Commit
b01d4cb
·
verified ·
1 Parent(s): 96e2104

Add cross-encoder-ms-marco-MiniLM-L-6-v2-E5-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-E5-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-E5-Pairwise/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f39f74ce4db223ab950dbb95a24d0c5b5d61447e672af80d452c3736c1b9adb5
3
+ size 45439306
cross-encoder-ms-marco-MiniLM-L-6-v2-E5-Pairwise/ranx_results.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model": "cross-encoder-ms-marco-MiniLM-L-6-v2",
3
+ "val_files": [
4
+ "../../data/val_data/13B1_golden.json",
5
+ "../../data/val_data/13B2_golden.json"
6
+ ],
7
+ "total": {
8
+ "ndcg@5": 0.7610321687437726,
9
+ "mrr": 0.8814197835997143,
10
+ "recall@10": 0.4885639881923411,
11
+ "recall@100": 0.9064535051545861,
12
+ "map@10": 0.42351815271867427,
13
+ "map-bioasq@10": 0.6373066383960992
14
+ },
15
+ "13B1_golden.json": {
16
+ "ndcg@5": 0.7034010229203814,
17
+ "mrr": 0.8484278024935463,
18
+ "recall@10": 0.4715925321406882,
19
+ "recall@100": 0.9173000151578273,
20
+ "map@10": 0.3970424184734813,
21
+ "map-bioasq@10": 0.5646531149496836
22
+ },
23
+ "13B2_golden.json": {
24
+ "ndcg@5": 0.8186633145671637,
25
+ "mrr": 0.9144117647058823,
26
+ "recall@10": 0.5055354442439941,
27
+ "recall@100": 0.8956069951513448,
28
+ "map@10": 0.44999388696386733,
29
+ "map-bioasq@10": 0.7099601618425148
30
+ }
31
+ }
cross-encoder-ms-marco-MiniLM-L-6-v2-E5-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-E5-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-E5-Pairwise/trainer_state.json ADDED
@@ -0,0 +1,963 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 5.0,
6
+ "eval_steps": 500,
7
+ "global_step": 6375,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.0392156862745098,
14
+ "grad_norm": 9.3125,
15
+ "learning_rate": 1.9846274509803922e-05,
16
+ "loss": 0.9331813812255859,
17
+ "step": 50
18
+ },
19
+ {
20
+ "epoch": 0.0784313725490196,
21
+ "grad_norm": 10.5625,
22
+ "learning_rate": 1.9689411764705883e-05,
23
+ "loss": 0.8077426147460938,
24
+ "step": 100
25
+ },
26
+ {
27
+ "epoch": 0.11764705882352941,
28
+ "grad_norm": 8.6875,
29
+ "learning_rate": 1.9532549019607844e-05,
30
+ "loss": 0.7292742156982421,
31
+ "step": 150
32
+ },
33
+ {
34
+ "epoch": 0.1568627450980392,
35
+ "grad_norm": 7.5625,
36
+ "learning_rate": 1.9375686274509805e-05,
37
+ "loss": 0.760586929321289,
38
+ "step": 200
39
+ },
40
+ {
41
+ "epoch": 0.19607843137254902,
42
+ "grad_norm": 8.375,
43
+ "learning_rate": 1.9218823529411766e-05,
44
+ "loss": 0.65364501953125,
45
+ "step": 250
46
+ },
47
+ {
48
+ "epoch": 0.23529411764705882,
49
+ "grad_norm": 8.0,
50
+ "learning_rate": 1.9061960784313727e-05,
51
+ "loss": 0.6206476211547851,
52
+ "step": 300
53
+ },
54
+ {
55
+ "epoch": 0.27450980392156865,
56
+ "grad_norm": 6.3125,
57
+ "learning_rate": 1.8905098039215688e-05,
58
+ "loss": 0.655108642578125,
59
+ "step": 350
60
+ },
61
+ {
62
+ "epoch": 0.3137254901960784,
63
+ "grad_norm": 6.75,
64
+ "learning_rate": 1.874823529411765e-05,
65
+ "loss": 0.6366973114013672,
66
+ "step": 400
67
+ },
68
+ {
69
+ "epoch": 0.35294117647058826,
70
+ "grad_norm": 7.96875,
71
+ "learning_rate": 1.8591372549019607e-05,
72
+ "loss": 0.6533857727050781,
73
