theblackcat102 commited on
Commit
b29044e
1 Parent(s): ce82195

push model at 1500

Browse files
added_tokens.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "<|assistant|>": 128002,
3
+ "<|prefix_begin|>": 128001,
4
+ "<|prefix_end|>": 128004,
5
+ "<|prompter|>": 128005,
6
+ "<|system|>": 128003,
7
+ "[MASK]": 128000
8
+ }
config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/deberta-v3-large",
3
+ "architectures": [
4
+ "DebertaV2ForSequenceClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "hidden_act": "gelu",
8
+ "hidden_dropout_prob": 0.1,
9
+ "hidden_size": 1024,
10
+ "id2label": {
11
+ "0": "LABEL_0",
12
+ "1": "LABEL_1",
13
+ "2": "LABEL_2",
14
+ "3": "LABEL_3",
15
+ "4": "LABEL_4",
16
+ "5": "LABEL_5",
17
+ "6": "LABEL_6",
18
+ "7": "LABEL_7",
19
+ "8": "LABEL_8"
20
+ },
21
+ "initializer_range": 0.02,
22
+ "intermediate_size": 4096,
23
+ "label2id": {
24
+ "LABEL_0": 0,
25
+ "LABEL_1": 1,
26
+ "LABEL_2": 2,
27
+ "LABEL_3": 3,
28
+ "LABEL_4": 4,
29
+ "LABEL_5": 5,
30
+ "LABEL_6": 6,
31
+ "LABEL_7": 7,
32
+ "LABEL_8": 8
33
+ },
34
+ "layer_norm_eps": 1e-07,
35
+ "max_position_embeddings": 512,
36
+ "max_relative_positions": -1,
37
+ "model_type": "deberta-v2",
38
+ "norm_rel_ebd": "layer_norm",
39
+ "num_attention_heads": 16,
40
+ "num_hidden_layers": 24,
41
+ "pad_token_id": 0,
42
+ "pooler_dropout": 0,
43
+ "pooler_hidden_act": "gelu",
44
+ "pooler_hidden_size": 1024,
45
+ "pos_att_type": [
46
+ "p2c",
47
+ "c2p"
48
+ ],
49
+ "position_biased_input": false,
50
+ "position_buckets": 256,
51
+ "relative_attention": true,
52
+ "share_att_key": true,
53
+ "torch_dtype": "float16",
54
+ "transformers_version": "4.28.0.dev0",
55
+ "type_vocab_size": 0,
56
+ "vocab_size": 128100
57
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e5af746ffc2697e5e393c527dc75fb07d08044408f0c6675e575495161e38425
3
+ size 870203519
special_tokens_map.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|prefix_begin|>",
4
+ "<|assistant|>",
5
+ "<|system|>",
6
+ "<|prefix_end|>",
7
+ "<|prompter|>"
8
+ ],
9
+ "bos_token": "[CLS]",
10
+ "cls_token": "[CLS]",
11
+ "eos_token": "[SEP]",
12
+ "mask_token": "[MASK]",
13
+ "pad_token": "[PAD]",
14
+ "sep_token": "[CLS]",
15
+ "unk_token": "[UNK]"
16
+ }
spm.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c679fbf93643d19aab7ee10c0b99e460bdbc02fedf34b92b05af343b4af586fd
3
+ size 2464616
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "[CLS]",
3
+ "clean_up_tokenization_spaces": true,
4
+ "cls_token": "[CLS]",
5
+ "do_lower_case": false,
6
+ "eos_token": "[SEP]",
7
+ "mask_token": "[MASK]",
8
+ "model_max_length": 1000000000000000019884624838656,
9
+ "pad_token": "[PAD]",
10
+ "sep_token": "[SEP]",
11
+ "sp_model_kwargs": {},
12
+ "special_tokens_map_file": null,
13
+ "split_by_punct": false,
14
+ "tokenizer_class": "DebertaV2Tokenizer",
15
+ "unk_token": "[UNK]",
16
+ "vocab_type": "spm"
17
+ }
trainer_state.json ADDED
@@ -0,0 +1,981 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.8449049873931491,
5
+ "global_step": 1500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 4.5801353078493935e-06,
