Hugging Face's logo

Stevenf232
/
hypencoder_8layer_SapBERT_context

PyTorch
Model card Files
xet
Community
Stevenf232 commited on
Commit
41282e7
·
verified ·
1 Parent(s): de58ce9

Upload trained model from Colab

Browse files
Files changed (12) hide show
  1. config.json +46 -0
  2. optimizer.pt +3 -0
  3. pytorch_model.bin +3 -0
  4. rng_state.pth +3 -0
  5. scaler.pt +3 -0
  6. scheduler.pt +3 -0
  7. special_tokens_map.json +7 -0
  8. tokenizer.json +0 -0
  9. tokenizer_config.json +59 -0
  10. trainer_state.json +334 -0
  11. training_args.bin +3 -0
  12. vocab.txt +0 -0
config.json ADDED
@@ -0,0 +1,46 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "HypencoderDualEncoder"
4
+ ],
5
+ "loss_kwargs": [
6
+ {
7
+ "only_use_first_item": false,
8
+ "use_in_batch_negatives": true
9
+ }
10
+ ],
11
+ "loss_type": [
12
+ "cross_entropy"
13
+ ],
14
+ "passage_encoder_kwargs": {
15
+ "freeze_transformer": false,
16
+ "model_name_or_path": "cambridgeltl/SapBERT-from-PubMedBERT-fulltext",
17
+ "pooling_type": "cls"
18
+ },
19
+ "passage_encoder_type": "",
20
+ "query_encoder_kwargs": {
21
+ "base_encoder_output_dim": 768,
22
+ "converter_kwargs": {
23
+ "activation_type": "relu",
24
+ "do_residual_on_last": false,
25
+ "vector_dimensions": [
26
+ 768,
27
+ 768,
28
+ 768,
29
+ 768,
30
+ 768,
31
+ 768,
32
+ 768,
33
+ 768,
34
+ 768,
35
+ 1
36
+ ]
37
+ },
38
+ "embedding_representation": null,
39
+ "freeze_transformer": false,
40
+ "model_name_or_path": "cambridgeltl/SapBERT-from-PubMedBERT-fulltext"
41
+ },
42
+ "query_encoder_type": "",
43
+ "shared_encoder": true,
44
+ "torch_dtype": "float32",
45
+ "transformers_version": "4.50.0"
46
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3133a26cc7ff413d41f19f454bc617ed14cf04da5d6fb285585fa72e14a0327
3
+ size 1187982411
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:858a5e57ce0063ca98017cd72b8eba21b31e1edfb540e538d5c3a8c78b8f3ce9
3
+ size 596364767
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfd8e669c36c6d3a4e4246c3e60675ed9c17fda37cfcf0a3134e73426741f37f
3
+ size 14645
scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b29029961a0544216116894b9b63bdfcf5345da31d71ade11acc37f64a54d0c8
3
+ size 1383
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70b9e8e4330d5f36cb2267aeed325e90ca0fbd87a3eefb66e476ecbfdaea635b
3
+ size 1465
special_tokens_map.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "cls_token": "[CLS]",
3
+ "mask_token": "[MASK]",
4
+ "pad_token": "[PAD]",
5
+ "sep_token": "[SEP]",
6
+ "unk_token": "[UNK]"
7
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "[PAD]",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "[UNK]",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "[CLS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "[SEP]",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "4": {
36
+ "content": "[MASK]",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ }
43
+ },
44
+ "clean_up_tokenization_spaces": true,
45
+ "cls_token": "[CLS]",
46
+ "do_basic_tokenize": true,
47
