Commit
a0ee100
·
verified ·
1 Parent(s): e013216

Add cross-encoder-ms-marco-MiniLM-L-6-v2-E3-S8-Mmulti_neg_pairwise

Browse files

Upload fine-tuned rerankers for BioASQ 14B

Co-authored-by: André Ribeiro <andrepedro2004@hotmail.com>
Co-authored-by: Rúben Garrido <rubengarrido@ua.pt>

cross-encoder-ms-marco-MiniLM-L-6-v2-E3-S8-Mmulti_neg_pairwise/config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_cross_attention": false,
3
+ "architectures": [
4
+ "BertForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": null,
8
+ "classifier_dropout": null,
9
+ "dtype": "bfloat16",
10
+ "eos_token_id": null,
11
+ "gradient_checkpointing": false,
12
+ "hidden_act": "gelu",
13
+ "hidden_dropout_prob": 0.1,
14
+ "hidden_size": 384,
15
+ "id2label": {
16
+ "0": "LABEL_0"
17
+ },
18
+ "initializer_range": 0.02,
19
+ "intermediate_size": 1536,
20
+ "is_decoder": false,
21
+ "label2id": {
22
+ "LABEL_0": 0
23
+ },
24
+ "layer_norm_eps": 1e-12,
25
+ "max_position_embeddings": 512,
26
+ "model_type": "bert",
27
+ "num_attention_heads": 12,
28
+ "num_hidden_layers": 6,
29
+ "pad_token_id": 0,
30
+ "position_embedding_type": "absolute",
31
+ "sbert_ce_default_activation_function": "torch.nn.modules.linear.Identity",
32
+ "tie_word_embeddings": true,
33
+ "transformers_version": "5.2.0",
34
+ "type_vocab_size": 2,
35
+ "use_cache": false,
36
+ "vocab_size": 30522
37
+ }
cross-encoder-ms-marco-MiniLM-L-6-v2-E3-S8-Mmulti_neg_pairwise/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d561edb509f80bf897e0d2bab72d448e85b1ec154c2899c72d762f2b3735ead7
3
+ size 45439306
cross-encoder-ms-marco-MiniLM-L-6-v2-E3-S8-Mmulti_neg_pairwise/ranx_results.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model": "cross-encoder-ms-marco-MiniLM-L-6-v2",
3
+ "val_files": [
4
+ "../../data/val_data/13B3_golden.json",
5
+ "../../data/val_data/13B1_golden.json",
6
+ "../../data/val_data/13B2_golden.json"
7
+ ],
8
+ "total": {
9
+ "ndcg@5": 0.7345278489266204,
10
+ "mrr": 0.8645173814833258,
11
+ "recall@10": 0.4616206216362286,
12
+ "recall@100": 0.9079625946465383,
13
+ "map@10": 0.39470150427606854,
14
+ "map-bioasq@10": 0.6097957256976865
15
+ },
16
+ "13B3_golden.json": {
17
+ "ndcg@5": 0.6743919612804384,
18
+ "mrr": 0.8226696832579186,
19
+ "recall@10": 0.40613561201616544,
20
+ "recall@100": 0.9092559827762909,
21
+ "map@10": 0.3367293648827149,
22
+ "map-bioasq@10": 0.5549048137773628
23
+ },
24
+ "13B1_golden.json": {
25
+ "ndcg@5": 0.7095794190343618,
26
+ "mrr": 0.8561765788391176,
27
+ "recall@10": 0.47273673751483974,
28
+ "recall@100": 0.91953492548711,
29
+ "map@10": 0.39891810218416235,
30
+ "map-bioasq@10": 0.5672834837638759
31
+ },
32
+ "13B2_golden.json": {
33
+ "ndcg@5": 0.819612166465061,
34
+ "mrr": 0.9147058823529411,
35
+ "recall@10": 0.5059895153776804,
36
+ "recall@100": 0.8950968756762139,
37
+ "map@10": 0.4484570457613284,
38
+ "map-bioasq@10": 0.7071988795518207
39
+ }
40
+ }