13
+ "loss": 0.8726,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.02,
18
+ "learning_rate": 7.087352805422317e-06,
19
+ "loss": 0.8672,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.04,
24
+ "learning_rate": 8.228161798644422e-06,
25
+ "loss": 0.8565,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.05,
30
+ "learning_rate": 9.014161010104347e-06,
31
+ "loss": 0.8213,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.06,
36
+ "learning_rate": 9.673229499590088e-06,
37
+ "loss": 0.7977,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.07,
42
+ "learning_rate": 9.987442444537463e-06,
43
+ "loss": 0.7521,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.09,
48
+ "learning_rate": 9.94558392632901e-06,
49
+ "loss": 0.7423,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.1,
54
+ "learning_rate": 9.903725408120554e-06,
55
+ "loss": 0.7026,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.11,
60
+ "learning_rate": 9.861866889912098e-06,
61
+ "loss": 0.6808,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.12,
66
+ "learning_rate": 9.820008371703642e-06,
67
+ "loss": 0.6901,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.14,
72
+ "learning_rate": 9.778149853495186e-06,
73
+ "loss": 0.6514,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.15,
78
+ "learning_rate": 9.736291335286732e-06,
79
+ "loss": 0.6467,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.16,
84
+ "learning_rate": 9.694432817078277e-06,
85
+ "loss": 0.6641,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.17,
90
+ "learning_rate": 9.652574298869821e-06,
91
+ "loss": 0.6616,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.18,
96
+ "learning_rate": 9.610715780661365e-06,
97
+ "loss": 0.6617,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.2,
102
+ "learning_rate": 9.56885726245291e-06,
103
+ "loss": 0.6295,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.21,
108
+ "learning_rate": 9.526998744244454e-06,
109
+ "loss": 0.6071,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.22,
114
+ "learning_rate": 9.485140226036e-06,
115
+ "loss": 0.6354,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.23,
120
+ "learning_rate": 9.443281707827544e-06,
121
+ "loss": 0.6211,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.25,
126
+ "learning_rate": 9.401423189619088e-06,
127
+ "loss": 0.6272,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.26,
132
+ "learning_rate": 9.359564671410633e-06,
133
+ "loss": 0.641,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.27,
138
+ "learning_rate": 9.317706153202177e-06,
139
+ "loss": 0.6205,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.28,
144
+ "learning_rate": 9.275847634993721e-06,
145
+ "loss": 0.6263,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.3,
150
+ "learning_rate": 9.233989116785267e-06,
151
+ "loss": 0.6148,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.31,
156
+ "learning_rate": 9.192130598576812e-06,
157
+ "loss": 0.6521,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.31,
162
+ "eval_oasst_export_w_label_accuracy": 0.7346153846153847,