+ "step": 450
74
+ },
75
+ {
76
+ "epoch": 0.39215686274509803,
77
+ "grad_norm": 6.5625,
78
+ "learning_rate": 1.843450980392157e-05,
79
+ "loss": 0.557676773071289,
80
+ "step": 500
81
+ },
82
+ {
83
+ "epoch": 0.43137254901960786,
84
+ "grad_norm": 7.4375,
85
+ "learning_rate": 1.8277647058823532e-05,
86
+ "loss": 0.6256840896606445,
87
+ "step": 550
88
+ },
89
+ {
90
+ "epoch": 0.47058823529411764,
91
+ "grad_norm": 4.78125,
92
+ "learning_rate": 1.8120784313725493e-05,
93
+ "loss": 0.5899119567871094,
94
+ "step": 600
95
+ },
96
+ {
97
+ "epoch": 0.5098039215686274,
98
+ "grad_norm": 7.71875,
99
+ "learning_rate": 1.796392156862745e-05,
100
+ "loss": 0.5704912567138671,
101
+ "step": 650
102
+ },
103
+ {
104
+ "epoch": 0.5490196078431373,
105
+ "grad_norm": 5.125,
106
+ "learning_rate": 1.7807058823529412e-05,
107
+ "loss": 0.545570068359375,
108
+ "step": 700
109
+ },
110
+ {
111
+ "epoch": 0.5882352941176471,
112
+ "grad_norm": 6.09375,
113
+ "learning_rate": 1.7650196078431373e-05,
114
+ "loss": 0.5427178955078125,
115
+ "step": 750
116
+ },
117
+ {
118
+ "epoch": 0.6274509803921569,
119
+ "grad_norm": 5.78125,
120
+ "learning_rate": 1.7493333333333334e-05,
121
+ "loss": 0.5405741882324219,
122
+ "step": 800
123
+ },
124
+ {
125
+ "epoch": 0.6666666666666666,
126
+ "grad_norm": 6.375,
127
+ "learning_rate": 1.7336470588235295e-05,
128
+ "loss": 0.5612873077392578,
129
+ "step": 850
130
+ },
131
+ {
132
+ "epoch": 0.7058823529411765,
133
+ "grad_norm": 5.34375,
134
+ "learning_rate": 1.7179607843137256e-05,
135
+ "loss": 0.5727352142333985,
136
+ "step": 900
137
+ },
138
+ {
139
+ "epoch": 0.7450980392156863,
140
+ "grad_norm": 4.71875,
141
+ "learning_rate": 1.7022745098039217e-05,
142
+ "loss": 0.5243006134033203,
143
+ "step": 950
144
+ },
145
+ {
146
+ "epoch": 0.7843137254901961,
147
+ "grad_norm": 6.40625,
148
+ "learning_rate": 1.6865882352941178e-05,
149
+ "loss": 0.5094789505004883,
150
+ "step": 1000
151
+ },
152
+ {
153
+ "epoch": 0.8235294117647058,
154
+ "grad_norm": 6.1875,
155
+ "learning_rate": 1.670901960784314e-05,
156
+ "loss": 0.5455001831054688,
157
+ "step": 1050
158
+ },
159
+ {
160
+ "epoch": 0.8627450980392157,
161
+ "grad_norm": 7.34375,
162
+ "learning_rate": 1.65521568627451e-05,
163
+ "loss": 0.599549560546875,
164
+ "step": 1100
165
+ },
166
+ {
167
+ "epoch": 0.9019607843137255,
168
+ "grad_norm": 5.875,
169
+ "learning_rate": 1.639529411764706e-05,
170
+ "loss": 0.4979535675048828,
171
+ "step": 1150
172
+ },
173
+ {
174
+ "epoch": 0.9411764705882353,
175
+ "grad_norm": 3.875,
176
+ "learning_rate": 1.6238431372549022e-05,
177
+ "loss": 0.5428482818603516,
178
+ "step": 1200
179
+ },
180
+ {
181
+ "epoch": 0.9803921568627451,
182
+ "grad_norm": 4.75,
183
+ "learning_rate": 1.608156862745098e-05,
184
+ "loss": 0.5192766571044922,
185
+ "step": 1250
186
+ },
187
+ {
188
+ "epoch": 1.0,
189
+ "eval_loss": 0.5822960734367371,
190
+ "eval_runtime": 0.2336,
191
+ "eval_samples_per_second": 719.15,
192
+ "eval_steps_per_second": 12.842,
193
+ "step": 1275
194
+ },
195
+ {
196
+ "epoch": 1.0196078431372548,
197
+ "grad_norm": 6.3125,
198
+ "learning_rate": 1.592470588235294e-05,
199
+ "loss": 0.5328455734252929,
200
+ "step": 1300
201
+ },
202
+ {
