theblackcat102 commited on
Commit
fbccd8e
1 Parent(s): 3e3d49b

[init] 1500 iterations

Browse files
added_tokens.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "<|assistant|>": 128003,
3
+ "<|prefix_begin|>": 128005,
4
+ "<|prefix_end|>": 128001,
5
+ "<|prompter|>": 128004,
6
+ "<|system|>": 128002,
7
+ "[MASK]": 128000
8
+ }
config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/deberta-v3-base",
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": 768,
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": 3072,
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": 12,
40
+ "num_hidden_layers": 12,
41
+ "pad_token_id": 0,
42
+ "pooler_dropout": 0,
43
+ "pooler_hidden_act": "gelu",
44
+ "pooler_hidden_size": 768,
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:fc77158664d449bb2ffd3f4d9cbc45e1c1e4a4b2eb03cdf45d22d06db05e9580
3
+ size 368891647
special_tokens_map.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|prefix_end|>",
4
+ "<|system|>",
5
+ "<|assistant|>",
6
+ "<|prompter|>",
7
+ "<|prefix_begin|>"
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": 1.8444512757454659,
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": 1.765775730203337e-05,
13
+ "loss": 0.878,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.02,
18
+ "learning_rate": 2.257992073472628e-05,
19
+ "loss": 0.8493,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.04,
24
+ "learning_rate": 2.5553565295129308e-05,
25
+ "loss": 0.8135,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.05,
30
+ "learning_rate": 2.7895435338793016e-05,
31
+ "loss": 0.7467,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.06,
36
+ "learning_rate": 2.9686949735748396e-05,
37
+ "loss": 0.6927,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.07,
42
+ "learning_rate": 2.992465466722478e-05,
43
+ "loss": 0.6988,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.09,
48
+ "learning_rate": 2.9811636668061953e-05,
49
+ "loss": 0.6967,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.1,
54
+ "learning_rate": 2.9686061113436587e-05,
55
+ "loss": 0.6456,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.11,
60
+ "learning_rate": 2.956048555881122e-05,
61
+ "loss": 0.6476,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.12,
66
+ "learning_rate": 2.944746755964839e-05,
67
+ "loss": 0.6664,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.14,
72
+ "learning_rate": 2.9321892005023023e-05,
73
+ "loss": 0.6349,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.15,
78
+ "learning_rate": 2.9196316450397654e-05,
79
+ "loss": 0.6442,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.16,
84
+ "learning_rate": 2.908329845123483e-05,
85
+ "loss": 0.6758,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.17,
90
+ "learning_rate": 2.8957722896609463e-05,
91
+ "loss": 0.6686,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.18,
96
+ "learning_rate": 2.8832147341984094e-05,
97
+ "loss": 0.6853,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.2,
102
+ "learning_rate": 2.870657178735873e-05,
103
+ "loss": 0.6428,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.21,
108
+ "learning_rate": 2.8580996232733363e-05,
109
+ "loss": 0.6093,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.22,
114
+ "learning_rate": 2.8455420678107998e-05,
115
+ "loss": 0.6382,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.23,
120
+ "learning_rate": 2.832984512348263e-05,
121
+ "loss": 0.6236,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.25,
126
+ "learning_rate": 2.820426956885726e-05,
127
+ "loss": 0.6172,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 0.26,
132
+ "learning_rate": 2.8078694014231895e-05,
133
+ "loss": 0.6481,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 0.27,
138
+ "learning_rate": 2.795311845960653e-05,
139
+ "loss": 0.6091,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 0.28,
144
+ "learning_rate": 2.7827542904981164e-05,
145
+ "loss": 0.6343,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 0.3,
150
+ "learning_rate": 2.7701967350355798e-05,
151
+ "loss": 0.6248,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 0.31,
156
+ "learning_rate": 2.7576391795730433e-05,
157
+ "loss": 0.6516,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 0.31,
