Hugging Face's logo

xshubhamx
/
tiny-mistral

Text Classification
Transformers
TensorBoard
Safetensors
mistral
Generated from Trainer
text-generation-inference
Inference Endpoints
Model card Files Metrics Community
xshubhamx commited on
Commit
c4904f3
•
1 Parent(s): 0cfdb5d

Upload folder using huggingface_hub

Browse files
Files changed (7) hide show
  1. training_checkpoints/checkpoint-7073/config.json +63 -0
  2. training_checkpoints/checkpoint-7073/model.safetensors +3 -0
  3. training_checkpoints/checkpoint-7073/optimizer.pt +3 -0
  4. training_checkpoints/checkpoint-7073/rng_state.pth +3 -0
  5. training_checkpoints/checkpoint-7073/scheduler.pt +3 -0
  6. training_checkpoints/checkpoint-7073/trainer_state.json +361 -0
  7. training_checkpoints/checkpoint-7073/training_args.bin +3 -0
training_checkpoints/checkpoint-7073/config.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "openaccess-ai-collective/tiny-mistral",
3
+ "architectures": [
4
+ "MistralForSequenceClassification"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 1,
8
+ "dropout_p": 0.1,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 512,
12
+ "id2label": {
13
+ "0": "Issue",
14
+ "1": "Court Discourse",
15
+ "2": "Conclusion",
16
+ "3": "Precedent Analysis",
17
+ "4": "Section Analysis",
18
+ "5": "Argument by Petitioner",
19
+ "6": "Fact",
20
+ "7": "Argument by Respondent",
21
+ "8": "Ratio",
22
+ "9": "Appellant",
23
+ "10": "Respondent",
24
+ "11": "Argument by Appellant",
25
+ "12": "Petitioner",
26
+ "13": "Judge",
27
+ "14": "Argument by Defendant"
28
+ },
29
+ "initializer_range": 0.02,
30
+ "intermediate_size": 14336,
31
+ "label2id": {
32
+ "Appellant": 9,
33
+ "Argument by Appellant": 11,
34
+ "Argument by Defendant": 14,
35
+ "Argument by Petitioner": 5,
36
+ "Argument by Respondent": 7,
37
+ "Conclusion": 2,
38
+ "Court Discourse": 1,
39
+ "Fact": 6,
40
+ "Issue": 0,
41
+ "Judge": 13,
42
+ "Petitioner": 12,
43
+ "Precedent Analysis": 3,
44
+ "Ratio": 8,
45
+ "Respondent": 10,
46
+ "Section Analysis": 4
47
+ },
48
+ "max_position_embeddings": 32768,
49
+ "model_type": "mistral",
50
+ "num_attention_heads": 16,
51
+ "num_hidden_layers": 8,
52
+ "num_key_value_heads": 4,
53
+ "pad_token_id": 32000,
54
+ "problem_type": "single_label_classification",
55
+ "rms_norm_eps": 1e-05,
56
+ "rope_theta": 10000.0,
57
+ "sliding_window": 4096,
58
+ "tie_word_embeddings": false,
59
+ "torch_dtype": "float32",
60
+ "transformers_version": "4.39.3",
61
+ "use_cache": true,
62
+ "vocab_size": 32001
63
+ }
training_checkpoints/checkpoint-7073/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bd6bbff49438f1bc609fd0062edab144a322b0578e3138db1ee15472b1b733b0
3
+ size 791226496
training_checkpoints/checkpoint-7073/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:52dc7a6130719cda9ca6815de685eb28db32cf34d1a730793c0eeeaadb761257
3
+ size 1582499834
training_checkpoints/checkpoint-7073/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ceca872404f155cf0d9853014eb18c0cc7e23cd52acf7a31c07149383dc6f9e
3
+ size 14244
training_checkpoints/checkpoint-7073/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17363695eb368ce0458bd0d062a4ea1888909c3415961f0a50ee5e01f371065a