163
+ "eval_oasst_export_w_label_kendalltau": 0.3859706959706959,
164
+ "eval_oasst_export_w_label_loss": 0.626953125,
165
+ "eval_oasst_export_w_label_neg_score": 0.2529296875,
166
+ "eval_oasst_export_w_label_pos_score": 0.91064453125,
167
+ "eval_oasst_export_w_label_runtime": 190.5015,
168
+ "eval_oasst_export_w_label_samples_per_second": 9.554,
169
+ "eval_oasst_export_w_label_score_diff": 0.65771484375,
170
+ "eval_oasst_export_w_label_steps_per_second": 2.388,
171
+ "step": 251
172
+ },
173
+ {
174
+ "epoch": 0.32,
175
+ "learning_rate": 9.150272080368356e-06,
176
+ "loss": 0.602,
177
+ "step": 260
178
+ },
179
+ {
180
+ "epoch": 0.33,
181
+ "learning_rate": 9.1084135621599e-06,
182
+ "loss": 0.5637,
183
+ "step": 270
184
+ },
185
+ {
186
+ "epoch": 0.34,
187
+ "learning_rate": 9.066555043951444e-06,
188
+ "loss": 0.5948,
189
+ "step": 280
190
+ },
191
+ {
192
+ "epoch": 0.36,
193
+ "learning_rate": 9.02469652574299e-06,
194
+ "loss": 0.6245,
195
+ "step": 290
196
+ },
197
+ {
198
+ "epoch": 0.37,
199
+ "learning_rate": 8.982838007534535e-06,
200
+ "loss": 0.5943,
201
+ "step": 300
202
+ },
203
+ {
204
+ "epoch": 0.38,
205
+ "learning_rate": 8.940979489326079e-06,
206
+ "loss": 0.5934,
207
+ "step": 310
208
+ },
209
+ {
210
+ "epoch": 0.39,
211
+ "learning_rate": 8.899120971117623e-06,
212
+ "loss": 0.5426,
213
+ "step": 320
214
+ },
215
+ {
216
+ "epoch": 0.41,
217
+ "learning_rate": 8.857262452909168e-06,
218
+ "loss": 0.6191,
219
+ "step": 330
220
+ },
221
+ {
222
+ "epoch": 0.42,
223
+ "learning_rate": 8.815403934700712e-06,
224
+ "loss": 0.6316,
225
+ "step": 340
226
+ },
227
+ {
228
+ "epoch": 0.43,
229
+ "learning_rate": 8.773545416492258e-06,
230
+ "loss": 0.611,
231
+ "step": 350
232
+ },
233
+ {
234
+ "epoch": 0.44,
235
+ "learning_rate": 8.7316868982838e-06,
236
+ "loss": 0.6434,
237
+ "step": 360
238
+ },
239
+ {
240
+ "epoch": 0.46,
241
+ "learning_rate": 8.689828380075346e-06,
242
+ "loss": 0.6327,
243
+ "step": 370
244
+ },
245
+ {
246
+ "epoch": 0.47,
247
+ "learning_rate": 8.64796986186689e-06,
248
+ "loss": 0.6021,
249
+ "step": 380
250
+ },
251
+ {
252
+ "epoch": 0.48,
253
+ "learning_rate": 8.606111343658435e-06,
254
+ "loss": 0.5986,
255
+ "step": 390
256
+ },
257
+ {
258
+ "epoch": 0.49,
259
+ "learning_rate": 8.564252825449981e-06,
260
+ "loss": 0.5798,
261
+ "step": 400
262
+ },
263
+ {
264
+ "epoch": 0.5,
265
+ "learning_rate": 8.522394307241524e-06,
266
+ "loss": 0.6055,
267
+ "step": 410
268
+ },
269
+ {
270
+ "epoch": 0.52,
271
+ "learning_rate": 8.480535789033068e-06,
272
+ "loss": 0.6172,
273
+ "step": 420
274
+ },
275
+ {
276
+ "epoch": 0.53,
277
+ "learning_rate": 8.438677270824614e-06,
278
+ "loss": 0.6657,
279
+ "step": 430
280
+ },
281
+ {
282
+ "epoch": 0.54,
283
+ "learning_rate": 8.396818752616158e-06,
284
+ "loss": 0.6206,
285
+ "step": 440
286
+ },
287
+ {
288
+ "epoch": 0.55,
289
+ "learning_rate": 8.354960234407702e-06,
290
+ "loss": 0.6177,
291
+ "step": 450
292
+ },
293
+ {
294
+ "epoch": 0.57,
295
+ "learning_rate": 8.313101716199248e-06,
296
+ "loss": 0.6201,
297
+ "step": 460
298
+ },
299
+ {
300