cross-encoder-ms-marco-MiniLM-L-6-v2-E3-S8-Mmulti_neg_pairwise/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
cross-encoder-ms-marco-MiniLM-L-6-v2-E3-S8-Mmulti_neg_pairwise/tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "backend": "tokenizers",
3
+ "clean_up_tokenization_spaces": true,
4
+ "cls_token": "[CLS]",
5
+ "do_basic_tokenize": true,
6
+ "do_lower_case": true,
7
+ "is_local": false,
8
+ "mask_token": "[MASK]",
9
+ "model_max_length": 512,
10
+ "never_split": null,
11
+ "pad_token": "[PAD]",
12
+ "sep_token": "[SEP]",
13
+ "strip_accents": null,
14
+ "tokenize_chinese_chars": true,
15
+ "tokenizer_class": "BertTokenizer",
16
+ "unk_token": "[UNK]"
17
+ }
cross-encoder-ms-marco-MiniLM-L-6-v2-E3-S8-Mmulti_neg_pairwise/trainer_state.json ADDED
@@ -0,0 +1,562 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 3.0,
6
+ "eval_steps": 500,
7
+ "global_step": 3636,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.041254125412541254,
14
+ "grad_norm": 83.0,
15
+ "learning_rate": 1.9730473047304733e-05,
16
+ "loss": 7.137315063476563,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.08250825082508251,
21
+ "grad_norm": 48.25,
22
+ "learning_rate": 1.9455445544554456e-05,
23
+ "loss": 5.915611572265625,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.12376237623762376,
28
+ "grad_norm": 43.75,
29
+ "learning_rate": 1.9180418041804182e-05,
30
+ "loss": 5.999351806640625,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.16501650165016502,
35
+ "grad_norm": 30.625,
36
+ "learning_rate": 1.890539053905391e-05,
37
+ "loss": 5.381943969726563,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 0.20627062706270627,
42
+ "grad_norm": 34.5,
43
+ "learning_rate": 1.863036303630363e-05,
44
+ "loss": 5.5327880859375,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 0.24752475247524752,
49
+ "grad_norm": 37.75,
50
+ "learning_rate": 1.8355335533553358e-05,
51
+ "loss": 4.7126007080078125,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.2887788778877888,
56
+ "grad_norm": 32.25,
57
+ "learning_rate": 1.808030803080308e-05,
58
+ "loss": 4.513934631347656,
59
+ "step": 350
60
+ },
61
+ {
62
+ "epoch": 0.33003300330033003,
63
+ "grad_norm": 37.0,
64
+ "learning_rate": 1.7805280528052807e-05,
65
+ "loss": 4.970782165527344,
66
+ "step": 400
67
+ },
68
+ {
69
+ "epoch": 0.3712871287128713,
70
+ "grad_norm": 34.75,
71
+ "learning_rate": 1.7530253025302533e-05,
72
+ "loss": 4.769329833984375,
73
+ "step": 450
74
+ },
75
+ {
76
+ "epoch": 0.41254125412541254,
77
+ "grad_norm": 30.375,
78
+ "learning_rate": 1.7255225522552256e-05,
79
+ "loss": 4.452649536132813,
80
+ "step": 500
81
+ },
82
+ {
83
+ "epoch": 0.4537953795379538,
84
+ "grad_norm": 37.5,
85
+ "learning_rate": 1.6980198019801982e-05,
86
+ "loss": 4.582575378417968,
87
+ "step": 550
88
+ },
89
+ {
90
+ "epoch": 0.49504950495049505,
91
+ "grad_norm": 35.5,
92
+ "learning_rate": 1.6705170517051708e-05,
93
+ "loss": 4.3151766967773435,
94
+ "step": 600
95
+ },
96
+ {
97