3
+ size 1064
training_checkpoints/checkpoint-7073/trainer_state.json ADDED
@@ -0,0 +1,361 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.6236987905363891,
3
+ "best_model_checkpoint": "tiny-mistral/checkpoint-5144",
4
+ "epoch": 11.0,
5
+ "eval_steps": 500,
6
+ "global_step": 7073,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.78,
13
+ "grad_norm": 13.069233894348145,
14
+ "learning_rate": 4.7413167444271646e-05,
15
+ "loss": 1.4479,
16
+ "step": 500
17
+ },
18
+ {
19
+ "epoch": 1.0,
20
+ "eval_accuracy": 0.6498838109992254,
21
+ "eval_f1_macro": 0.45467871431664447,
22
+ "eval_f1_micro": 0.6498838109992254,
23
+ "eval_f1_weighted": 0.6214028764904677,
24
+ "eval_loss": 1.118202805519104,
25
+ "eval_macro_fpr": 0.039027492204575735,
26
+ "eval_macro_sensitivity": 0.4743680961365506,
27
+ "eval_macro_specificity": 0.9731278674075015,
28
+ "eval_precision": 0.6258370043663808,
29
+ "eval_precision_macro": 0.4712458606651466,
30
+ "eval_recall": 0.6498838109992254,
31
+ "eval_recall_macro": 0.4743680961365506,
32
+ "eval_runtime": 52.8829,
33
+ "eval_samples_per_second": 24.412,
34
+ "eval_steps_per_second": 3.063,
35
+ "eval_weighted_fpr": 0.03705525495982948,
36
+ "eval_weighted_sensitivity": 0.6498838109992254,
37
+ "eval_weighted_specificity": 0.9470342001132973,
38
+ "step": 643
39
+ },
40
+ {
41
+ "epoch": 1.56,
42
+ "grad_norm": 42.20966720581055,
43
+ "learning_rate": 4.482115085536548e-05,
44
+ "loss": 0.8133,
45
+ "step": 1000
46
+ },
47
+ {
48
+ "epoch": 2.0,
49
+ "eval_accuracy": 0.6986831913245546,
50
+ "eval_f1_macro": 0.5473679856051097,
51
+ "eval_f1_micro": 0.6986831913245546,
52
+ "eval_f1_weighted": 0.6969580552726153,
53
+ "eval_loss": 1.0854355096817017,
54
+ "eval_macro_fpr": 0.030488809177152586,
55
+ "eval_macro_sensitivity": 0.552765390116517,
56
+ "eval_macro_specificity": 0.9773020581125585,
57
+ "eval_precision": 0.7196764053623382,
58
+ "eval_precision_macro": 0.5876945028566362,
59
+ "eval_recall": 0.6986831913245546,
60
+ "eval_recall_macro": 0.552765390116517,
61
+ "eval_runtime": 93.8123,
62
+ "eval_samples_per_second": 13.762,
63
+ "eval_steps_per_second": 1.727,
64
+ "eval_weighted_fpr": 0.029883997848966736,
65
+ "eval_weighted_sensitivity": 0.6986831913245546,
66
+ "eval_weighted_specificity": 0.9608476803638234,
67
+ "step": 1286
68
+ },
69
+ {
70
+ "epoch": 2.33,
71
+ "grad_norm": 26.4113712310791,
72
+ "learning_rate": 4.223950233281493e-05,
73
+ "loss": 0.5592,
74
+ "step": 1500
75
+ },
76
+ {
77
+ "epoch": 3.0,
78
+ "eval_accuracy": 0.6986831913245546,
79
+ "eval_f1_macro": 0.6012751387344262,
80
+ "eval_f1_micro": 0.6986831913245546,
81
+ "eval_f1_weighted": 0.6997681579317561,
82
+ "eval_loss": 1.6113793849945068,
83
+ "eval_macro_fpr": 0.030358926871943602,
84
+ "eval_macro_sensitivity": 0.5880627120151358,
85
+ "eval_macro_specificity": 0.9773067719065487,
86
+ "eval_precision": 0.7107021565435693,
87
+ "eval_precision_macro": 0.6368062038606529,
88
+ "eval_recall": 0.6986831913245546,
89
+ "eval_recall_macro": 0.5880627120151358,
90
+ "eval_runtime": 96.7771,
91
+ "eval_samples_per_second": 13.34,