+ "do_lower_case": true,
48
+ "extra_special_tokens": {},
49
+ "full_tokenizer_file": null,
50
+ "mask_token": "[MASK]",
51
+ "model_max_length": 1000000000000000019884624838656,
52
+ "never_split": null,
53
+ "pad_token": "[PAD]",
54
+ "sep_token": "[SEP]",
55
+ "strip_accents": null,
56
+ "tokenize_chinese_chars": true,
57
+ "tokenizer_class": "BertTokenizer",
58
+ "unk_token": "[UNK]"
59
+ }
trainer_state.json ADDED
@@ -0,0 +1,334 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 2.3529411764705883,
6
+ "eval_steps": 10,
7
+ "global_step": 200,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 0.11764705882352941,
14
+ "grad_norm": 169.93887329101562,
15
+ "learning_rate": 6e-06,
16
+ "loss": 4.8387,
17
+ "step": 10
18
+ },
19
+ {
20
+ "epoch": 0.11764705882352941,
21
+ "eval_loss": 4.6919026374816895,
22
+ "eval_runtime": 15.5369,
23
+ "eval_samples_per_second": 350.456,
24
+ "eval_steps_per_second": 5.535,
25
+ "step": 10
26
+ },
27
+ {
28
+ "epoch": 0.23529411764705882,
29
+ "grad_norm": 103.59686279296875,
30
+ "learning_rate": 1.6000000000000003e-05,
31
+ "loss": 4.3444,
32
+ "step": 20
33
+ },
34
+ {
35
+ "epoch": 0.23529411764705882,
36
+ "eval_loss": 3.9225239753723145,
37
+ "eval_runtime": 16.2767,
38
+ "eval_samples_per_second": 334.526,
39
+ "eval_steps_per_second": 5.284,
40
+ "step": 20
41
+ },
42
+ {
43
+ "epoch": 0.35294117647058826,
44
+ "grad_norm": 96.45588684082031,
45
+ "learning_rate": 1.9333333333333333e-05,
46
+ "loss": 3.5537,
47
+ "step": 30
48
+ },
49
+ {
50
+ "epoch": 0.35294117647058826,
51
+ "eval_loss": 3.1923811435699463,
52
+ "eval_runtime": 17.7651,
53
+ "eval_samples_per_second": 306.5,
54
+ "eval_steps_per_second": 4.841,
55
+ "step": 30
56
+ },
57
+ {
58
+ "epoch": 0.47058823529411764,
59
+ "grad_norm": 116.93628692626953,
60
+ "learning_rate": 1.8222222222222224e-05,
61
+ "loss": 2.4677,
62
+ "step": 40
63
+ },
64
+ {
65
+ "epoch": 0.47058823529411764,
66
+ "eval_loss": 2.3630847930908203,
67
+ "eval_runtime": 19.3532,
68
+ "eval_samples_per_second": 281.349,
69
+ "eval_steps_per_second": 4.444,
70
+ "step": 40
71
+ },
72
+ {
73
+ "epoch": 0.5882352941176471,
74
+ "grad_norm": 123.0512924194336,
75
+ "learning_rate": 1.7111111111111112e-05,
76
+ "loss": 1.5596,
77
+ "step": 50
78
+ },
79
+ {
80
+ "epoch": 0.5882352941176471,
81
+ "eval_loss": 1.779540777206421,
82
+ "eval_runtime": 17.9711,
83
+ "eval_samples_per_second": 302.987,
84
+ "eval_steps_per_second": 4.785,
85
+ "step": 50
86
+ },
87
+ {
88
+ "epoch": 0.7058823529411765,
89
+ "grad_norm": 116.81210327148438,
90
+ "learning_rate": 1.6000000000000003e-05,
91
+ "loss": 0.9314,
92
+ "step": 60
93
+ },
94
+ {
95
+ "epoch": 0.7058823529411765,
96
+ "eval_loss": 1.3096877336502075,
97
+ "eval_runtime": 17.7299,
98
+ "eval_samples_per_second": 307.109,
99
+ "eval_steps_per_second": 4.851,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.8235294117647058,
104
+ "grad_norm": 78.21733093261719,
105
+ "learning_rate": 1.488888888888889e-05,
106
+ "loss": 0.4562,
107
+ "step": 70
108