+ "epoch": 0.58,
301
+ "learning_rate": 8.271243197990791e-06,
302
+ "loss": 0.5299,
303
+ "step": 470
304
+ },
305
+ {
306
+ "epoch": 0.59,
307
+ "learning_rate": 8.229384679782337e-06,
308
+ "loss": 0.6733,
309
+ "step": 480
310
+ },
311
+ {
312
+ "epoch": 0.6,
313
+ "learning_rate": 8.187526161573881e-06,
314
+ "loss": 0.6546,
315
+ "step": 490
316
+ },
317
+ {
318
+ "epoch": 0.61,
319
+ "learning_rate": 8.145667643365426e-06,
320
+ "loss": 0.6234,
321
+ "step": 500
322
+ },
323
+ {
324
+ "epoch": 0.62,
325
+ "eval_oasst_export_w_label_accuracy": 0.7423076923076923,
326
+ "eval_oasst_export_w_label_kendalltau": 0.3931501831501834,
327
+ "eval_oasst_export_w_label_loss": 0.60888671875,
328
+ "eval_oasst_export_w_label_neg_score": 0.53515625,
329
+ "eval_oasst_export_w_label_pos_score": 1.1513671875,
330
+ "eval_oasst_export_w_label_runtime": 190.2556,
331
+ "eval_oasst_export_w_label_samples_per_second": 9.566,
332
+ "eval_oasst_export_w_label_score_diff": 0.6162109375,
333
+ "eval_oasst_export_w_label_steps_per_second": 2.392,
334
+ "step": 502
335
+ },
336
+ {
337
+ "epoch": 0.63,
338
+ "learning_rate": 8.10380912515697e-06,
339
+ "loss": 0.6087,
340
+ "step": 510
341
+ },
342
+ {
343
+ "epoch": 0.64,
344
+ "learning_rate": 8.061950606948514e-06,
345
+ "loss": 0.6345,
346
+ "step": 520
347
+ },
348
+ {
349
+ "epoch": 0.65,
350
+ "learning_rate": 8.020092088740058e-06,
351
+ "loss": 0.5881,
352
+ "step": 530
353
+ },
354
+ {
355
+ "epoch": 0.66,
356
+ "learning_rate": 7.978233570531604e-06,
357
+ "loss": 0.5954,
358
+ "step": 540
359
+ },
360
+ {
361
+ "epoch": 0.68,
362
+ "learning_rate": 7.936375052323149e-06,
363
+ "loss": 0.61,
364
+ "step": 550
365
+ },
366
+ {
367
+ "epoch": 0.69,
368
+ "learning_rate": 7.894516534114693e-06,
369
+ "loss": 0.5808,
370
+ "step": 560
371
+ },
372
+ {
373
+ "epoch": 0.7,
374
+ "learning_rate": 7.852658015906237e-06,
375
+ "loss": 0.6579,
376
+ "step": 570
377
+ },
378
+ {
379
+ "epoch": 0.71,
380
+ "learning_rate": 7.810799497697782e-06,
381
+ "loss": 0.6155,
382
+ "step": 580
383
+ },
384
+ {
385
+ "epoch": 0.73,
386
+ "learning_rate": 7.768940979489327e-06,
387
+ "loss": 0.6049,
388
+ "step": 590
389
+ },
390
+ {
391
+ "epoch": 0.74,
392
+ "learning_rate": 7.727082461280872e-06,
393
+ "loss": 0.6135,
394
+ "step": 600
395
+ },
396
+ {
397
+ "epoch": 0.75,
398
+ "learning_rate": 7.685223943072416e-06,
399
+ "loss": 0.5793,
400
+ "step": 610
401
+ },
402
+ {
403
+ "epoch": 0.76,
404
+ "learning_rate": 7.64336542486396e-06,
405
+ "loss": 0.6091,
406
+ "step": 620
407
+ },
408
+ {
409
+ "epoch": 0.77,
410
+ "learning_rate": 7.601506906655505e-06,
411
+ "loss": 0.6196,
412
+ "step": 630
413
+ },
414
+ {
415
+ "epoch": 0.79,
416
+ "learning_rate": 7.55964838844705e-06,
417
+ "loss": 0.5951,
418
+ "step": 640
419
+ },
420
+ {
421
+ "epoch": 0.8,
422
+ "learning_rate": 7.517789870238594e-06,
423
+ "loss": 0.6366,
424
+ "step": 650
425
+ },
426
+ {
427
+ "epoch": 0.81,
428
+ "learning_rate": 7.475931352030139e-06,
429
+ "loss": 0.6109,
430
+ "step": 660
431
+ },
432
+ {
433
+ "epoch": 0.82,
434
+ "learning_rate": 7.434072833821683e-06,