92
+ "eval_steps_per_second": 1.674,
93
+ "eval_weighted_fpr": 0.029883997848966736,
94
+ "eval_weighted_sensitivity": 0.6986831913245546,
95
+ "eval_weighted_specificity": 0.9609183872736765,
96
+ "step": 1929
97
+ },
98
+ {
99
+ "epoch": 3.11,
100
+ "grad_norm": 0.03971128165721893,
101
+ "learning_rate": 3.964748574390876e-05,
102
+ "loss": 0.3916,
103
+ "step": 2000
104
+ },
105
+ {
106
+ "epoch": 3.89,
107
+ "grad_norm": 13.533605575561523,
108
+ "learning_rate": 3.705546915500259e-05,
109
+ "loss": 0.2375,
110
+ "step": 2500
111
+ },
112
+ {
113
+ "epoch": 4.0,
114
+ "eval_accuracy": 0.6955848179705655,
115
+ "eval_f1_macro": 0.5698961254173763,
116
+ "eval_f1_micro": 0.6955848179705655,
117
+ "eval_f1_weighted": 0.6923294650516646,
118
+ "eval_loss": 1.7778925895690918,
119
+ "eval_macro_fpr": 0.030968414851007965,
120
+ "eval_macro_sensitivity": 0.5666678921166624,
121
+ "eval_macro_specificity": 0.9768109367511323,
122
+ "eval_precision": 0.7000704251293273,
123
+ "eval_precision_macro": 0.584042243736088,
124
+ "eval_recall": 0.6955848179705655,
125
+ "eval_recall_macro": 0.5666678921166624,
126
+ "eval_runtime": 94.9154,
127
+ "eval_samples_per_second": 13.602,
128
+ "eval_steps_per_second": 1.707,
129
+ "eval_weighted_fpr": 0.030312379483224065,
130
+ "eval_weighted_sensitivity": 0.6955848179705655,
131
+ "eval_weighted_specificity": 0.956579233296421,
132
+ "step": 2572
133
+ },
134
+ {
135
+ "epoch": 4.67,
136
+ "grad_norm": 0.2196340560913086,
137
+ "learning_rate": 3.446345256609643e-05,
138
+ "loss": 0.1586,
139
+ "step": 3000
140
+ },
141
+ {
142
+ "epoch": 5.0,
143
+ "eval_accuracy": 0.6948102246320682,
144
+ "eval_f1_macro": 0.5694948674292386,
145
+ "eval_f1_micro": 0.6948102246320682,
146
+ "eval_f1_weighted": 0.691666971595594,
147
+ "eval_loss": 2.175213575363159,
148
+ "eval_macro_fpr": 0.03156565589564158,
149
+ "eval_macro_sensitivity": 0.5798857481261962,
150
+ "eval_macro_specificity": 0.9769938659292313,
151
+ "eval_precision": 0.7011163809357478,
152
+ "eval_precision_macro": 0.5797124163320468,
153
+ "eval_recall": 0.6948102246320682,
154
+ "eval_recall_macro": 0.5798857481261962,
155
+ "eval_runtime": 96.1916,
156
+ "eval_samples_per_second": 13.421,
157
+ "eval_steps_per_second": 1.684,
158
+ "eval_weighted_fpr": 0.03042001235330451,
159
+ "eval_weighted_sensitivity": 0.6948102246320682,
160
+ "eval_weighted_specificity": 0.9600977643064038,
161
+ "step": 3215
162
+ },
163
+ {
164
+ "epoch": 5.44,
165
+ "grad_norm": 0.0076691824942827225,
166
+ "learning_rate": 3.1881804043545884e-05,
167
+ "loss": 0.0956,
168
+ "step": 3500
169
+ },
170
+ {
171
+ "epoch": 6.0,
172
+ "eval_accuracy": 0.7079783113865221,
173
+ "eval_f1_macro": 0.6114593419504739,
174
+ "eval_f1_micro": 0.7079783113865221,
175
+ "eval_f1_weighted": 0.710498142821041,
176
+ "eval_loss": 2.326085329055786,
177
+ "eval_macro_fpr": 0.029146593037922882,
178
+ "eval_macro_sensitivity": 0.6191389353797161,
179
+ "eval_macro_specificity": 0.9781721436893602,
180
+ "eval_precision": 0.7213132077015695,
181
+ "eval_precision_macro": 0.6169445235048693,
182
+ "eval_recall": 0.7079783113865221,