203
+ "epoch": 1.0588235294117647,
204
+ "grad_norm": 5.8125,
205
+ "learning_rate": 1.5767843137254902e-05,
206
+ "loss": 0.5731172180175781,
207
+ "step": 1350
208
+ },
209
+ {
210
+ "epoch": 1.0980392156862746,
211
+ "grad_norm": 5.34375,
212
+ "learning_rate": 1.5610980392156866e-05,
213
+ "loss": 0.49518077850341796,
214
+ "step": 1400
215
+ },
216
+ {
217
+ "epoch": 1.1372549019607843,
218
+ "grad_norm": 5.96875,
219
+ "learning_rate": 1.5454117647058824e-05,
220
+ "loss": 0.5210482788085937,
221
+ "step": 1450
222
+ },
223
+ {
224
+ "epoch": 1.1764705882352942,
225
+ "grad_norm": 5.09375,
226
+ "learning_rate": 1.5297254901960785e-05,
227
+ "loss": 0.5553556060791016,
228
+ "step": 1500
229
+ },
230
+ {
231
+ "epoch": 1.215686274509804,
232
+ "grad_norm": 5.90625,
233
+ "learning_rate": 1.5140392156862746e-05,
234
+ "loss": 0.5134529113769531,
235
+ "step": 1550
236
+ },
237
+ {
238
+ "epoch": 1.2549019607843137,
239
+ "grad_norm": 5.4375,
240
+ "learning_rate": 1.4983529411764707e-05,
241
+ "loss": 0.5423490142822266,
242
+ "step": 1600
243
+ },
244
+ {
245
+ "epoch": 1.2941176470588236,
246
+ "grad_norm": 5.1875,
247
+ "learning_rate": 1.4826666666666666e-05,
248
+ "loss": 0.5546124267578125,
249
+ "step": 1650
250
+ },
251
+ {
252
+ "epoch": 1.3333333333333333,
253
+ "grad_norm": 5.34375,
254
+ "learning_rate": 1.4669803921568627e-05,
255
+ "loss": 0.532197494506836,
256
+ "step": 1700
257
+ },
258
+ {
259
+ "epoch": 1.3725490196078431,
260
+ "grad_norm": 6.65625,
261
+ "learning_rate": 1.451294117647059e-05,
262
+ "loss": 0.5223103332519531,
263
+ "step": 1750
264
+ },
265
+ {
266
+ "epoch": 1.4117647058823528,
267
+ "grad_norm": 4.71875,
268
+ "learning_rate": 1.4356078431372551e-05,
269
+ "loss": 0.5102643966674805,
270
+ "step": 1800
271
+ },
272
+ {
273
+ "epoch": 1.4509803921568627,
274
+ "grad_norm": 4.6875,
275
+ "learning_rate": 1.419921568627451e-05,
276
+ "loss": 0.5427158355712891,
277
+ "step": 1850
278
+ },
279
+ {
280
+ "epoch": 1.4901960784313726,
281
+ "grad_norm": 5.53125,
282
+ "learning_rate": 1.4042352941176472e-05,
283
+ "loss": 0.543260498046875,
284
+ "step": 1900
285
+ },
286
+ {
287
+ "epoch": 1.5294117647058822,
288
+ "grad_norm": 4.78125,
289
+ "learning_rate": 1.3885490196078433e-05,
290
+ "loss": 0.48865962982177735,
291
+ "step": 1950
292
+ },
293
+ {
294
+ "epoch": 1.5686274509803921,
295
+ "grad_norm": 6.1875,
296
+ "learning_rate": 1.3728627450980394e-05,
297
+ "loss": 0.5582750320434571,
298
+ "step": 2000
299
+ },
300
+ {
301
+ "epoch": 1.607843137254902,
302
+ "grad_norm": 4.96875,
303
+ "learning_rate": 1.3571764705882353e-05,
304
+ "loss": 0.5200865554809571,
305
+ "step": 2050
306
+ },
307
+ {
308
+ "epoch": 1.6470588235294117,
309
+ "grad_norm": 4.875,
310
+ "learning_rate": 1.3414901960784314e-05,
311
+ "loss": 0.5551679229736328,
312
+ "step": 2100
313
+ },
314
+ {
315
+ "epoch": 1.6862745098039216,
316
+ "grad_norm": 7.8125,
317
+ "learning_rate": 1.3258039215686275e-05,
318
+ "loss": 0.48414222717285155,
319
+ "step": 2150
320
+ },
321
+ {
322
+ "epoch": 1.7254901960784315,
323
+ "grad_norm": 5.28125,
324
+ "learning_rate": 1.3101176470588238e-05,
325
+ "loss": 0.5371945190429688,
326
+ "step": 2200
327
+ },
328
+ {
329