435
+ "loss": 0.5869,
436
+ "step": 670
437
+ },
438
+ {
439
+ "epoch": 0.84,
440
+ "learning_rate": 7.392214315613228e-06,
441
+ "loss": 0.612,
442
+ "step": 680
443
+ },
444
+ {
445
+ "epoch": 0.85,
446
+ "learning_rate": 7.350355797404772e-06,
447
+ "loss": 0.6019,
448
+ "step": 690
449
+ },
450
+ {
451
+ "epoch": 0.86,
452
+ "learning_rate": 7.308497279196317e-06,
453
+ "loss": 0.6265,
454
+ "step": 700
455
+ },
456
+ {
457
+ "epoch": 0.87,
458
+ "learning_rate": 7.2666387609878615e-06,
459
+ "loss": 0.5994,
460
+ "step": 710
461
+ },
462
+ {
463
+ "epoch": 0.89,
464
+ "learning_rate": 7.224780242779407e-06,
465
+ "loss": 0.5607,
466
+ "step": 720
467
+ },
468
+ {
469
+ "epoch": 0.9,
470
+ "learning_rate": 7.18292172457095e-06,
471
+ "loss": 0.5985,
472
+ "step": 730
473
+ },
474
+ {
475
+ "epoch": 0.91,
476
+ "learning_rate": 7.141063206362495e-06,
477
+ "loss": 0.5915,
478
+ "step": 740
479
+ },
480
+ {
481
+ "epoch": 0.92,
482
+ "learning_rate": 7.09920468815404e-06,
483
+ "loss": 0.5442,
484
+ "step": 750
485
+ },
486
+ {
487
+ "epoch": 0.93,
488
+ "eval_oasst_export_w_label_accuracy": 0.7478021978021978,
489
+ "eval_oasst_export_w_label_kendalltau": 0.4016483516483517,
490
+ "eval_oasst_export_w_label_loss": 0.62255859375,
491
+ "eval_oasst_export_w_label_neg_score": -0.1964111328125,
492
+ "eval_oasst_export_w_label_pos_score": 0.82470703125,
493
+ "eval_oasst_export_w_label_runtime": 190.3739,
494
+ "eval_oasst_export_w_label_samples_per_second": 9.56,
495
+ "eval_oasst_export_w_label_score_diff": 1.021484375,
496
+ "eval_oasst_export_w_label_steps_per_second": 2.39,
497
+ "step": 753
498
+ },
499
+ {
500
+ "epoch": 0.93,
501
+ "learning_rate": 7.057346169945585e-06,
502
+ "loss": 0.6559,
503
+ "step": 760
504
+ },
505
+ {
506
+ "epoch": 0.95,
507
+ "learning_rate": 7.01548765173713e-06,
508
+ "loss": 0.5909,
509
+ "step": 770
510
+ },
511
+ {
512
+ "epoch": 0.96,
513
+ "learning_rate": 6.973629133528673e-06,
514
+ "loss": 0.6263,
515
+ "step": 780
516
+ },
517
+ {
518
+ "epoch": 0.97,
519
+ "learning_rate": 6.931770615320218e-06,
520
+ "loss": 0.6288,
521
+ "step": 790
522
+ },
523
+ {
524
+ "epoch": 0.98,
525
+ "learning_rate": 6.889912097111763e-06,
526
+ "loss": 0.5627,
527
+ "step": 800
528
+ },
529
+ {
530
+ "epoch": 1.0,
531
+ "learning_rate": 6.848053578903308e-06,
532
+ "loss": 0.6076,
533
+ "step": 810
534
+ },
535
+ {
536
+ "epoch": 1.01,
537
+ "learning_rate": 6.806195060694852e-06,
538
+ "loss": 0.567,
539
+ "step": 820
540
+ },
541
+ {
542
+ "epoch": 1.02,
543
+ "learning_rate": 6.764336542486396e-06,
544
+ "loss": 0.565,
545
+ "step": 830
546
+ },
547
+ {
548
+ "epoch": 1.03,
549
+ "learning_rate": 6.722478024277941e-06,
550
+ "loss": 0.6091,
551
+ "step": 840
552
+ },
553
+ {
554
+ "epoch": 1.05,
555
+ "learning_rate": 6.680619506069486e-06,
556
+ "loss": 0.5579,
557
+ "step": 850
558
+ },
559
+ {
560
+ "epoch": 1.06,
561
+ "learning_rate": 6.63876098786103e-06,
562
+ "loss": 0.5943,
563
+ "step": 860
564
+ },
565
+ {
566
+ "epoch": 1.07,
567
+ "learning_rate": 6.596902469652575e-06,
568
+ "loss": 0.566,
569
+ "step": 870
