xshubhamx commited on
Commit
aff385f
1 Parent(s): 5537da3

Upload folder using huggingface_hub

Browse files
training_checkpoints/checkpoint-2250/README.md CHANGED
@@ -1,6 +1,6 @@
1
  ---
2
  library_name: peft
3
- base_model: facebook/bart-base
4
  ---
5
 
6
  # Model Card for Model ID
@@ -81,7 +81,7 @@ Use the code below to get started with the model.
81
 
82
  [More Information Needed]
83
 
84
- ### Training Procedure
85
 
86
  <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
 
@@ -197,8 +197,6 @@ Carbon emissions can be estimated using the [Machine Learning Impact calculator]
197
  ## Model Card Contact
198
 
199
  [More Information Needed]
200
-
201
-
202
  ### Framework versions
203
 
204
  - PEFT 0.10.0
 
1
  ---
2
  library_name: peft
3
+ base_model: facebook/bart-large
4
  ---
5
 
6
  # Model Card for Model ID
 
81
 
82
  [More Information Needed]
83
 
84
+ ### Training Procedure
85
 
86
  <!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
87
 
 
197
  ## Model Card Contact
198
 
199
  [More Information Needed]
 
 
200
  ### Framework versions
201
 
202
  - PEFT 0.10.0
training_checkpoints/checkpoint-2250/adapter_config.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
  "alpha_pattern": {},
3
  "auto_mapping": null,
4
- "base_model_name_or_path": "facebook/bart-base",
5
  "bias": "none",
6
  "fan_in_fan_out": false,
7
  "inference_mode": true,
@@ -20,13 +20,13 @@
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
23
- "dense",
24
- "v_proj",
25
  "fc1",
26
- "fc2",
27
- "out_proj",
28
  "q_proj",
29
- "k_proj"
 
 
 
 
30
  ],
31
  "task_type": "SEQ_CLS",
32
  "use_dora": false,
 
1
  {
2
  "alpha_pattern": {},
3
  "auto_mapping": null,
4
+ "base_model_name_or_path": "facebook/bart-large",
5
  "bias": "none",
6
  "fan_in_fan_out": false,
7
  "inference_mode": true,
 