+ "epoch": 1.7647058823529411,
330
+ "grad_norm": 6.75,
331
+ "learning_rate": 1.2944313725490197e-05,
332
+ "loss": 0.5146623229980469,
333
+ "step": 2250
334
+ },
335
+ {
336
+ "epoch": 1.803921568627451,
337
+ "grad_norm": 5.0,
338
+ "learning_rate": 1.2787450980392158e-05,
339
+ "loss": 0.48903770446777345,
340
+ "step": 2300
341
+ },
342
+ {
343
+ "epoch": 1.843137254901961,
344
+ "grad_norm": 7.34375,
345
+ "learning_rate": 1.2630588235294119e-05,
346
+ "loss": 0.5088069915771485,
347
+ "step": 2350
348
+ },
349
+ {
350
+ "epoch": 1.8823529411764706,
351
+ "grad_norm": 6.0,
352
+ "learning_rate": 1.247372549019608e-05,
353
+ "loss": 0.48639339447021485,
354
+ "step": 2400
355
+ },
356
+ {
357
+ "epoch": 1.9215686274509802,
358
+ "grad_norm": 5.5,
359
+ "learning_rate": 1.231686274509804e-05,
360
+ "loss": 0.532833480834961,
361
+ "step": 2450
362
+ },
363
+ {
364
+ "epoch": 1.9607843137254903,
365
+ "grad_norm": 4.03125,
366
+ "learning_rate": 1.216e-05,
367
+ "loss": 0.4956613922119141,
368
+ "step": 2500
369
+ },
370
+ {
371
+ "epoch": 2.0,
372
+ "grad_norm": 9.1875,
373
+ "learning_rate": 1.2003137254901961e-05,
374
+ "loss": 0.5517213821411133,
375
+ "step": 2550
376
+ },
377
+ {
378
+ "epoch": 2.0,
379
+ "eval_loss": 0.5724225640296936,
380
+ "eval_runtime": 0.2831,
381
+ "eval_samples_per_second": 593.362,
382
+ "eval_steps_per_second": 10.596,
383
+ "step": 2550
384
+ },
385
+ {
386
+ "epoch": 2.0392156862745097,
387
+ "grad_norm": 4.625,
388
+ "learning_rate": 1.1846274509803923e-05,
389
+ "loss": 0.5009288787841797,
390
+ "step": 2600
391
+ },
392
+ {
393
+ "epoch": 2.0784313725490198,
394
+ "grad_norm": 6.75,
395
+ "learning_rate": 1.1689411764705882e-05,
396
+ "loss": 0.5402259826660156,
397
+ "step": 2650
398
+ },
399
+ {
400
+ "epoch": 2.1176470588235294,
401
+ "grad_norm": 5.28125,
402
+ "learning_rate": 1.1532549019607845e-05,
403
+ "loss": 0.5598141860961914,
404
+ "step": 2700
405
+ },
406
+ {
407
+ "epoch": 2.156862745098039,
408
+ "grad_norm": 5.0,
409
+ "learning_rate": 1.1375686274509806e-05,
410
+ "loss": 0.44724628448486325,
411
+ "step": 2750
412
+ },
413
+ {
414
+ "epoch": 2.196078431372549,
415
+ "grad_norm": 6.03125,
416
+ "learning_rate": 1.1218823529411767e-05,
417
+ "loss": 0.5136179733276367,
418
+ "step": 2800
419
+ },
420
+ {
421
+ "epoch": 2.235294117647059,
422
+ "grad_norm": 6.5,
423
+ "learning_rate": 1.1061960784313726e-05,
424
+ "loss": 0.5157033157348633,
425
+ "step": 2850
426
+ },
427
+ {
428
+ "epoch": 2.2745098039215685,
429
+ "grad_norm": 7.96875,
430
+ "learning_rate": 1.0905098039215687e-05,
431
+ "loss": 0.547960205078125,
432
+ "step": 2900
433
+ },
434
+ {
435
+ "epoch": 2.313725490196078,
436
+ "grad_norm": 5.875,
437
+ "learning_rate": 1.0748235294117648e-05,
438
+ "loss": 0.5126950836181641,
439
+ "step": 2950
440
+ },
441
+ {
442
+ "epoch": 2.3529411764705883,
443
+ "grad_norm": 6.59375,
444
+ "learning_rate": 1.0591372549019609e-05,
445
+ "loss": 0.5805511474609375,
446
+ "step": 3000
447
+ },
448
+ {
449
+ "epoch": 2.392156862745098,
450
+ "grad_norm": 5.09375,
451
+ "learning_rate": 1.0434509803921568e-05,
452
+ "loss": 0.5385345458984375,
453
+ "step": 3050
454
+ },
455
+ {
456
+ "epoch": 2.431372549019608,
457
+ "grad_norm": 6.71875,