+ },
109
+ {
110
+ "epoch": 0.8235294117647058,
111
+ "eval_loss": 1.0128556489944458,
112
+ "eval_runtime": 18.4122,
113
+ "eval_samples_per_second": 295.729,
114
+ "eval_steps_per_second": 4.671,
115
+ "step": 70
116
+ },
117
+ {
118
+ "epoch": 0.9411764705882353,
119
+ "grad_norm": 65.50968933105469,
120
+ "learning_rate": 1.377777777777778e-05,
121
+ "loss": 0.3587,
122
+ "step": 80
123
+ },
124
+ {
125
+ "epoch": 0.9411764705882353,
126
+ "eval_loss": 0.9688291549682617,
127
+ "eval_runtime": 18.5333,
128
+ "eval_samples_per_second": 293.795,
129
+ "eval_steps_per_second": 4.64,
130
+ "step": 80
131
+ },
132
+ {
133
+ "epoch": 1.0588235294117647,
134
+ "grad_norm": 61.12400817871094,
135
+ "learning_rate": 1.2666666666666667e-05,
136
+ "loss": 0.3667,
137
+ "step": 90
138
+ },
139
+ {
140
+ "epoch": 1.0588235294117647,
141
+ "eval_loss": 0.9096461534500122,
142
+ "eval_runtime": 18.1026,
143
+ "eval_samples_per_second": 300.785,
144
+ "eval_steps_per_second": 4.751,
145
+ "step": 90
146
+ },
147
+ {
148
+ "epoch": 1.1764705882352942,
149
+ "grad_norm": 50.41886901855469,
150
+ "learning_rate": 1.1555555555555556e-05,
151
+ "loss": 0.3104,
152
+ "step": 100
153
+ },
154
+ {
155
+ "epoch": 1.1764705882352942,
156
+ "eval_loss": 0.9075976610183716,
157
+ "eval_runtime": 18.2417,
158
+ "eval_samples_per_second": 298.492,
159
+ "eval_steps_per_second": 4.714,
160
+ "step": 100
161
+ },
162
+ {
163
+ "epoch": 1.2941176470588236,
164
+ "grad_norm": 49.723411560058594,
165
+ "learning_rate": 1.0444444444444445e-05,
166
+ "loss": 0.3057,
167
+ "step": 110
168
+ },
169
+ {
170
+ "epoch": 1.2941176470588236,
171
+ "eval_loss": 0.8963654637336731,
172
+ "eval_runtime": 18.3711,
173
+ "eval_samples_per_second": 296.389,
174
+ "eval_steps_per_second": 4.681,
175
+ "step": 110
176
+ },
177
+ {
178
+ "epoch": 1.4117647058823528,
179
+ "grad_norm": 66.96435546875,
180
+ "learning_rate": 9.333333333333334e-06,
181
+ "loss": 0.2852,
182
+ "step": 120
183
+ },
184
+ {
185
+ "epoch": 1.4117647058823528,
186
+ "eval_loss": 0.8938003778457642,
187
+ "eval_runtime": 18.4942,
188
+ "eval_samples_per_second": 294.417,
189
+ "eval_steps_per_second": 4.65,
190
+ "step": 120
191
+ },
192
+ {
193
+ "epoch": 1.5294117647058822,
194
+ "grad_norm": 67.33085632324219,
195
+ "learning_rate": 8.222222222222222e-06,
196
+ "loss": 0.2527,
197
+ "step": 130
198
+ },
199
+ {
200
+ "epoch": 1.5294117647058822,
201
+ "eval_loss": 0.9134606122970581,
202
+ "eval_runtime": 18.4294,
203
+ "eval_samples_per_second": 295.452,
204
+ "eval_steps_per_second": 4.666,
205
+ "step": 130
206
+ },
207
+ {
208
+ "epoch": 1.6470588235294117,
209
+ "grad_norm": 60.442684173583984,
210
+ "learning_rate": 7.111111111111112e-06,
211
+ "loss": 0.3592,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 1.6470588235294117,
216
+ "eval_loss": 0.8834967017173767,
217
+ "eval_runtime": 18.3629,
218
+ "eval_samples_per_second": 296.522,
219
+ "eval_steps_per_second": 4.683,
220
+ "step": 140
221
+ },
222
+ {
223
+ "epoch": 1.7647058823529411,
