+ size 1056059590
runs/May12_12-00-02_cs-01jv249ybzat2v926zrdde5asg/events.out.tfevents.1747051203.cs-01jv249ybzat2v926zrdde5asg.28951.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f32ddc5d0eb575f398aac027fd19aa2c153d48f87df951b0404fd45dad4c0ff1
3
+ size 14110