458
+ "learning_rate": 1.027764705882353e-05,
459
+ "loss": 0.5520645141601562,
460
+ "step": 3100
461
+ },
462
+ {
463
+ "epoch": 2.4705882352941178,
464
+ "grad_norm": 5.71875,
465
+ "learning_rate": 1.0120784313725492e-05,
466
+ "loss": 0.526230583190918,
467
+ "step": 3150
468
+ },
469
+ {
470
+ "epoch": 2.5098039215686274,
471
+ "grad_norm": 4.125,
472
+ "learning_rate": 9.963921568627451e-06,
473
+ "loss": 0.4954509735107422,
474
+ "step": 3200
475
+ },
476
+ {
477
+ "epoch": 2.549019607843137,
478
+ "grad_norm": 5.96875,
479
+ "learning_rate": 9.807058823529412e-06,
480
+ "loss": 0.542619743347168,
481
+ "step": 3250
482
+ },
483
+ {
484
+ "epoch": 2.588235294117647,
485
+ "grad_norm": 6.5625,
486
+ "learning_rate": 9.650196078431373e-06,
487
+ "loss": 0.5098217010498047,
488
+ "step": 3300
489
+ },
490
+ {
491
+ "epoch": 2.627450980392157,
492
+ "grad_norm": 8.0,
493
+ "learning_rate": 9.493333333333334e-06,
494
+ "loss": 0.5336943054199219,
495
+ "step": 3350
496
+ },
497
+ {
498
+ "epoch": 2.6666666666666665,
499
+ "grad_norm": 4.78125,
500
+ "learning_rate": 9.336470588235294e-06,
501
+ "loss": 0.5197631072998047,
502
+ "step": 3400
503
+ },
504
+ {
505
+ "epoch": 2.7058823529411766,
506
+ "grad_norm": 5.15625,
507
+ "learning_rate": 9.179607843137257e-06,
508
+ "loss": 0.49386444091796877,
509
+ "step": 3450
510
+ },
511
+ {
512
+ "epoch": 2.7450980392156863,
513
+ "grad_norm": 6.34375,
514
+ "learning_rate": 9.022745098039216e-06,
515
+ "loss": 0.5673449325561524,
516
+ "step": 3500
517
+ },
518
+ {
519
+ "epoch": 2.784313725490196,
520
+ "grad_norm": 6.5625,
521
+ "learning_rate": 8.865882352941177e-06,
522
+ "loss": 0.48973052978515624,
523
+ "step": 3550
524
+ },
525
+ {
526
+ "epoch": 2.8235294117647056,
527
+ "grad_norm": 4.5625,
528
+ "learning_rate": 8.709019607843138e-06,
529
+ "loss": 0.5525439834594726,
530
+ "step": 3600
531
+ },
532
+ {
533
+ "epoch": 2.8627450980392157,
534
+ "grad_norm": 6.21875,
535
+ "learning_rate": 8.552156862745099e-06,
536
+ "loss": 0.5162571716308594,
537
+ "step": 3650
538
+ },
539
+ {
540
+ "epoch": 2.9019607843137254,
541
+ "grad_norm": 5.75,
542
+ "learning_rate": 8.39529411764706e-06,
543
+ "loss": 0.5150326156616211,
544
+ "step": 3700
545
+ },
546
+ {
547
+ "epoch": 2.9411764705882355,
548
+ "grad_norm": 6.375,
549
+ "learning_rate": 8.238431372549021e-06,
550
+ "loss": 0.5412918853759766,
551
+ "step": 3750
552
+ },
553
+ {
554
+ "epoch": 2.980392156862745,
555
+ "grad_norm": 3.703125,
556
+ "learning_rate": 8.08156862745098e-06,
557
+ "loss": 0.47865005493164064,
558
+ "step": 3800
559
+ },
560
+ {
561
+ "epoch": 3.0,
562
+ "eval_loss": 0.5673609972000122,
563
+ "eval_runtime": 0.2837,
564
+ "eval_samples_per_second": 592.149,
565
+ "eval_steps_per_second": 10.574,
566
+ "step": 3825
567
+ },
568
+ {
569
+ "epoch": 3.019607843137255,
570
+ "grad_norm": 4.875,
571
+ "learning_rate": 7.924705882352941e-06,
572
+ "loss": 0.5531900024414063,
573
+ "step": 3850
574
+ },
575
+ {
576
+ "epoch": 3.0588235294117645,
577
+ "grad_norm": 4.96875,
578
+ "learning_rate": 7.767843137254902e-06,
579
+ "loss": 0.5600380706787109,
580
+ "step": 3900
581
+ },
582
+ {
583
+ "epoch": 3.0980392156862746,
584
+ "grad_norm": 7.15625,