20
  "rank_pattern": {},
21
  "revision": null,
22
  "target_modules": [
 
 
23
  "fc1",
 
 
24
  "q_proj",
25
+ "dense",
26
+ "k_proj",
27
+ "out_proj",
28
+ "fc2",
29
+ "v_proj"
30
  ],
31
  "task_type": "SEQ_CLS",
32
  "use_dora": false,
training_checkpoints/checkpoint-2250/adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2156c22df6454648fd71b35e07a6ef7863527e78c366bb4894646200a6fc4ea3
3
- size 13151448
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0bc0bc82733dcf987182e419e15aad2ccc549f57c5c716d8ab9808220652375c
3
+ size 34854640
training_checkpoints/checkpoint-2250/optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1fc500942f535cce4098032106a4a50139315ce5f6f1c5a52c1437ffd89131a9
3
- size 26413626
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85b862b533b9db43f863a0fe7ceafda9cf7c745c794e1025e1f8b0e4d213c93b
3
+ size 69927198
training_checkpoints/checkpoint-2250/rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4b2516983cbfa7c23f1e8a27e2206e31b023c26e78bf8458fddff912ab16ac65
3
  size 14244
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2b359903c3cdc3122a7c07934e0565b0f57a6739b41ecaf5ccb061620b204b17
3
  size 14244
training_checkpoints/checkpoint-2250/trainer_state.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
- "best_metric": 0.6261172643852387,
3
- "best_model_checkpoint": "bart-base-lora/checkpoint-2250",
4
  "epoch": 13.996889580093313,
5
  "eval_steps": 500,
6
  "global_step": 2250,
@@ -10,334 +10,334 @@
10
  "log_history": [
11
  {
12
  "epoch": 1.0,
13
- "eval_accuracy": 0.611154144074361,
14
- "eval_f1_macro": 0.28705545756004525,
15
- "eval_f1_micro": 0.611154144074361,
16
- "eval_f1_weighted": 0.5574695078229231,
17
- "eval_loss": 1.3204560279846191,
18
- "eval_macro_fpr": 0.046390243375546196,
19
- "eval_macro_sensitivity": 0.302378554842108,
20
- "eval_macro_specificity": 0.9691815961629444,
21
- "eval_precision": 0.5321554731456403,
22
- "eval_precision_macro": 0.2887439395071009,
23
- "eval_recall": 0.611154144074361,
24
- "eval_recall_macro": 0.302378554842108,
25
- "eval_runtime": 45.1674,
26
- "eval_samples_per_second": 28.583,
27
- "eval_steps_per_second": 3.587,
28
- "eval_weighted_fpr": 0.04347073086248701,
29
- "eval_weighted_sensitivity": 0.611154144074361,
30
- "eval_weighted_specificity": 0.9265697983698051,
31
  "step": 160
32
  },
33
  {
34
  "epoch": 2.0,
35
- "eval_accuracy": 0.6994577846630519,
36
- "eval_f1_macro": 0.3947617401663212,
37
- "eval_f1_micro": 0.6994577846630519,
38
- "eval_f1_weighted": 0.6807748226338182,
39
- "eval_loss": 0.887535810470581,
40
- "eval_macro_fpr": 0.03062685085400977,
41
- "eval_macro_sensitivity": 0.4253981018601158,
42
- "eval_macro_specificity": 0.977359054380045,
43
- "eval_precision": 0.6728416067370143,
44
- "eval_precision_macro": 0.3822291983692299,
45
- "eval_recall": 0.6994577846630519,
46
- "eval_recall_macro": 0.4253981018601158,
47
- "eval_runtime": 44.0045,
48
- "eval_samples_per_second": 29.338,
49
- "eval_steps_per_second": 3.681,
50
- "eval_weighted_fpr": 0.02977743668457406,
51
- "eval_weighted_sensitivity": 0.6994577846630519,
52
- "eval_weighted_specificity": 0.9609280310376243,
53
  "step": 321
54
  },
55
  {
56
  "epoch": 3.0,
57
- "eval_accuracy": 0.7064291247095275,
58
- "eval_f1_macro": 0.39687975393644837,
59
- "eval_f1_micro": 0.7064291247095275,
60
- "eval_f1_weighted": 0.6751887743731854,
61
- "eval_loss": 0.842707633972168,
62
- "eval_macro_fpr": 0.029525759570689643,
63
- "eval_macro_sensitivity": 0.44418263304043937,
64
- "eval_macro_specificity": 0.9780349023882488,