570
+ },
571
+ {
572
+ "epoch": 1.08,
573
+ "learning_rate": 6.555043951444119e-06,
574
+ "loss": 0.5757,
575
+ "step": 880
576
+ },
577
+ {
578
+ "epoch": 1.09,
579
+ "learning_rate": 6.513185433235664e-06,
580
+ "loss": 0.5752,
581
+ "step": 890
582
+ },
583
+ {
584
+ "epoch": 1.11,
585
+ "learning_rate": 6.471326915027208e-06,
586
+ "loss": 0.5595,
587
+ "step": 900
588
+ },
589
+ {
590
+ "epoch": 1.12,
591
+ "learning_rate": 6.429468396818753e-06,
592
+ "loss": 0.6051,
593
+ "step": 910
594
+ },
595
+ {
596
+ "epoch": 1.13,
597
+ "learning_rate": 6.387609878610298e-06,
598
+ "loss": 0.5308,
599
+ "step": 920
600
+ },
601
+ {
602
+ "epoch": 1.14,
603
+ "learning_rate": 6.345751360401843e-06,
604
+ "loss": 0.5341,
605
+ "step": 930
606
+ },
607
+ {
608
+ "epoch": 1.16,
609
+ "learning_rate": 6.303892842193386e-06,
610
+ "loss": 0.5976,
611
+ "step": 940
612
+ },
613
+ {
614
+ "epoch": 1.17,
615
+ "learning_rate": 6.262034323984931e-06,
616
+ "loss": 0.568,
617
+ "step": 950
618
+ },
619
+ {
620
+ "epoch": 1.18,
621
+ "learning_rate": 6.220175805776476e-06,
622
+ "loss": 0.5865,
623
+ "step": 960
624
+ },
625
+ {
626
+ "epoch": 1.19,
627
+ "learning_rate": 6.178317287568021e-06,
628
+ "loss": 0.5486,
629
+ "step": 970
630
+ },
631
+ {
632
+ "epoch": 1.21,
633
+ "learning_rate": 6.136458769359566e-06,
634
+ "loss": 0.5699,
635
+ "step": 980
636
+ },
637
+ {
638
+ "epoch": 1.22,
639
+ "learning_rate": 6.094600251151109e-06,
640
+ "loss": 0.5831,
641
+ "step": 990
642
+ },
643
+ {
644
+ "epoch": 1.23,
645
+ "learning_rate": 6.052741732942654e-06,
646
+ "loss": 0.5558,
647
+ "step": 1000
648
+ },
649
+ {
650
+ "epoch": 1.23,
651
+ "eval_oasst_export_w_label_accuracy": 0.743956043956044,
652
+ "eval_oasst_export_w_label_kendalltau": 0.3847985347985353,
653
+ "eval_oasst_export_w_label_loss": 0.61865234375,
654
+ "eval_oasst_export_w_label_neg_score": -0.09515380859375,
655
+ "eval_oasst_export_w_label_pos_score": 0.93212890625,
656
+ "eval_oasst_export_w_label_runtime": 190.324,
657
+ "eval_oasst_export_w_label_samples_per_second": 9.563,
658
+ "eval_oasst_export_w_label_score_diff": 1.02734375,
659
+ "eval_oasst_export_w_label_steps_per_second": 2.391,
660
+ "step": 1004
661
+ },
662
+ {
663
+ "epoch": 1.24,
664
+ "learning_rate": 6.010883214734199e-06,
665
+ "loss": 0.5286,
666
+ "step": 1010
667
+ },
668
+ {
669
+ "epoch": 1.25,
670
+ "learning_rate": 5.969024696525744e-06,
671
+ "loss": 0.5657,
672
+ "step": 1020
673
+ },
674
+ {
675
+ "epoch": 1.27,
676
+ "learning_rate": 5.927166178317288e-06,
677
+ "loss": 0.5376,
678
+ "step": 1030
679
+ },
680
+ {
681
+ "epoch": 1.28,
682
+ "learning_rate": 5.885307660108832e-06,
683
+ "loss": 0.5375,
684
+ "step": 1040
685
+ },
686
+ {
687
+ "epoch": 1.29,
688
+ "learning_rate": 5.843449141900377e-06,
689
+ "loss": 0.5513,
690
+ "step": 1050
691
+ },
692
+ {
693
+ "epoch": 1.3,
694
+ "learning_rate": 5.801590623691922e-06,
695
+ "loss": 0.5778,
696
+ "step": 1060
697
+ },
698
+ {
699
+ "epoch": 1.32,
700
+ "learning_rate": 5.759732105483467e-06,
701
+ "loss": 0.5445,
702
+ "step": 1070
703
+ },
704
+ {
705