585
+ "learning_rate": 7.610980392156863e-06,
586
+ "loss": 0.5152308654785156,
587
+ "step": 3950
588
+ },
589
+ {
590
+ "epoch": 3.1372549019607843,
591
+ "grad_norm": 5.4375,
592
+ "learning_rate": 7.4541176470588236e-06,
593
+ "loss": 0.5683317565917969,
594
+ "step": 4000
595
+ },
596
+ {
597
+ "epoch": 3.176470588235294,
598
+ "grad_norm": 5.4375,
599
+ "learning_rate": 7.2972549019607854e-06,
600
+ "loss": 0.5181949615478516,
601
+ "step": 4050
602
+ },
603
+ {
604
+ "epoch": 3.215686274509804,
605
+ "grad_norm": 4.8125,
606
+ "learning_rate": 7.140392156862746e-06,
607
+ "loss": 0.5614780807495117,
608
+ "step": 4100
609
+ },
610
+ {
611
+ "epoch": 3.2549019607843137,
612
+ "grad_norm": 5.8125,
613
+ "learning_rate": 6.983529411764707e-06,
614
+ "loss": 0.5389285278320313,
615
+ "step": 4150
616
+ },
617
+ {
618
+ "epoch": 3.2941176470588234,
619
+ "grad_norm": 5.03125,
620
+ "learning_rate": 6.826666666666667e-06,
621
+ "loss": 0.5733447265625,
622
+ "step": 4200
623
+ },
624
+ {
625
+ "epoch": 3.3333333333333335,
626
+ "grad_norm": 6.0625,
627
+ "learning_rate": 6.669803921568629e-06,
628
+ "loss": 0.6258538818359375,
629
+ "step": 4250
630
+ },
631
+ {
632
+ "epoch": 3.372549019607843,
633
+ "grad_norm": 7.5,
634
+ "learning_rate": 6.512941176470589e-06,
635
+ "loss": 0.5843626403808594,
636
+ "step": 4300
637
+ },
638
+ {
639
+ "epoch": 3.411764705882353,
640
+ "grad_norm": 5.78125,
641
+ "learning_rate": 6.35607843137255e-06,
642
+ "loss": 0.5407603454589843,
643
+ "step": 4350
644
+ },
645
+ {
646
+ "epoch": 3.450980392156863,
647
+ "grad_norm": 5.375,
648
+ "learning_rate": 6.19921568627451e-06,
649
+ "loss": 0.5244369506835938,
650
+ "step": 4400
651
+ },
652
+ {
653
+ "epoch": 3.4901960784313726,
654
+ "grad_norm": 6.96875,
655
+ "learning_rate": 6.042352941176471e-06,
656
+ "loss": 0.5704220581054688,
657
+ "step": 4450
658
+ },
659
+ {
660
+ "epoch": 3.5294117647058822,
661
+ "grad_norm": 5.625,
662
+ "learning_rate": 5.885490196078431e-06,
663
+ "loss": 0.5458635711669921,
664
+ "step": 4500
665
+ },
666
+ {
667
+ "epoch": 3.568627450980392,
668
+ "grad_norm": 7.65625,
669
+ "learning_rate": 5.728627450980393e-06,
670
+ "loss": 0.5404665756225586,
671
+ "step": 4550
672
+ },
673
+ {
674
+ "epoch": 3.607843137254902,
675
+ "grad_norm": 5.71875,
676
+ "learning_rate": 5.571764705882353e-06,
677
+ "loss": 0.5152708053588867,
678
+ "step": 4600
679
+ },
680
+ {
681
+ "epoch": 3.6470588235294117,
682
+ "grad_norm": 4.34375,
683
+ "learning_rate": 5.414901960784314e-06,
684
+ "loss": 0.5370127868652343,
685
+ "step": 4650
686
+ },
687
+ {
688
+ "epoch": 3.686274509803922,
689
+ "grad_norm": 6.28125,
690
+ "learning_rate": 5.2580392156862745e-06,
691
+ "loss": 0.5682061767578125,
692
+ "step": 4700
693
+ },
694
+ {
695
+ "epoch": 3.7254901960784315,
696
+ "grad_norm": 6.40625,
697
+ "learning_rate": 5.101176470588236e-06,
698
+ "loss": 0.566998291015625,
699
+ "step": 4750
700
+ },
701
+ {
702
+ "epoch": 3.764705882352941,
703
+ "grad_norm": 5.5,
704
+ "learning_rate": 4.9443137254901966e-06,
705
+ "loss": 0.5921047973632813,
706
+ "step": 4800
707
+ },
708
+ {
709
+ "epoch": 3.803921568627451,
710
+ "grad_norm": 4.1875,
711
+ "learning_rate": 4.787450980392157e-06,
712