65
- "eval_precision": 0.6951917464608133,
66
- "eval_precision_macro": 0.4131313886422618,
67
- "eval_recall": 0.7064291247095275,
68
- "eval_recall_macro": 0.44418263304043937,
69
- "eval_runtime": 44.2756,
70
- "eval_samples_per_second": 29.158,
71
- "eval_steps_per_second": 3.659,
72
- "eval_weighted_fpr": 0.02882786947592607,
73
- "eval_weighted_sensitivity": 0.7064291247095275,
74
- "eval_weighted_specificity": 0.9640944111142056,
75
  "step": 482
76
  },
77
  {
78
  "epoch": 3.11,
79
  "learning_rate": 3.958333333333333e-05,
80
- "loss": 1.2895,
81
  "step": 500
82
  },
83
  {
84
  "epoch": 4.0,
85
- "eval_accuracy": 0.7273431448489543,
86
- "eval_f1_macro": 0.42839565617146974,
87
- "eval_f1_micro": 0.7273431448489543,
88
- "eval_f1_weighted": 0.7166656051521775,
89
- "eval_loss": 0.7718679904937744,
90
- "eval_macro_fpr": 0.026373610346420835,
91
- "eval_macro_sensitivity": 0.4597947061276887,
92
- "eval_macro_specificity": 0.9797578098165848,
93
- "eval_precision": 0.7132141916978711,
94
- "eval_precision_macro": 0.41981312370683244,
95
- "eval_recall": 0.7273431448489543,
96
- "eval_recall_macro": 0.4597947061276887,
97
- "eval_runtime": 44.0471,
98
- "eval_samples_per_second": 29.31,
99
- "eval_steps_per_second": 3.678,
100
- "eval_weighted_fpr": 0.026077937472218107,
101
- "eval_weighted_sensitivity": 0.7273431448489543,
102
- "eval_weighted_specificity": 0.9690240023998159,
103
  "step": 643
104
  },
105
  {
106
  "epoch": 5.0,
107
- "eval_accuracy": 0.750580945003873,
108
- "eval_f1_macro": 0.5071625387211971,
109
- "eval_f1_micro": 0.750580945003873,
110
- "eval_f1_weighted": 0.7368350875727435,
111
- "eval_loss": 0.7387559413909912,
112
- "eval_macro_fpr": 0.02390193870216418,
113
- "eval_macro_sensitivity": 0.5165232292614671,
114
- "eval_macro_specificity": 0.9813510652812688,
115
- "eval_precision": 0.739991301984922,
116
- "eval_precision_macro": 0.5733174750951799,
117
- "eval_recall": 0.750580945003873,
118
- "eval_recall_macro": 0.5165232292614671,
119
- "eval_runtime": 43.5191,
120
- "eval_samples_per_second": 29.665,
121
- "eval_steps_per_second": 3.723,
122
- "eval_weighted_fpr": 0.023185483870967742,
123
- "eval_weighted_sensitivity": 0.750580945003873,
124
- "eval_weighted_specificity": 0.9696850342151593,
125
  "step": 803
126
  },
127
  {
128
  "epoch": 6.0,
129
- "eval_accuracy": 0.7443841982958946,
130
- "eval_f1_macro": 0.508766573370445,
131
- "eval_f1_micro": 0.7443841982958946,
132
- "eval_f1_weighted": 0.726841801268282,
133
- "eval_loss": 0.7526118755340576,
134
- "eval_macro_fpr": 0.024741953499129225,
135
- "eval_macro_sensitivity": 0.5229557335240369,
136
- "eval_macro_specificity": 0.980899976318631,
137
- "eval_precision": 0.7337183019873953,
138
- "eval_precision_macro": 0.5703412270657953,
139
- "eval_recall": 0.7443841982958946,
140
- "eval_recall_macro": 0.5229557335240369,
141
- "eval_runtime": 43.7116,
142
- "eval_samples_per_second": 29.534,
143
- "eval_steps_per_second": 3.706,
144
- "eval_weighted_fpr": 0.023940800928612886,
145
- "eval_weighted_sensitivity": 0.7443841982958946,
146
- "eval_weighted_specificity": 0.9691154464835722,
147
  "step": 964
148
  },
149
  {
150
  "epoch": 6.22,
151
  "learning_rate": 2.916666666666667e-05,
152
- "loss": 0.7332,
153
  "step": 1000
154
  },
155
  {
156
  "epoch": 7.0,
157
- "eval_accuracy": 0.7552285050348567,
158
- "eval_f1_macro": 0.5608894910772386,
159
- "eval_f1_micro": 0.7552285050348567,
160
- "eval_f1_weighted": 0.7460771491633669,
161
- "eval_loss": 0.7082085609436035,
162
- "eval_macro_fpr": 0.02329269548734161,