+ "epoch": 0.5363036303630363,
98
+ "grad_norm": 35.0,
99
+ "learning_rate": 1.643014301430143e-05,
100
+ "loss": 4.3324758911132815,
101
+ "step": 650
102
+ },
103
+ {
104
+ "epoch": 0.5775577557755776,
105
+ "grad_norm": 35.5,
106
+ "learning_rate": 1.6155115511551157e-05,
107
+ "loss": 4.510329895019531,
108
+ "step": 700
109
+ },
110
+ {
111
+ "epoch": 0.6188118811881188,
112
+ "grad_norm": 40.75,
113
+ "learning_rate": 1.5880088008800884e-05,
114
+ "loss": 4.240176696777343,
115
+ "step": 750
116
+ },
117
+ {
118
+ "epoch": 0.6600660066006601,
119
+ "grad_norm": 29.5,
120
+ "learning_rate": 1.5605060506050606e-05,
121
+ "loss": 3.992020263671875,
122
+ "step": 800
123
+ },
124
+ {
125
+ "epoch": 0.7013201320132013,
126
+ "grad_norm": 23.625,
127
+ "learning_rate": 1.533003300330033e-05,
128
+ "loss": 4.241922607421875,
129
+ "step": 850
130
+ },
131
+ {
132
+ "epoch": 0.7425742574257426,
133
+ "grad_norm": 33.5,
134
+ "learning_rate": 1.5055005500550057e-05,
135
+ "loss": 4.440846862792969,
136
+ "step": 900
137
+ },
138
+ {
139
+ "epoch": 0.7838283828382838,
140
+ "grad_norm": 32.0,
141
+ "learning_rate": 1.4779977997799782e-05,
142
+ "loss": 3.99255859375,
143
+ "step": 950
144
+ },
145
+ {
146
+ "epoch": 0.8250825082508251,
147
+ "grad_norm": 31.875,
148
+ "learning_rate": 1.4504950495049506e-05,
149
+ "loss": 4.233996276855469,
150
+ "step": 1000
151
+ },
152
+ {
153
+ "epoch": 0.8663366336633663,
154
+ "grad_norm": 26.5,
155
+ "learning_rate": 1.4229922992299232e-05,
156
+ "loss": 4.023433227539062,
157
+ "step": 1050
158
+ },
159
+ {
160
+ "epoch": 0.9075907590759076,
161
+ "grad_norm": 28.0,
162
+ "learning_rate": 1.3954895489548957e-05,
163
+ "loss": 4.419351196289062,
164
+ "step": 1100
165
+ },
166
+ {
167
+ "epoch": 0.9488448844884488,
168
+ "grad_norm": 32.75,
169
+ "learning_rate": 1.367986798679868e-05,
170
+ "loss": 4.055094299316406,
171
+ "step": 1150
172
+ },
173
+ {
174
+ "epoch": 0.9900990099009901,
175
+ "grad_norm": 33.0,
176
+ "learning_rate": 1.3404840484048408e-05,
177
+ "loss": 4.113250122070313,
178
+ "step": 1200
179
+ },
180
+ {
181
+ "epoch": 1.0,
182
+ "eval_loss": 0.5890524387359619,
183
+ "eval_runtime": 0.5028,
184
+ "eval_samples_per_second": 503.156,
185
+ "eval_steps_per_second": 7.955,
186
+ "step": 1212
187
+ },
188
+ {
189
+ "epoch": 1.0313531353135315,
190
+ "grad_norm": 22.25,
191
+ "learning_rate": 1.312981298129813e-05,
192
+ "loss": 4.054990234375,
193
+ "step": 1250
194
+ },
195
+ {
196
+ "epoch": 1.0726072607260726,
197
+ "grad_norm": 21.25,
198
+ "learning_rate": 1.2854785478547855e-05,
199
+ "loss": 4.209974975585937,
200
+ "step": 1300
201
+ },
202
+ {
203
+ "epoch": 1.113861386138614,
204
+ "grad_norm": 19.875,
205
+ "learning_rate": 1.2579757975797581e-05,
206
+ "loss": 4.038520202636719,
207
+ "step": 1350
208
+ },
209
+ {
210
+ "epoch": 1.155115511551155,
211
+ "grad_norm": 28.25,
212
+ "learning_rate": 1.2304730473047306e-05,
213
+ "loss": 3.918459167480469,
214