224
+ "grad_norm": 59.17316818237305,
225
+ "learning_rate": 6e-06,
226
+ "loss": 0.1998,
227
+ "step": 150
228
+ },
229
+ {
230
+ "epoch": 1.7647058823529411,
231
+ "eval_loss": 0.862636923789978,
232
+ "eval_runtime": 18.2959,
233
+ "eval_samples_per_second": 297.608,
234
+ "eval_steps_per_second": 4.701,
235
+ "step": 150
236
+ },
237
+ {
238
+ "epoch": 1.8823529411764706,
239
+ "grad_norm": 33.52590560913086,
240
+ "learning_rate": 4.888888888888889e-06,
241
+ "loss": 0.2258,
242
+ "step": 160
243
+ },
244
+ {
245
+ "epoch": 1.8823529411764706,
246
+ "eval_loss": 0.8706350326538086,
247
+ "eval_runtime": 18.3223,
248
+ "eval_samples_per_second": 297.179,
249
+ "eval_steps_per_second": 4.694,
250
+ "step": 160
251
+ },
252
+ {
253
+ "epoch": 2.0,
254
+ "grad_norm": 48.038719177246094,
255
+ "learning_rate": 3.777777777777778e-06,
256
+ "loss": 0.1933,
257
+ "step": 170
258
+ },
259
+ {
260
+ "epoch": 2.0,
261
+ "eval_loss": 0.875869870185852,
262
+ "eval_runtime": 18.2474,
263
+ "eval_samples_per_second": 298.399,
264
+ "eval_steps_per_second": 4.713,
265
+ "step": 170
266
+ },
267
+ {
268
+ "epoch": 2.1176470588235294,
269
+ "grad_norm": 42.39509201049805,
270
+ "learning_rate": 2.666666666666667e-06,
271
+ "loss": 0.1987,
272
+ "step": 180
273
+ },
274
+ {
275
+ "epoch": 2.1176470588235294,
276
+ "eval_loss": 0.875755250453949,
277
+ "eval_runtime": 18.2272,
278
+ "eval_samples_per_second": 298.729,
279
+ "eval_steps_per_second": 4.718,
280
+ "step": 180
281
+ },
282
+ {
283
+ "epoch": 2.235294117647059,
284
+ "grad_norm": 24.651744842529297,
285
+ "learning_rate": 1.5555555555555558e-06,
286
+ "loss": 0.1684,
287
+ "step": 190
288
+ },
289
+ {
290
+ "epoch": 2.235294117647059,
291
+ "eval_loss": 0.8625762462615967,
292
+ "eval_runtime": 18.1576,
293
+ "eval_samples_per_second": 299.874,
294
+ "eval_steps_per_second": 4.736,
295
+ "step": 190
296
+ },
297
+ {
298
+ "epoch": 2.3529411764705883,
299
+ "grad_norm": 27.64773941040039,
300
+ "learning_rate": 4.444444444444445e-07,
301
+ "loss": 0.1612,
302
+ "step": 200
303
+ },
304
+ {
305
+ "epoch": 2.3529411764705883,
306
+ "eval_loss": 0.8577004075050354,
307
+ "eval_runtime": 18.1098,
308
+ "eval_samples_per_second": 300.666,
309
+ "eval_steps_per_second": 4.749,
310
+ "step": 200
311
+ }
312
+ ],
313
+ "logging_steps": 10,
314
+ "max_steps": 200,
315
+ "num_input_tokens_seen": 0,
316
+ "num_train_epochs": 3,
317
+ "save_steps": 2500,
318
+ "stateful_callbacks": {
319
+ "TrainerControl": {
320
+ "args": {
321
+ "should_epoch_stop": false,
322
+ "should_evaluate": false,
323
+ "should_log": false,
324
+ "should_save": true,
325
+ "should_training_stop": true
326
+ },
327
+ "attributes": {}
328
+ }
329
+ },
330
+ "total_flos": 0.0,
331
+ "train_batch_size": 64,
332
+ "trial_name": null,
333
+ "trial_params": null
334
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ec332a8e89e7566e54e694db9abb34df273e201d5d0e5ea19a4a4ff7b47b0076
3
+ size 5713
vocab.txt ADDED
The diff for this file is too large to render. See raw diff