163
- "eval_macro_sensitivity": 0.5727565701904757,
164
- "eval_macro_specificity": 0.9817624646436656,
165
- "eval_precision": 0.7436383222975608,
166
- "eval_precision_macro": 0.5665418243254761,
167
- "eval_recall": 0.7552285050348567,
168
- "eval_recall_macro": 0.5727565701904757,
169
- "eval_runtime": 43.8592,
170
- "eval_samples_per_second": 29.435,
171
- "eval_steps_per_second": 3.694,
172
- "eval_weighted_fpr": 0.02262637834741515,
173
- "eval_weighted_sensitivity": 0.7552285050348567,
174
- "eval_weighted_specificity": 0.9712084646201266,
175
  "step": 1125
176
  },
177
  {
178
  "epoch": 8.0,
179
- "eval_accuracy": 0.7583268783888458,
180
- "eval_f1_macro": 0.5755845408528306,
181
- "eval_f1_micro": 0.7583268783888458,
182
- "eval_f1_weighted": 0.7502605827076951,
183
- "eval_loss": 0.7160602807998657,
184
- "eval_macro_fpr": 0.022834923601538912,
185
- "eval_macro_sensitivity": 0.5974764193374731,
186
- "eval_macro_specificity": 0.9820281006961107,
187
- "eval_precision": 0.7489019370925815,
188
- "eval_precision_macro": 0.5641453776326693,
189
- "eval_recall": 0.7583268783888458,
190
- "eval_recall_macro": 0.5974764193374731,
191
- "eval_runtime": 47.4544,
192
- "eval_samples_per_second": 27.205,
193
- "eval_steps_per_second": 3.414,
194
- "eval_weighted_fpr": 0.022257098016835498,
195
- "eval_weighted_sensitivity": 0.7583268783888458,
196
- "eval_weighted_specificity": 0.9720946320528147,
197
  "step": 1286
198
  },
199
  {
200
  "epoch": 9.0,
201
- "eval_accuracy": 0.7776917118512781,
202
- "eval_f1_macro": 0.5874833340927881,
203
- "eval_f1_micro": 0.7776917118512781,
204
- "eval_f1_weighted": 0.7653015782806978,
205
- "eval_loss": 0.683149516582489,
206
- "eval_macro_fpr": 0.020831516213897273,
207
- "eval_macro_sensitivity": 0.606896137249408,
208
- "eval_macro_specificity": 0.9832816294136357,
209
- "eval_precision": 0.7586915140412427,
210
- "eval_precision_macro": 0.5780981033617572,
211
- "eval_recall": 0.7776917118512781,
212
- "eval_recall_macro": 0.606896137249408,
213
- "eval_runtime": 50.31,
214
- "eval_samples_per_second": 25.661,
215
- "eval_steps_per_second": 3.22,
216
- "eval_weighted_fpr": 0.02000976085895559,
217
- "eval_weighted_sensitivity": 0.7776917118512781,
218
- "eval_weighted_specificity": 0.9715327293532582,
219
  "step": 1446
220
  },
221
  {
222
  "epoch": 9.33,
223
  "learning_rate": 1.8750000000000002e-05,
224
- "loss": 0.6167,
225
  "step": 1500
226
  },
227
  {
228
  "epoch": 10.0,
229
- "eval_accuracy": 0.7862122385747483,
230
- "eval_f1_macro": 0.5987413950993845,
231
- "eval_f1_micro": 0.7862122385747482,
232
- "eval_f1_weighted": 0.7753590541094936,
233
- "eval_loss": 0.6683156490325928,
234
- "eval_macro_fpr": 0.019819293314469463,
235
- "eval_macro_sensitivity": 0.617364154098103,
236
- "eval_macro_specificity": 0.9839313497952282,
237
- "eval_precision": 0.7714272913975577,
238
- "eval_precision_macro": 0.5916700769971839,
239
- "eval_recall": 0.7862122385747483,
240
- "eval_recall_macro": 0.617364154098103,
241
- "eval_runtime": 46.3094,
242
- "eval_samples_per_second": 27.878,
243
- "eval_steps_per_second": 3.498,
244
- "eval_weighted_fpr": 0.019052878641446915,
245
- "eval_weighted_sensitivity": 0.7862122385747483,
246
- "eval_weighted_specificity": 0.972758008353676,
247
  "step": 1607
248
  },
249
  {
250
  "epoch": 11.0,
251
- "eval_accuracy": 0.7761425251742835,
252
- "eval_f1_macro": 0.5945805235654043,
253
- "eval_f1_micro": 0.7761425251742835,
254
- "eval_f1_weighted": 0.7642426809512617,
255
- "eval_loss": 0.6884666085243225,
256