+ "step": 1400
215
+ },
216
+ {
217
+ "epoch": 1.1963696369636962,
218
+ "grad_norm": 30.625,
219
+ "learning_rate": 1.202970297029703e-05,
220
+ "loss": 4.268034973144531,
221
+ "step": 1450
222
+ },
223
+ {
224
+ "epoch": 1.2376237623762376,
225
+ "grad_norm": 35.0,
226
+ "learning_rate": 1.1754675467546757e-05,
227
+ "loss": 4.156949157714844,
228
+ "step": 1500
229
+ },
230
+ {
231
+ "epoch": 1.278877887788779,
232
+ "grad_norm": 30.5,
233
+ "learning_rate": 1.1479647964796481e-05,
234
+ "loss": 4.086359252929688,
235
+ "step": 1550
236
+ },
237
+ {
238
+ "epoch": 1.3201320132013201,
239
+ "grad_norm": 32.5,
240
+ "learning_rate": 1.1204620462046206e-05,
241
+ "loss": 4.243644714355469,
242
+ "step": 1600
243
+ },
244
+ {
245
+ "epoch": 1.3613861386138613,
246
+ "grad_norm": 45.25,
247
+ "learning_rate": 1.0929592959295932e-05,
248
+ "loss": 4.332274475097656,
249
+ "step": 1650
250
+ },
251
+ {
252
+ "epoch": 1.4026402640264026,
253
+ "grad_norm": 32.5,
254
+ "learning_rate": 1.0654565456545657e-05,
255
+ "loss": 3.9054995727539064,
256
+ "step": 1700
257
+ },
258
+ {
259
+ "epoch": 1.443894389438944,
260
+ "grad_norm": 26.875,
261
+ "learning_rate": 1.037953795379538e-05,
262
+ "loss": 4.0744476318359375,
263
+ "step": 1750
264
+ },
265
+ {
266
+ "epoch": 1.4851485148514851,
267
+ "grad_norm": 32.25,
268
+ "learning_rate": 1.0104510451045106e-05,
269
+ "loss": 3.759951477050781,
270
+ "step": 1800
271
+ },
272
+ {
273
+ "epoch": 1.5264026402640263,
274
+ "grad_norm": 25.0,
275
+ "learning_rate": 9.82948294829483e-06,
276
+ "loss": 4.088746948242187,
277
+ "step": 1850
278
+ },
279
+ {
280
+ "epoch": 1.5676567656765676,
281
+ "grad_norm": 28.75,
282
+ "learning_rate": 9.554455445544555e-06,
283
+ "loss": 4.369046325683594,
284
+ "step": 1900
285
+ },
286
+ {
287
+ "epoch": 1.608910891089109,
288
+ "grad_norm": 40.0,
289
+ "learning_rate": 9.279427942794281e-06,
290
+ "loss": 3.942612609863281,
291
+ "step": 1950
292
+ },
293
+ {
294
+ "epoch": 1.6501650165016502,
295
+ "grad_norm": 25.625,
296
+ "learning_rate": 9.004400440044006e-06,
297
+ "loss": 3.8802056884765626,
298
+ "step": 2000
299
+ },
300
+ {
301
+ "epoch": 1.6914191419141913,
302
+ "grad_norm": 35.0,
303
+ "learning_rate": 8.72937293729373e-06,
304
+ "loss": 3.846055908203125,
305
+ "step": 2050
306
+ },
307
+ {
308
+ "epoch": 1.7326732673267327,
309
+ "grad_norm": 36.0,
310
+ "learning_rate": 8.454345434543455e-06,
311
+ "loss": 3.6827587890625,
312
+ "step": 2100
313
+ },
314
+ {
315
+ "epoch": 1.773927392739274,
316
+ "grad_norm": 25.5,
317
+ "learning_rate": 8.17931793179318e-06,
318
+ "loss": 4.043742980957031,
319
+ "step": 2150
320
+ },
321
+ {
322
+ "epoch": 1.8151815181518152,
323
+ "grad_norm": 30.875,
324
+ "learning_rate": 7.904290429042905e-06,
325
+ "loss": 4.189202880859375,
326
+ "step": 2200
327
+ },
328
+ {
329
+ "epoch": 1.8564356435643563,
330
+ "grad_norm": 30.375,
331
+ "learning_rate": 7.62926292629263e-06,
332