+ "epoch": 1.33,
706
+ "learning_rate": 5.717873587275011e-06,
707
+ "loss": 0.5657,
708
+ "step": 1080
709
+ },
710
+ {
711
+ "epoch": 1.34,
712
+ "learning_rate": 5.676015069066555e-06,
713
+ "loss": 0.5808,
714
+ "step": 1090
715
+ },
716
+ {
717
+ "epoch": 1.35,
718
+ "learning_rate": 5.6341565508581e-06,
719
+ "loss": 0.5723,
720
+ "step": 1100
721
+ },
722
+ {
723
+ "epoch": 1.37,
724
+ "learning_rate": 5.592298032649645e-06,
725
+ "loss": 0.5413,
726
+ "step": 1110
727
+ },
728
+ {
729
+ "epoch": 1.38,
730
+ "learning_rate": 5.550439514441189e-06,
731
+ "loss": 0.6384,
732
+ "step": 1120
733
+ },
734
+ {
735
+ "epoch": 1.39,
736
+ "learning_rate": 5.508580996232734e-06,
737
+ "loss": 0.5336,
738
+ "step": 1130
739
+ },
740
+ {
741
+ "epoch": 1.4,
742
+ "learning_rate": 5.466722478024279e-06,
743
+ "loss": 0.5754,
744
+ "step": 1140
745
+ },
746
+ {
747
+ "epoch": 1.41,
748
+ "learning_rate": 5.424863959815823e-06,
749
+ "loss": 0.5171,
750
+ "step": 1150
751
+ },
752
+ {
753
+ "epoch": 1.43,
754
+ "learning_rate": 5.383005441607367e-06,
755
+ "loss": 0.5536,
756
+ "step": 1160
757
+ },
758
+ {
759
+ "epoch": 1.44,
760
+ "learning_rate": 5.341146923398912e-06,
761
+ "loss": 0.5812,
762
+ "step": 1170
763
+ },
764
+ {
765
+ "epoch": 1.45,
766
+ "learning_rate": 5.299288405190457e-06,
767
+ "loss": 0.5604,
768
+ "step": 1180
769
+ },
770
+ {
771
+ "epoch": 1.46,
772
+ "learning_rate": 5.257429886982002e-06,
773
+ "loss": 0.5433,
774
+ "step": 1190
775
+ },
776
+ {
777
+ "epoch": 1.48,
778
+ "learning_rate": 5.215571368773545e-06,
779
+ "loss": 0.5704,
780
+ "step": 1200
781
+ },
782
+ {
783
+ "epoch": 1.49,
784
+ "learning_rate": 5.17371285056509e-06,
785
+ "loss": 0.592,
786
+ "step": 1210
787
+ },
788
+ {
789
+ "epoch": 1.5,
790
+ "learning_rate": 5.131854332356635e-06,
791
+ "loss": 0.5406,
792
+ "step": 1220
793
+ },
794
+ {
795
+ "epoch": 1.51,
796
+ "learning_rate": 5.08999581414818e-06,
797
+ "loss": 0.5597,
798
+ "step": 1230
799
+ },
800
+ {
801
+ "epoch": 1.53,
802
+ "learning_rate": 5.048137295939725e-06,
803
+ "loss": 0.6011,
804
+ "step": 1240
805
+ },
806
+ {
807
+ "epoch": 1.54,
808
+ "learning_rate": 5.006278777731268e-06,
809
+ "loss": 0.5071,
810
+ "step": 1250
811
+ },
812
+ {
813
+ "epoch": 1.54,
814
+ "eval_oasst_export_w_label_accuracy": 0.7478021978021978,
815
+ "eval_oasst_export_w_label_kendalltau": 0.4045054945054951,
816
+ "eval_oasst_export_w_label_loss": 0.61669921875,
817
+ "eval_oasst_export_w_label_neg_score": -0.1842041015625,
818
+ "eval_oasst_export_w_label_pos_score": 1.02734375,
819
+ "eval_oasst_export_w_label_runtime": 190.3207,
820
+ "eval_oasst_export_w_label_samples_per_second": 9.563,
821
+ "eval_oasst_export_w_label_score_diff": 1.2109375,
822
+ "eval_oasst_export_w_label_steps_per_second": 2.391,
823
+ "step": 1255
824
+ },
825
+ {
826
+ "epoch": 1.55,
827
+ "learning_rate": 4.964420259522813e-06,
828
+ "loss": 0.5374,
829
+ "step": 1260
830
+ },
831
+ {
832
+ "epoch": 1.56,
833
+ "learning_rate": 4.922561741314358e-06,
834
+ "loss": 0.5613,
835
+ "step": 1270
836
+ },
837
+ {
838
+ "epoch": 1.57,
839