- "eval_macro_fpr": 0.020959556505082782,
257
- "eval_macro_sensitivity": 0.6219516621776988,
258
- "eval_macro_specificity": 0.9832325403042509,
259
- "eval_precision": 0.762752032263232,
260
- "eval_precision_macro": 0.581742682634886,
261
- "eval_recall": 0.7761425251742835,
262
- "eval_recall_macro": 0.6219516621776988,
263
- "eval_runtime": 44.4944,
264
- "eval_samples_per_second": 29.015,
265
- "eval_steps_per_second": 3.641,
266
- "eval_weighted_fpr": 0.02018579311308235,
267
- "eval_weighted_sensitivity": 0.7761425251742835,
268
- "eval_weighted_specificity": 0.9723455793894815,
269
  "step": 1768
270
  },
271
  {
272
  "epoch": 12.0,
273
- "eval_accuracy": 0.7869868319132456,
274
- "eval_f1_macro": 0.6214234200731922,
275
- "eval_f1_micro": 0.7869868319132455,
276
- "eval_f1_weighted": 0.7764436224395941,
277
- "eval_loss": 0.6829659938812256,
278
- "eval_macro_fpr": 0.01971781574074696,
279
- "eval_macro_sensitivity": 0.6463770170214457,
280
- "eval_macro_specificity": 0.9840270123264785,
281
- "eval_precision": 0.7826408792667282,
282
- "eval_precision_macro": 0.6626773170335125,
283
- "eval_recall": 0.7869868319132456,
284
- "eval_recall_macro": 0.6463770170214457,
285
- "eval_runtime": 43.9944,
286
- "eval_samples_per_second": 29.345,
287
- "eval_steps_per_second": 3.682,
288
- "eval_weighted_fpr": 0.018966825298296436,
289
- "eval_weighted_sensitivity": 0.7869868319132456,
290
- "eval_weighted_specificity": 0.9734183529839338,
291
  "step": 1929
292
  },
293
  {
294
  "epoch": 12.44,
295
  "learning_rate": 8.333333333333334e-06,
296
- "loss": 0.5314,
297
  "step": 2000
298
  },
299
  {
300
  "epoch": 13.0,
301
- "eval_accuracy": 0.7916343919442292,
302
- "eval_f1_macro": 0.6110606025775643,
303
- "eval_f1_micro": 0.7916343919442292,
304
- "eval_f1_weighted": 0.7817623460591858,
305
- "eval_loss": 0.660524845123291,
306
- "eval_macro_fpr": 0.01915789146680066,
307
- "eval_macro_sensitivity": 0.6358134270226833,
308
- "eval_macro_specificity": 0.984380207900859,
309
- "eval_precision": 0.7769679723590022,
310
- "eval_precision_macro": 0.5964718375735727,
311
- "eval_recall": 0.7916343919442292,
312
- "eval_recall_macro": 0.6358134270226833,
313
- "eval_runtime": 44.1283,
314
- "eval_samples_per_second": 29.256,
315
- "eval_steps_per_second": 3.671,
316
- "eval_weighted_fpr": 0.01845372847636688,
317
- "eval_weighted_sensitivity": 0.7916343919442292,
318
- "eval_weighted_specificity": 0.9740687265686558,
319
  "step": 2089
320
  },
321
  {
322
  "epoch": 14.0,
323
- "eval_accuracy": 0.790859798605732,
324
- "eval_f1_macro": 0.6261172643852387,
325
- "eval_f1_micro": 0.790859798605732,
326
- "eval_f1_weighted": 0.780302820536599,
327
- "eval_loss": 0.6614473462104797,
328
- "eval_macro_fpr": 0.019344545099537428,
329
- "eval_macro_sensitivity": 0.6477733385910524,
330
- "eval_macro_specificity": 0.9842490309995476,
331
- "eval_precision": 0.7794397894982683,
332
- "eval_precision_macro": 0.6367821774875876,
333
- "eval_recall": 0.790859798605732,
334
- "eval_recall_macro": 0.6477733385910524,
335
- "eval_runtime": 52.6504,
336
- "eval_samples_per_second": 24.52,
337
- "eval_steps_per_second": 3.077,
338
- "eval_weighted_fpr": 0.018538862949739082,
339
- "eval_weighted_sensitivity": 0.790859798605732,
340
- "eval_weighted_specificity": 0.9728756663874838,
341
  "step": 2250
342
  }
343
  ],
@@ -345,7 +345,7 @@
345
  "max_steps": 2400,
346
  "num_train_epochs": 15,
347
  "save_steps": 500,
348
- "total_flos": 2.278788590234419e+16,
349
  "trial_name": null,
350
  "trial_params": null
351
  }
 