+ "loss": 3.9661297607421875,
333
+ "step": 2250
334
+ },
335
+ {
336
+ "epoch": 1.8976897689768977,
337
+ "grad_norm": 32.25,
338
+ "learning_rate": 7.354235423542355e-06,
339
+ "loss": 3.683446960449219,
340
+ "step": 2300
341
+ },
342
+ {
343
+ "epoch": 1.938943894389439,
344
+ "grad_norm": 28.375,
345
+ "learning_rate": 7.079207920792079e-06,
346
+ "loss": 3.7715469360351563,
347
+ "step": 2350
348
+ },
349
+ {
350
+ "epoch": 1.9801980198019802,
351
+ "grad_norm": 32.25,
352
+ "learning_rate": 6.804180418041804e-06,
353
+ "loss": 4.085735473632813,
354
+ "step": 2400
355
+ },
356
+ {
357
+ "epoch": 2.0,
358
+ "eval_loss": 0.5830715298652649,
359
+ "eval_runtime": 0.7174,
360
+ "eval_samples_per_second": 352.646,
361
+ "eval_steps_per_second": 5.575,
362
+ "step": 2424
363
+ },
364
+ {
365
+ "epoch": 2.0214521452145213,
366
+ "grad_norm": 32.0,
367
+ "learning_rate": 6.52915291529153e-06,
368
+ "loss": 4.060396728515625,
369
+ "step": 2450
370
+ },
371
+ {
372
+ "epoch": 2.062706270627063,
373
+ "grad_norm": 41.0,
374
+ "learning_rate": 6.254125412541254e-06,
375
+ "loss": 3.9967327880859376,
376
+ "step": 2500
377
+ },
378
+ {
379
+ "epoch": 2.103960396039604,
380
+ "grad_norm": 53.25,
381
+ "learning_rate": 5.97909790979098e-06,
382
+ "loss": 4.0453948974609375,
383
+ "step": 2550
384
+ },
385
+ {
386
+ "epoch": 2.145214521452145,
387
+ "grad_norm": 31.125,
388
+ "learning_rate": 5.704070407040705e-06,
389
+ "loss": 4.580944213867188,
390
+ "step": 2600
391
+ },
392
+ {
393
+ "epoch": 2.1864686468646863,
394
+ "grad_norm": 27.875,
395
+ "learning_rate": 5.429042904290429e-06,
396
+ "loss": 4.358104858398438,
397
+ "step": 2650
398
+ },
399
+ {
400
+ "epoch": 2.227722772277228,
401
+ "grad_norm": 24.375,
402
+ "learning_rate": 5.154015401540154e-06,
403
+ "loss": 4.2383575439453125,
404
+ "step": 2700
405
+ },
406
+ {
407
+ "epoch": 2.268976897689769,
408
+ "grad_norm": 32.5,
409
+ "learning_rate": 4.8789878987898795e-06,
410
+ "loss": 4.47793212890625,
411
+ "step": 2750
412
+ },
413
+ {
414
+ "epoch": 2.31023102310231,
415
+ "grad_norm": 24.875,
416
+ "learning_rate": 4.603960396039605e-06,
417
+ "loss": 4.445508422851563,
418
+ "step": 2800
419
+ },
420
+ {
421
+ "epoch": 2.3514851485148514,
422
+ "grad_norm": 25.875,
423
+ "learning_rate": 4.3289328932893295e-06,
424
+ "loss": 4.297294921875,
425
+ "step": 2850
426
+ },
427
+ {
428
+ "epoch": 2.3927392739273925,
429
+ "grad_norm": 28.375,
430
+ "learning_rate": 4.053905390539054e-06,
431
+ "loss": 4.231211242675781,
432
+ "step": 2900
433
+ },
434
+ {
435
+ "epoch": 2.433993399339934,
436
+ "grad_norm": 28.75,
437
+ "learning_rate": 3.7788778877887794e-06,
438
+ "loss": 3.899722900390625,
439
+ "step": 2950
440
+ },
441
+ {
442
+ "epoch": 2.4752475247524752,
443
+ "grad_norm": 41.5,
444
+ "learning_rate": 3.5038503850385044e-06,
445
+ "loss": 4.3706597900390625,
446
+ "step": 3000
447
+ },
448
+ {
449
+ "epoch": 2.5165016501650164,
450
+ "grad_norm": 31.0,
451