183
+ "eval_recall_macro": 0.6191389353797161,
184
+ "eval_runtime": 108.9589,
185
+ "eval_samples_per_second": 11.849,
186
+ "eval_steps_per_second": 1.487,
187
+ "eval_weighted_fpr": 0.028619145221285964,
188
+ "eval_weighted_sensitivity": 0.7079783113865221,
189
+ "eval_weighted_specificity": 0.9646038439538819,
190
+ "step": 3858
191
+ },
192
+ {
193
+ "epoch": 6.22,
194
+ "grad_norm": 0.003827617969363928,
195
+ "learning_rate": 2.928978745463971e-05,
196
+ "loss": 0.0623,
197
+ "step": 4000
198
+ },
199
+ {
200
+ "epoch": 7.0,
201
+ "grad_norm": 0.007828492671251297,
202
+ "learning_rate": 2.669777086573354e-05,
203
+ "loss": 0.044,
204
+ "step": 4500
205
+ },
206
+ {
207
+ "epoch": 7.0,
208
+ "eval_accuracy": 0.715724244771495,
209
+ "eval_f1_macro": 0.6014076418730009,
210
+ "eval_f1_micro": 0.715724244771495,
211
+ "eval_f1_weighted": 0.7131079309029965,
212
+ "eval_loss": 2.330761671066284,
213
+ "eval_macro_fpr": 0.028468870564731498,
214
+ "eval_macro_sensitivity": 0.5938828421663026,
215
+ "eval_macro_specificity": 0.9784561846662306,
216
+ "eval_precision": 0.7142878695652373,
217
+ "eval_precision_macro": 0.6184199211909682,
218
+ "eval_recall": 0.715724244771495,
219
+ "eval_recall_macro": 0.5938828421663026,
220
+ "eval_runtime": 106.6283,
221
+ "eval_samples_per_second": 12.107,
222
+ "eval_steps_per_second": 1.519,
223
+ "eval_weighted_fpr": 0.027587762158911525,
224
+ "eval_weighted_sensitivity": 0.715724244771495,
225
+ "eval_weighted_specificity": 0.9611185252219651,
226
+ "step": 4501
227
+ },
228
+ {
229
+ "epoch": 7.78,
230
+ "grad_norm": 0.0054510668851435184,
231
+ "learning_rate": 2.4105754276827372e-05,
232
+ "loss": 0.0212,
233
+ "step": 5000
234
+ },
235
+ {
236
+ "epoch": 8.0,
237
+ "eval_accuracy": 0.7126258714175058,
238
+ "eval_f1_macro": 0.6236987905363891,
239
+ "eval_f1_micro": 0.7126258714175058,
240
+ "eval_f1_weighted": 0.7047150280912466,
241
+ "eval_loss": 2.5606603622436523,
242
+ "eval_macro_fpr": 0.029426308648760717,
243
+ "eval_macro_sensitivity": 0.6174588332474265,
244
+ "eval_macro_specificity": 0.9780489131013995,
245
+ "eval_precision": 0.7033345220138081,
246
+ "eval_precision_macro": 0.6494012124714204,
247
+ "eval_recall": 0.7126258714175058,
248
+ "eval_recall_macro": 0.6174588332474265,
249
+ "eval_runtime": 94.9282,
250
+ "eval_samples_per_second": 13.6,
251
+ "eval_steps_per_second": 1.707,
252
+ "eval_weighted_fpr": 0.027997886951928157,
253
+ "eval_weighted_sensitivity": 0.7126258714175058,
254
+ "eval_weighted_specificity": 0.9581078251034848,
255
+ "step": 5144
256
+ },
257
+ {
258
+ "epoch": 8.55,
259
+ "grad_norm": 0.002751357154920697,
260
+ "learning_rate": 2.1513737687921205e-05,
261
+ "loss": 0.0183,
262
+ "step": 5500
263
+ },
264
+ {
265
+ "epoch": 9.0,
266
+ "eval_accuracy": 0.7118512780790085,
267
+ "eval_f1_macro": 0.5935171263710702,
268
+ "eval_f1_micro": 0.7118512780790084,
269
+ "eval_f1_weighted": 0.7088386468311138,
270
+ "eval_loss": 2.640535354614258,
271
+ "eval_macro_fpr": 0.02906658876848228,
272
+ "eval_macro_sensitivity": 0.5850361401718444,
273
+ "eval_macro_specificity": 0.978116781696577,
274