+ "loss": 0.5128079223632812,
713
+ "step": 4850
714
+ },
715
+ {
716
+ "epoch": 3.843137254901961,
717
+ "grad_norm": 5.65625,
718
+ "learning_rate": 4.630588235294118e-06,
719
+ "loss": 0.5532876586914063,
720
+ "step": 4900
721
+ },
722
+ {
723
+ "epoch": 3.8823529411764706,
724
+ "grad_norm": 6.875,
725
+ "learning_rate": 4.473725490196079e-06,
726
+ "loss": 0.5848858642578125,
727
+ "step": 4950
728
+ },
729
+ {
730
+ "epoch": 3.9215686274509802,
731
+ "grad_norm": 5.375,
732
+ "learning_rate": 4.316862745098039e-06,
733
+ "loss": 0.5643892669677735,
734
+ "step": 5000
735
+ },
736
+ {
737
+ "epoch": 3.9607843137254903,
738
+ "grad_norm": 7.03125,
739
+ "learning_rate": 4.16e-06,
740
+ "loss": 0.565797348022461,
741
+ "step": 5050
742
+ },
743
+ {
744
+ "epoch": 4.0,
745
+ "grad_norm": 10.3125,
746
+ "learning_rate": 4.003137254901961e-06,
747
+ "loss": 0.5881293869018555,
748
+ "step": 5100
749
+ },
750
+ {
751
+ "epoch": 4.0,
752
+ "eval_loss": 0.5672411322593689,
753
+ "eval_runtime": 0.2361,
754
+ "eval_samples_per_second": 711.627,
755
+ "eval_steps_per_second": 12.708,
756
+ "step": 5100
757
+ },
758
+ {
759
+ "epoch": 4.03921568627451,
760
+ "grad_norm": 4.4375,
761
+ "learning_rate": 3.846274509803922e-06,
762
+ "loss": 0.5876416015625,
763
+ "step": 5150
764
+ },
765
+ {
766
+ "epoch": 4.078431372549019,
767
+ "grad_norm": 6.1875,
768
+ "learning_rate": 3.6894117647058826e-06,
769
+ "loss": 0.5970059204101562,
770
+ "step": 5200
771
+ },
772
+ {
773
+ "epoch": 4.117647058823529,
774
+ "grad_norm": 6.375,
775
+ "learning_rate": 3.5325490196078432e-06,
776
+ "loss": 0.6093558502197266,
777
+ "step": 5250
778
+ },
779
+ {
780
+ "epoch": 4.1568627450980395,
781
+ "grad_norm": 6.75,
782
+ "learning_rate": 3.375686274509804e-06,
783
+ "loss": 0.5946432113647461,
784
+ "step": 5300
785
+ },
786
+ {
787
+ "epoch": 4.196078431372549,
788
+ "grad_norm": 6.03125,
789
+ "learning_rate": 3.218823529411765e-06,
790
+ "loss": 0.5788070678710937,
791
+ "step": 5350
792
+ },
793
+ {
794
+ "epoch": 4.235294117647059,
795
+ "grad_norm": 5.0625,
796
+ "learning_rate": 3.0619607843137255e-06,
797
+ "loss": 0.5740170288085937,
798
+ "step": 5400
799
+ },
800
+ {
801
+ "epoch": 4.2745098039215685,
802
+ "grad_norm": 7.46875,
803
+ "learning_rate": 2.9050980392156865e-06,
804
+ "loss": 0.571316795349121,
805
+ "step": 5450
806
+ },
807
+ {
808
+ "epoch": 4.313725490196078,
809
+ "grad_norm": 6.59375,
810
+ "learning_rate": 2.748235294117647e-06,
811
+ "loss": 0.5775479125976563,
812
+ "step": 5500
813
+ },
814
+ {
815
+ "epoch": 4.352941176470588,
816
+ "grad_norm": 5.03125,
817
+ "learning_rate": 2.5913725490196077e-06,
818
+ "loss": 0.6153933715820312,
819
+ "step": 5550
820
+ },
821
+ {
822
+ "epoch": 4.392156862745098,
823
+ "grad_norm": 5.09375,
824
+ "learning_rate": 2.434509803921569e-06,
825
+ "loss": 0.5949542236328125,
826
+ "step": 5600
827
+ },
828
+ {
829
+ "epoch": 4.431372549019608,
830
+ "grad_norm": 7.25,
831
+ "learning_rate": 2.2776470588235297e-06,
832
+ "loss": 0.59134033203125,
833
+ "step": 5650
834
+ },
835
+ {
836
+ "epoch": 4.470588235294118,
837
+ "grad_norm": 6.5,
838
+ "learning_rate": 2.1207843137254903e-06,
839
+ "loss": 0.5656556701660156,
840
+ "step": 5700