+ "learning_rate": 3.228822882288229e-06,
452
+ "loss": 4.427062377929688,
453
+ "step": 3050
454
+ },
455
+ {
456
+ "epoch": 2.557755775577558,
457
+ "grad_norm": 24.25,
458
+ "learning_rate": 2.9537953795379543e-06,
459
+ "loss": 4.363355102539063,
460
+ "step": 3100
461
+ },
462
+ {
463
+ "epoch": 2.599009900990099,
464
+ "grad_norm": 37.25,
465
+ "learning_rate": 2.678767876787679e-06,
466
+ "loss": 4.356752319335937,
467
+ "step": 3150
468
+ },
469
+ {
470
+ "epoch": 2.6402640264026402,
471
+ "grad_norm": 29.875,
472
+ "learning_rate": 2.4037403740374042e-06,
473
+ "loss": 4.191433410644532,
474
+ "step": 3200
475
+ },
476
+ {
477
+ "epoch": 2.6815181518151814,
478
+ "grad_norm": 31.5,
479
+ "learning_rate": 2.1287128712871288e-06,
480
+ "loss": 4.142337646484375,
481
+ "step": 3250
482
+ },
483
+ {
484
+ "epoch": 2.7227722772277225,
485
+ "grad_norm": 39.25,
486
+ "learning_rate": 1.853685368536854e-06,
487
+ "loss": 4.291042785644532,
488
+ "step": 3300
489
+ },
490
+ {
491
+ "epoch": 2.764026402640264,
492
+ "grad_norm": 27.5,
493
+ "learning_rate": 1.578657865786579e-06,
494
+ "loss": 4.2790087890625,
495
+ "step": 3350
496
+ },
497
+ {
498
+ "epoch": 2.8052805280528053,
499
+ "grad_norm": 25.75,
500
+ "learning_rate": 1.3036303630363036e-06,
501
+ "loss": 4.274718627929688,
502
+ "step": 3400
503
+ },
504
+ {
505
+ "epoch": 2.8465346534653464,
506
+ "grad_norm": 34.25,
507
+ "learning_rate": 1.0286028602860288e-06,
508
+ "loss": 4.3562008666992185,
509
+ "step": 3450
510
+ },
511
+ {
512
+ "epoch": 2.887788778877888,
513
+ "grad_norm": 34.0,
514
+ "learning_rate": 7.535753575357537e-07,
515
+ "loss": 4.399241027832031,
516
+ "step": 3500
517
+ },
518
+ {
519
+ "epoch": 2.929042904290429,
520
+ "grad_norm": 29.5,
521
+ "learning_rate": 4.785478547854786e-07,
522
+ "loss": 4.297891235351562,
523
+ "step": 3550
524
+ },
525
+ {
526
+ "epoch": 2.9702970297029703,
527
+ "grad_norm": 29.375,
528
+ "learning_rate": 2.0352035203520353e-07,
529
+ "loss": 4.296119995117188,
530
+ "step": 3600
531
+ },
532
+ {
533
+ "epoch": 3.0,
534
+ "eval_loss": 0.5826700925827026,
535
+ "eval_runtime": 0.6856,
536
+ "eval_samples_per_second": 369.039,
537
+ "eval_steps_per_second": 5.835,
538
+ "step": 3636
539
+ }
540
+ ],
541
+ "logging_steps": 50,
542
+ "max_steps": 3636,
543
+ "num_input_tokens_seen": 0,
544
+ "num_train_epochs": 3,
545
+ "save_steps": 500,
546
+ "stateful_callbacks": {
547
+ "TrainerControl": {
548
+ "args": {
549
+ "should_epoch_stop": false,
550
+ "should_evaluate": false,
551
+ "should_log": false,
552
+ "should_save": true,
553
+ "should_training_stop": true
554
+ },
555
+ "attributes": {}
556
+ }
557
+ },
558
+ "total_flos": 0.0,
559
+ "train_batch_size": 32,
560
+ "trial_name": null,
561
+ "trial_params": null
562
+ }
cross-encoder-ms-marco-MiniLM-L-6-v2-E3-S8-Mmulti_neg_pairwise/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a7528b157e492557b3d8206abdfab9b6b93263702df6ec6dd5fc2217e6fada8
3
+ size 5329