1
  {
2
+ "best_metric": 0.7393184269658342,
3
+ "best_model_checkpoint": "bart-base-lora/checkpoint-1125",
4
  "epoch": 13.996889580093313,
5
  "eval_steps": 500,
6
  "global_step": 2250,
 
10
  "log_history": [
11
  {
12
  "epoch": 1.0,
13
+ "eval_accuracy": 0.715724244771495,
14
+ "eval_f1_macro": 0.3957532686340496,
15
+ "eval_f1_micro": 0.715724244771495,
16
+ "eval_f1_weighted": 0.6834645556416842,
17
+ "eval_loss": 0.9525313973426819,
18
+ "eval_macro_fpr": 0.028501626692807605,
19
+ "eval_macro_sensitivity": 0.4415906678049911,
20
+ "eval_macro_specificity": 0.9786637894661117,
21
+ "eval_precision": 0.6787596592263883,
22
+ "eval_precision_macro": 0.38751339695744513,
23
+ "eval_recall": 0.715724244771495,
24
+ "eval_recall_macro": 0.4415906678049911,
25
+ "eval_runtime": 120.9594,
26
+ "eval_samples_per_second": 10.673,
27
+ "eval_steps_per_second": 1.339,
28
+ "eval_weighted_fpr": 0.027587762158911525,
29
+ "eval_weighted_sensitivity": 0.715724244771495,
30
+ "eval_weighted_specificity": 0.9642325972201805,
31
  "step": 160
32
  },
33
  {
34
  "epoch": 2.0,
35
+ "eval_accuracy": 0.7412858249419055,
36
+ "eval_f1_macro": 0.43370046411892155,
37
+ "eval_f1_micro": 0.7412858249419055,
38
+ "eval_f1_weighted": 0.7231329193382661,
39
+ "eval_loss": 0.7733433842658997,
40
+ "eval_macro_fpr": 0.025167370687133015,
41
+ "eval_macro_sensitivity": 0.4686708989603177,
42
+ "eval_macro_specificity": 0.9805374797463638,
43
+ "eval_precision": 0.729625861926521,
44
+ "eval_precision_macro": 0.4491323101141271,
45
+ "eval_recall": 0.7412858249419055,
46
+ "eval_recall_macro": 0.4686708989603177,
47
+ "eval_runtime": 119.8649,
48
+ "eval_samples_per_second": 10.77,
49
+ "eval_steps_per_second": 1.352,
50
+ "eval_weighted_fpr": 0.024322749781532187,
51
+ "eval_weighted_sensitivity": 0.7412858249419055,
52
+ "eval_weighted_specificity": 0.9667763712535515,
53
  "step": 321
54
  },
55
  {
56
  "epoch": 3.0,
57
+ "eval_accuracy": 0.7738187451587917,
58
+ "eval_f1_macro": 0.5270593942960143,
59
+ "eval_f1_micro": 0.7738187451587917,
60
+ "eval_f1_weighted": 0.7610767338978833,
61
+ "eval_loss": 0.7104586958885193,
62
+ "eval_macro_fpr": 0.021211805484311928,
63
+ "eval_macro_sensitivity": 0.5407847836684627,
64
+ "eval_macro_specificity": 0.9830897092527341,
65
+ "eval_precision": 0.7630858215778352,
66
+ "eval_precision_macro": 0.5565386466479504,
67
+ "eval_recall": 0.7738187451587917,
68
+ "eval_recall_macro": 0.5407847836684627,
69
+ "eval_runtime": 119.5419,
70
+ "eval_samples_per_second": 10.8,
71
+ "eval_steps_per_second": 1.355,
72
+ "eval_weighted_fpr": 0.020451043563524302,
73
+ "eval_weighted_sensitivity": 0.7738187451587917,
74
+ "eval_weighted_specificity": 0.9725268936322198,
75
  "step": 482
76
  },
77
  {
78
  "epoch": 3.11,
79
  "learning_rate": 3.958333333333333e-05,
80
+ "loss": 1.08,
81
  "step": 500
82
  },
83
  {
84
  "epoch": 4.0,
85
+ "eval_accuracy": 0.7575522850503486,
86
+ "eval_f1_macro": 0.5497163954183316,
87
+ "eval_f1_micro": 0.7575522850503486,
88
+ "eval_f1_weighted": 0.7437674055775699,
89
+ "eval_loss": 0.7538753151893616,
90
+ "eval_macro_fpr": 0.023393901824967867,
91
+ "eval_macro_sensitivity": 0.5613408412198008,
92
+ "eval_macro_specificity": 0.9817091819688993,
93
+ "eval_precision": 0.7583906478192325,
94
+ "eval_precision_macro": 0.5791225123984725,
95
+ "eval_recall": 0.7575522850503486,
96
+ "eval_recall_macro": 0.5613408412198008,
97
+ "eval_runtime": 120.3095,
98
+ "eval_samples_per_second": 10.731,
99
+ "eval_steps_per_second": 1.347,
100
+ "eval_weighted_fpr": 0.022349161013923598,
101
+ "eval_weighted_sensitivity": 0.7575522850503486,
102
+ "eval_weighted_specificity": 0.9680854444831427,
103
  "step": 643
104
  },
105
  {
106
  "epoch": 5.0,
107
+ "eval_accuracy": 0.7831138652207591,
108
+ "eval_f1_macro": 0.6353946839596228,
109
+ "eval_f1_micro": 0.783113865220759,
110
+ "eval_f1_weighted": 0.770250750407997,
111
+ "eval_loss": 0.6978110074996948,
112
+ "eval_macro_fpr": 0.020328728105215688,
113
+ "eval_macro_sensitivity": 0.6491872351233146,
114
+ "eval_macro_specificity": 0.983610303127039,
115
+ "eval_precision": 0.790031827496624,
116
+ "eval_precision_macro": 0.7409525878258343,
117
+ "eval_recall": 0.7831138652207591,
118
+ "eval_recall_macro": 0.6491872351233146,
119
+ "eval_runtime": 119.9411,
120
+ "eval_samples_per_second": 10.764,
121
+ "eval_steps_per_second": 1.351,
122