+ "eval_precision": 0.7091923605282856,
275
+ "eval_precision_macro": 0.6132653330348934,
276
+ "eval_recall": 0.7118512780790085,
277
+ "eval_recall_macro": 0.5850361401718444,
278
+ "eval_runtime": 97.5564,
279
+ "eval_samples_per_second": 13.233,
280
+ "eval_steps_per_second": 1.661,
281
+ "eval_weighted_fpr": 0.02810092158936395,
282
+ "eval_weighted_sensitivity": 0.7118512780790085,
283
+ "eval_weighted_specificity": 0.9599004473696481,
284
+ "step": 5787
285
+ },
286
+ {
287
+ "epoch": 9.33,
288
+ "grad_norm": 0.002242760267108679,
289
+ "learning_rate": 1.8921721099015034e-05,
290
+ "loss": 0.0145,
291
+ "step": 6000
292
+ },
293
+ {
294
+ "epoch": 10.0,
295
+ "eval_accuracy": 0.7087529047250194,
296
+ "eval_f1_macro": 0.6038805077899971,
297
+ "eval_f1_micro": 0.7087529047250194,
298
+ "eval_f1_weighted": 0.7051494126143734,
299
+ "eval_loss": 2.7267849445343018,
300
+ "eval_macro_fpr": 0.0296548766519362,
301
+ "eval_macro_sensitivity": 0.594490599401291,
302
+ "eval_macro_specificity": 0.9777462293254275,
303
+ "eval_precision": 0.7057517403420687,
304
+ "eval_precision_macro": 0.6235388470549638,
305
+ "eval_recall": 0.7087529047250194,
306
+ "eval_recall_macro": 0.594490599401291,
307
+ "eval_runtime": 109.4092,
308
+ "eval_samples_per_second": 11.8,
309
+ "eval_steps_per_second": 1.481,
310
+ "eval_weighted_fpr": 0.028515091763992112,
311
+ "eval_weighted_sensitivity": 0.7087529047250194,
312
+ "eval_weighted_specificity": 0.9574405351563919,
313
+ "step": 6430
314
+ },
315
+ {
316
+ "epoch": 10.11,
317
+ "grad_norm": 0.001886312267743051,
318
+ "learning_rate": 1.6329704510108863e-05,
319
+ "loss": 0.0052,
320
+ "step": 6500
321
+ },
322
+ {
323
+ "epoch": 10.89,
324
+ "grad_norm": 0.002035327022895217,
325
+ "learning_rate": 1.374287195438051e-05,
326
+ "loss": 0.0065,
327
+ "step": 7000
328
+ },
329
+ {
330
+ "epoch": 11.0,
331
+ "eval_accuracy": 0.7149496514329977,
332
+ "eval_f1_macro": 0.6067795155065238,
333
+ "eval_f1_micro": 0.7149496514329978,
334
+ "eval_f1_weighted": 0.7122837081983867,
335
+ "eval_loss": 2.756761312484741,
336
+ "eval_macro_fpr": 0.028612398833029382,
337
+ "eval_macro_sensitivity": 0.5966419871344679,
338
+ "eval_macro_specificity": 0.9783906945848296,
339
+ "eval_precision": 0.7133091176874246,
340
+ "eval_precision_macro": 0.6341724921390148,
341
+ "eval_recall": 0.7149496514329977,
342
+ "eval_recall_macro": 0.5966419871344679,
343
+ "eval_runtime": 100.225,
344
+ "eval_samples_per_second": 12.881,
345
+ "eval_steps_per_second": 1.616,
346
+ "eval_weighted_fpr": 0.027689992475545523,
347
+ "eval_weighted_sensitivity": 0.7149496514329977,
348
+ "eval_weighted_specificity": 0.9609107673394471,
349
+ "step": 7073
350
+ }
351
+ ],
352
+ "logging_steps": 500,
353
+ "max_steps": 9645,
354
+ "num_input_tokens_seen": 0,
355
+ "num_train_epochs": 15,
356
+ "save_steps": 500,
357
+ "total_flos": 3.149874391626547e+16,
358
+ "train_batch_size": 8,
359
+ "trial_name": null,
360
+ "trial_params": null
361
+ }
training_checkpoints/checkpoint-7073/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ae13135576265f1180bcf8daa3dbfb279a0746197d617e38954842ee54fec7e1
3
+ size 4920