841
+ },
842
+ {
843
+ "epoch": 4.509803921568627,
844
+ "grad_norm": 4.625,
845
+ "learning_rate": 1.9639215686274514e-06,
846
+ "loss": 0.6088162231445312,
847
+ "step": 5750
848
+ },
849
+ {
850
+ "epoch": 4.549019607843137,
851
+ "grad_norm": 5.53125,
852
+ "learning_rate": 1.807058823529412e-06,
853
+ "loss": 0.5470469665527343,
854
+ "step": 5800
855
+ },
856
+ {
857
+ "epoch": 4.588235294117647,
858
+ "grad_norm": 6.15625,
859
+ "learning_rate": 1.6501960784313728e-06,
860
+ "loss": 0.6141192245483399,
861
+ "step": 5850
862
+ },
863
+ {
864
+ "epoch": 4.627450980392156,
865
+ "grad_norm": 6.46875,
866
+ "learning_rate": 1.4933333333333336e-06,
867
+ "loss": 0.5964439392089844,
868
+ "step": 5900
869
+ },
870
+ {
871
+ "epoch": 4.666666666666667,
872
+ "grad_norm": 5.875,
873
+ "learning_rate": 1.3364705882352944e-06,
874
+ "loss": 0.5732437515258789,
875
+ "step": 5950
876
+ },
877
+ {
878
+ "epoch": 4.705882352941177,
879
+ "grad_norm": 5.03125,
880
+ "learning_rate": 1.179607843137255e-06,
881
+ "loss": 0.5962744140625,
882
+ "step": 6000
883
+ },
884
+ {
885
+ "epoch": 4.745098039215686,
886
+ "grad_norm": 5.875,
887
+ "learning_rate": 1.0227450980392158e-06,
888
+ "loss": 0.5646636962890625,
889
+ "step": 6050
890
+ },
891
+ {
892
+ "epoch": 4.784313725490196,
893
+ "grad_norm": 4.875,
894
+ "learning_rate": 8.658823529411765e-07,
895
+ "loss": 0.5924723815917968,
896
+ "step": 6100
897
+ },
898
+ {
899
+ "epoch": 4.823529411764706,
900
+ "grad_norm": 4.1875,
901
+ "learning_rate": 7.090196078431373e-07,
902
+ "loss": 0.5885520935058594,
903
+ "step": 6150
904
+ },
905
+ {
906
+ "epoch": 4.862745098039216,
907
+ "grad_norm": 7.90625,
908
+ "learning_rate": 5.521568627450981e-07,
909
+ "loss": 0.6044607543945313,
910
+ "step": 6200
911
+ },
912
+ {
913
+ "epoch": 4.901960784313726,
914
+ "grad_norm": 5.375,
915
+ "learning_rate": 3.9529411764705884e-07,
916
+ "loss": 0.613191909790039,
917
+ "step": 6250
918
+ },
919
+ {
920
+ "epoch": 4.9411764705882355,
921
+ "grad_norm": 5.28125,
922
+ "learning_rate": 2.3843137254901965e-07,
923
+ "loss": 0.6047885131835937,
924
+ "step": 6300
925
+ },
926
+ {
927
+ "epoch": 4.980392156862745,
928
+ "grad_norm": 4.9375,
929
+ "learning_rate": 8.156862745098041e-08,
930
+ "loss": 0.5925003051757812,
931
+ "step": 6350
932
+ },
933
+ {
934
+ "epoch": 5.0,
935
+ "eval_loss": 0.5668168067932129,
936
+ "eval_runtime": 0.2338,
937
+ "eval_samples_per_second": 718.45,
938
+ "eval_steps_per_second": 12.829,
939
+ "step": 6375
940
+ }
941
+ ],
942
+ "logging_steps": 50,
943
+ "max_steps": 6375,
944
+ "num_input_tokens_seen": 0,
945
+ "num_train_epochs": 5,
946
+ "save_steps": 500,
947
+ "stateful_callbacks": {
948
+ "TrainerControl": {
949
+ "args": {
950
+ "should_epoch_stop": false,
951
+ "should_evaluate": false,
952
+ "should_log": false,
953
+ "should_save": true,
954
+ "should_training_stop": true
955
+ },
956
+ "attributes": {}
957
+ }
958
+ },
959
+ "total_flos": 0.0,
960
+ "train_batch_size": 32,
961
+ "trial_name": null,
962
+ "trial_params": null
963
+ }
cross-encoder-ms-marco-MiniLM-L-6-v2-E5-Pairwise/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f6e4f82fd68ad5fd3ad84298fa5c82de5947790f3858093c6891d2455ad2980
3
+ size 5265