+ "eval_weighted_fpr": 0.019398642095053348,
123
+ "eval_weighted_sensitivity": 0.7831138652207591,
124
+ "eval_weighted_specificity": 0.9710406816848265,
125
  "step": 803
126
  },
127
  {
128
  "epoch": 6.0,
129
+ "eval_accuracy": 0.8156467854376452,
130
+ "eval_f1_macro": 0.686011065139391,
131
+ "eval_f1_micro": 0.8156467854376452,
132
+ "eval_f1_weighted": 0.8087620976325791,
133
+ "eval_loss": 0.5920090675354004,
134
+ "eval_macro_fpr": 0.016634568189926056,
135
+ "eval_macro_sensitivity": 0.6889168814632094,
136
+ "eval_macro_specificity": 0.9860196319752603,
137
+ "eval_precision": 0.8053079350996695,
138
+ "eval_precision_macro": 0.7051361586363433,
139
+ "eval_recall": 0.8156467854376452,
140
+ "eval_recall_macro": 0.6889168814632094,
141
+ "eval_runtime": 121.4244,
142
+ "eval_samples_per_second": 10.632,
143
+ "eval_steps_per_second": 1.334,
144
+ "eval_weighted_fpr": 0.01588785046728972,
145
+ "eval_weighted_sensitivity": 0.8156467854376452,
146
+ "eval_weighted_specificity": 0.974647694191258,
147
  "step": 964
148
  },
149
  {
150
  "epoch": 6.22,
151
  "learning_rate": 2.916666666666667e-05,
152
+ "loss": 0.5581,
153
  "step": 1000
154
  },
155
  {
156
  "epoch": 7.0,
157
+ "eval_accuracy": 0.8187451587916343,
158
+ "eval_f1_macro": 0.7393184269658342,
159
+ "eval_f1_micro": 0.8187451587916342,
160
+ "eval_f1_weighted": 0.8146848849127107,
161
+ "eval_loss": 0.6230757236480713,
162
+ "eval_macro_fpr": 0.016206558639896132,
163
+ "eval_macro_sensitivity": 0.7424574536250088,
164
+ "eval_macro_specificity": 0.9863587851767341,
165
+ "eval_precision": 0.8177894369587694,
166
+ "eval_precision_macro": 0.7626654223433325,
167
+ "eval_recall": 0.8187451587916343,
168
+ "eval_recall_macro": 0.7424574536250088,
169
+ "eval_runtime": 122.7775,
170
+ "eval_samples_per_second": 10.515,
171
+ "eval_steps_per_second": 1.319,
172
+ "eval_weighted_fpr": 0.015566790846194785,
173
+ "eval_weighted_sensitivity": 0.8187451587916343,
174
+ "eval_weighted_specificity": 0.9766366188593788,
175
  "step": 1125
176
  },
177
  {
178
  "epoch": 8.0,
179
+ "eval_accuracy": 0.8140975987606507,
180
+ "eval_f1_macro": 0.7328585966804847,
181
+ "eval_f1_micro": 0.8140975987606507,
182
+ "eval_f1_weighted": 0.8088643288992207,
183
+ "eval_loss": 0.6290707588195801,
184
+ "eval_macro_fpr": 0.01671628651715199,
185
+ "eval_macro_sensitivity": 0.730719512015592,
186
+ "eval_macro_specificity": 0.985990627132147,
187
+ "eval_precision": 0.8133783130550384,
188
+ "eval_precision_macro": 0.7636063765511063,
189
+ "eval_recall": 0.8140975987606507,
190
+ "eval_recall_macro": 0.730719512015592,
191
+ "eval_runtime": 122.8754,
192
+ "eval_samples_per_second": 10.507,
193
+ "eval_steps_per_second": 1.318,
194
+ "eval_weighted_fpr": 0.016049217600641968,
195
+ "eval_weighted_sensitivity": 0.8140975987606507,
196
+ "eval_weighted_specificity": 0.975761808221554,
197
  "step": 1286
198
  },
199
  {
200
  "epoch": 9.0,
201
+ "eval_accuracy": 0.8241673121611154,
202
+ "eval_f1_macro": 0.7364711129039813,
203
+ "eval_f1_micro": 0.8241673121611154,
204
+ "eval_f1_weighted": 0.8190822404938951,
205
+ "eval_loss": 0.6226187348365784,
206
+ "eval_macro_fpr": 0.015759087941173786,
207
+ "eval_macro_sensitivity": 0.7340167949600996,
208
+ "eval_macro_specificity": 0.9866810242390173,
209
+ "eval_precision": 0.8212295601295609,
210
+ "eval_precision_macro": 0.7665878963082912,
211
+ "eval_recall": 0.8241673121611154,
212
+ "eval_recall_macro": 0.7340167949600996,
213
+ "eval_runtime": 119.2726,
214
+ "eval_samples_per_second": 10.824,
215
+ "eval_steps_per_second": 1.358,
216
+ "eval_weighted_fpr": 0.015010249289162203,
217
+ "eval_weighted_sensitivity": 0.8241673121611154,
218
+ "eval_weighted_specificity": 0.9760480514241424,
219
  "step": 1446
220
  },
221
  {
222
  "epoch": 9.33,
223
  "learning_rate": 1.8750000000000002e-05,
224
+ "loss": 0.3924,
225
  "step": 1500
226
  },
227
  {
228
  "epoch": 10.0,
229
+ "eval_accuracy": 0.8109992254066615,
230
+ "eval_f1_macro": 0.7240137823928982,
231
+ "eval_f1_micro": 0.8109992254066615,
232
+ "eval_f1_weighted": 0.8047604071311991,
233
+ "eval_loss": 0.672822892665863,
234
+ "eval_macro_fpr": 0.017004063571327103,
235
+ "eval_macro_sensitivity": 0.7288750805513627,
236
+ "eval_macro_specificity": 0.9858130697255809,
237
+ "eval_precision": 0.8123042127541057,
238
+ "eval_precision_macro": 0.7417780746962291,
239
+ "eval_recall": 0.8109992254066615,
240