+ "learning_rate": 4.880703223105903e-06,
840
+ "loss": 0.5643,
841
+ "step": 1280
842
+ },
843
+ {
844
+ "epoch": 1.59,
845
+ "learning_rate": 4.838844704897447e-06,
846
+ "loss": 0.5755,
847
+ "step": 1290
848
+ },
849
+ {
850
+ "epoch": 1.6,
851
+ "learning_rate": 4.7969861866889915e-06,
852
+ "loss": 0.5631,
853
+ "step": 1300
854
+ },
855
+ {
856
+ "epoch": 1.61,
857
+ "learning_rate": 4.755127668480537e-06,
858
+ "loss": 0.5557,
859
+ "step": 1310
860
+ },
861
+ {
862
+ "epoch": 1.62,
863
+ "learning_rate": 4.713269150272081e-06,
864
+ "loss": 0.5173,
865
+ "step": 1320
866
+ },
867
+ {
868
+ "epoch": 1.64,
869
+ "learning_rate": 4.671410632063625e-06,
870
+ "loss": 0.5938,
871
+ "step": 1330
872
+ },
873
+ {
874
+ "epoch": 1.65,
875
+ "learning_rate": 4.6295521138551695e-06,
876
+ "loss": 0.6386,
877
+ "step": 1340
878
+ },
879
+ {
880
+ "epoch": 1.66,
881
+ "learning_rate": 4.587693595646715e-06,
882
+ "loss": 0.5652,
883
+ "step": 1350
884
+ },
885
+ {
886
+ "epoch": 1.67,
887
+ "learning_rate": 4.545835077438259e-06,
888
+ "loss": 0.6097,
889
+ "step": 1360
890
+ },
891
+ {
892
+ "epoch": 1.69,
893
+ "learning_rate": 4.503976559229803e-06,
894
+ "loss": 0.5369,
895
+ "step": 1370
896
+ },
897
+ {
898
+ "epoch": 1.7,
899
+ "learning_rate": 4.462118041021348e-06,
900
+ "loss": 0.539,
901
+ "step": 1380
902
+ },
903
+ {
904
+ "epoch": 1.71,
905
+ "learning_rate": 4.4202595228128935e-06,
906
+ "loss": 0.5854,
907
+ "step": 1390
908
+ },
909
+ {
910
+ "epoch": 1.72,
911
+ "learning_rate": 4.378401004604437e-06,
912
+ "loss": 0.5574,
913
+ "step": 1400
914
+ },
915
+ {
916
+ "epoch": 1.73,
917
+ "learning_rate": 4.336542486395982e-06,
918
+ "loss": 0.5079,
919
+ "step": 1410
920
+ },
921
+ {
922
+ "epoch": 1.75,
923
+ "learning_rate": 4.294683968187526e-06,
924
+ "loss": 0.5889,
925
+ "step": 1420
926
+ },
927
+ {
928
+ "epoch": 1.76,
929
+ "learning_rate": 4.2528254499790715e-06,
930
+ "loss": 0.5908,
931
+ "step": 1430
932
+ },
933
+ {
934
+ "epoch": 1.77,
935
+ "learning_rate": 4.210966931770616e-06,
936
+ "loss": 0.5254,
937
+ "step": 1440
938
+ },
939
+ {
940
+ "epoch": 1.78,
941
+ "learning_rate": 4.16910841356216e-06,
942
+ "loss": 0.5566,
943
+ "step": 1450
944
+ },
945
+ {
946
+ "epoch": 1.8,
947
+ "learning_rate": 4.127249895353705e-06,
948
+ "loss": 0.5743,
949
+ "step": 1460
950
+ },
951
+ {
952
+ "epoch": 1.81,
953
+ "learning_rate": 4.0853913771452495e-06,
954
+ "loss": 0.5208,
955
+ "step": 1470
956
+ },
957
+ {
958
+ "epoch": 1.82,
959
+ "learning_rate": 4.043532858936794e-06,
960
+ "loss": 0.5486,
961
+ "step": 1480
962
+ },
963
+ {
964
+ "epoch": 1.83,
965
+ "learning_rate": 4.001674340728339e-06,
966
+ "loss": 0.587,
967
+ "step": 1490
968
+ },
969
+ {
970
+ "epoch": 1.84,
971
+ "learning_rate": 3.959815822519883e-06,
972
+ "loss": 0.5781,
973
+ "step": 1500
974
+ }
975
+ ],
976
+ "max_steps": 2439,
977
+ "num_train_epochs": 3,
978
+ "total_flos": 0.0,
979
+ "trial_name": null,
980
+ "trial_params": null
981
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f5700a4afb75b927bce7ba70711d04ded78b4b39072e14a647f2798bcd9ea3f4
3
+ size 4667