+ "eval_recall_macro": 0.7288750805513627,
241
+ "eval_runtime": 119.3649,
242
+ "eval_samples_per_second": 10.816,
243
+ "eval_steps_per_second": 1.357,
244
+ "eval_weighted_fpr": 0.016373641121997046,
245
+ "eval_weighted_sensitivity": 0.8109992254066615,
246
+ "eval_weighted_specificity": 0.976196820477052,
247
  "step": 1607
248
  },
249
  {
250
  "epoch": 11.0,
251
+ "eval_accuracy": 0.8094500387296669,
252
+ "eval_f1_macro": 0.7262574438118878,
253
+ "eval_f1_micro": 0.8094500387296669,
254
+ "eval_f1_weighted": 0.8026446222230169,
255
+ "eval_loss": 0.6805150508880615,
256
+ "eval_macro_fpr": 0.017269448415116812,
257
+ "eval_macro_sensitivity": 0.7303446738424816,
258
+ "eval_macro_specificity": 0.9856433903766489,
259
+ "eval_precision": 0.8123272064418959,
260
+ "eval_precision_macro": 0.7389776120628855,
261
+ "eval_recall": 0.8094500387296669,
262
+ "eval_recall_macro": 0.7303446738424816,
263
+ "eval_runtime": 119.049,
264
+ "eval_samples_per_second": 10.844,
265
+ "eval_steps_per_second": 1.361,
266
+ "eval_weighted_fpr": 0.016536703414896478,
267
+ "eval_weighted_sensitivity": 0.8094500387296669,
268
+ "eval_weighted_specificity": 0.9752008169200658,
269
  "step": 1768
270
  },
271
  {
272
  "epoch": 12.0,
273
+ "eval_accuracy": 0.8133230054221534,
274
+ "eval_f1_macro": 0.7283700928598721,
275
+ "eval_f1_micro": 0.8133230054221534,
276
+ "eval_f1_weighted": 0.8089952243674349,
277
+ "eval_loss": 0.670982301235199,
278
+ "eval_macro_fpr": 0.016803062086639158,
279
+ "eval_macro_sensitivity": 0.7306161057668827,
280
+ "eval_macro_specificity": 0.9859491188467968,
281
+ "eval_precision": 0.813705327621,
282
+ "eval_precision_macro": 0.7396207219727483,
283
+ "eval_recall": 0.8133230054221534,
284
+ "eval_recall_macro": 0.7306161057668827,
285
+ "eval_runtime": 179.3033,
286
+ "eval_samples_per_second": 7.2,
287
+ "eval_steps_per_second": 0.903,
288
+ "eval_weighted_fpr": 0.016130111772973698,
289
+ "eval_weighted_sensitivity": 0.8133230054221534,
290
+ "eval_weighted_specificity": 0.9759137772797981,
291
  "step": 1929
292
  },
293
  {
294
  "epoch": 12.44,
295
  "learning_rate": 8.333333333333334e-06,
296
+ "loss": 0.2929,
297
  "step": 2000
298
  },
299
  {
300
  "epoch": 13.0,
301
+ "eval_accuracy": 0.8187451587916343,
302
+ "eval_f1_macro": 0.73684785586885,
303
+ "eval_f1_micro": 0.8187451587916342,
304
+ "eval_f1_weighted": 0.8150888108444543,
305
+ "eval_loss": 0.6740049719810486,
306
+ "eval_macro_fpr": 0.01621356652381048,
307
+ "eval_macro_sensitivity": 0.7360046444502426,
308
+ "eval_macro_specificity": 0.9863212120014355,
309
+ "eval_precision": 0.8169592155695241,
310
+ "eval_precision_macro": 0.7644187374870298,
311
+ "eval_recall": 0.8187451587916343,
312
+ "eval_recall_macro": 0.7360046444502426,
313
+ "eval_runtime": 119.1613,
314
+ "eval_samples_per_second": 10.834,
315
+ "eval_steps_per_second": 1.36,
316
+ "eval_weighted_fpr": 0.015566790846194785,
317
+ "eval_weighted_sensitivity": 0.8187451587916343,
318
+ "eval_weighted_specificity": 0.9760730212298985,
319
  "step": 2089
320
  },
321
  {
322
  "epoch": 14.0,
323
+ "eval_accuracy": 0.8179705654531371,
324
+ "eval_f1_macro": 0.7360855331051521,
325
+ "eval_f1_micro": 0.817970565453137,
326
+ "eval_f1_weighted": 0.8136558284635095,
327
+ "eval_loss": 0.6823435425758362,
328
+ "eval_macro_fpr": 0.016364294303971197,
329
+ "eval_macro_sensitivity": 0.7336475818062481,
330
+ "eval_macro_specificity": 0.9862178349089701,
331
+ "eval_precision": 0.8158757496725866,
332
+ "eval_precision_macro": 0.7657492044841306,
333
+ "eval_recall": 0.8179705654531371,
334
+ "eval_recall_macro": 0.7336475818062481,
335
+ "eval_runtime": 119.9116,
336
+ "eval_samples_per_second": 10.766,
337
+ "eval_steps_per_second": 1.351,
338
+ "eval_weighted_fpr": 0.01564684732671949,
339
+ "eval_weighted_sensitivity": 0.8179705654531371,
340
+ "eval_weighted_specificity": 0.9752969581814132,
341
  "step": 2250
342
  }
343
  ],
 
345
  "max_steps": 2400,
346
  "num_train_epochs": 15,
347
  "save_steps": 500,
348
+ "total_flos": 8.009993230615757e+16,
349
  "trial_name": null,
350
  "trial_params": null
351
  }
training_checkpoints/checkpoint-2250/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cfb230625284b71a8582d0495ae9a462b745f8cb976f9d7932c4f91357944609
3
  size 4600
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8cfeeb4079e0edbd60cad2df3b73a192a8e64f61f57cc52cfabb80fd9ff6105d
3
  size 4600