162
+ "eval_oasst_export_w_label_accuracy": 0.7197802197802198,
163
+ "eval_oasst_export_w_label_kendalltau": 0.35809523809523786,
164
+ "eval_oasst_export_w_label_loss": 0.64404296875,
165
+ "eval_oasst_export_w_label_neg_score": -0.31201171875,
166
+ "eval_oasst_export_w_label_pos_score": 0.261962890625,
167
+ "eval_oasst_export_w_label_runtime": 69.9812,
168
+ "eval_oasst_export_w_label_samples_per_second": 26.007,
169
+ "eval_oasst_export_w_label_score_diff": 0.57421875,
170
+ "eval_oasst_export_w_label_steps_per_second": 6.502,
171
+ "step": 251
172
+ },
173
+ {
174
+ "epoch": 0.32,
175
+ "learning_rate": 2.7450816241105064e-05,
176
+ "loss": 0.6195,
177
+ "step": 260
178
+ },
179
+ {
180
+ "epoch": 0.33,
181
+ "learning_rate": 2.73252406864797e-05,
182
+ "loss": 0.5911,
183
+ "step": 270
184
+ },
185
+ {
186
+ "epoch": 0.34,
187
+ "learning_rate": 2.7199665131854333e-05,
188
+ "loss": 0.621,
189
+ "step": 280
190
+ },
191
+ {
192
+ "epoch": 0.36,
193
+ "learning_rate": 2.7074089577228968e-05,
194
+ "loss": 0.6102,
195
+ "step": 290
196
+ },
197
+ {
198
+ "epoch": 0.37,
199
+ "learning_rate": 2.6948514022603602e-05,
200
+ "loss": 0.5947,
201
+ "step": 300
202
+ },
203
+ {
204
+ "epoch": 0.38,
205
+ "learning_rate": 2.6822938467978233e-05,
206
+ "loss": 0.6253,
207
+ "step": 310
208
+ },
209
+ {
210
+ "epoch": 0.39,
211
+ "learning_rate": 2.6697362913352868e-05,
212
+ "loss": 0.57,
213
+ "step": 320
214
+ },
215
+ {
216
+ "epoch": 0.41,
217
+ "learning_rate": 2.6571787358727503e-05,
218
+ "loss": 0.6439,
219
+ "step": 330
220
+ },
221
+ {
222
+ "epoch": 0.42,
223
+ "learning_rate": 2.6446211804102137e-05,
224
+ "loss": 0.611,
225
+ "step": 340
226
+ },
227
+ {
228
+ "epoch": 0.43,
229
+ "learning_rate": 2.632063624947677e-05,
230
+ "loss": 0.6451,
231
+ "step": 350
232
+ },
233
+ {
234
+ "epoch": 0.44,
235
+ "learning_rate": 2.6195060694851403e-05,
236
+ "loss": 0.6299,
237
+ "step": 360
238
+ },
239
+ {
240
+ "epoch": 0.45,
241
+ "learning_rate": 2.6069485140226037e-05,
242
+ "loss": 0.6554,
243
+ "step": 370
244
+ },
245
+ {
246
+ "epoch": 0.47,
247
+ "learning_rate": 2.594390958560067e-05,
248
+ "loss": 0.6024,
249
+ "step": 380
250
+ },
251
+ {
252
+ "epoch": 0.48,
253
+ "learning_rate": 2.5818334030975303e-05,
254
+ "loss": 0.6215,
255
+ "step": 390
256
+ },
257
+ {
258
+ "epoch": 0.49,
259
+ "learning_rate": 2.5705316031812474e-05,
260
+ "loss": 0.594,
261
+ "step": 400
262
+ },
263
+ {
264
+ "epoch": 0.5,
265
+ "learning_rate": 2.5579740477187108e-05,
266
+ "loss": 0.5889,
267
+ "step": 410
268
+ },
269
+ {
270
+ "epoch": 0.52,
271
+ "learning_rate": 2.5454164922561743e-05,
272
+ "loss": 0.6088,
273
+ "step": 420
274
+ },
275
+ {
276
+ "epoch": 0.53,
277
+ "learning_rate": 2.5328589367936377e-05,
278
+ "loss": 0.6793,
279
+ "step": 430
280
+ },
281
+ {
282
+ "epoch": 0.54,
283
+ "learning_rate": 2.5203013813311012e-05,
284
+ "loss": 0.6468,
285
+ "step": 440
286
+ },
287
+ {
288
+ "epoch": 0.55,
289
+ "learning_rate": 2.5077438258685643e-05,
290
+ "loss": 0.6457,
291
+ "step": 450
292
+ },
293
+ {
294
+ "epoch": 0.57,
295
+ "learning_rate": 2.4951862704060278e-05,
296
+ "loss": 0.6246,
297
+ "step": 460
298
+ },
299
+ {
300
+ "epoch": 0.58,
301
+ "learning_rate": 2.4826287149434912e-05,
302
+ "loss": 0.5629,
303
+ "step": 470
304
+ },
305
+ {
306
+ "epoch": 0.59,
307
+ "learning_rate": 2.4700711594809547e-05,
308
+ "loss": 0.6508,
309
+ "step": 480
310
+ },
311
+ {
312
+ "epoch": 0.6,
313
+ "learning_rate": 2.457513604018418e-05,
314
+ "loss": 0.635,
315
+ "step": 490
316
+ },
317
+ {
318
+ "epoch": 0.61,
319
+ "learning_rate": 2.444956048555881e-05,
320
+ "loss": 0.6358,
321
+ "step": 500
322
+ },
323
+ {
324
+ "epoch": 0.62,
325
+ "eval_oasst_export_w_label_accuracy": 0.7346153846153847,
326
+ "eval_oasst_export_w_label_kendalltau": 0.3878754578754578,
327
+ "eval_oasst_export_w_label_loss": 0.6220703125,
328
+ "eval_oasst_export_w_label_neg_score": 0.30078125,
329
+ "eval_oasst_export_w_label_pos_score": 0.83984375,
330
+ "eval_oasst_export_w_label_runtime": 69.9591,
331
+ "eval_oasst_export_w_label_samples_per_second": 26.015,
332
+ "eval_oasst_export_w_label_score_diff": 0.5390625,
333
+ "eval_oasst_export_w_label_steps_per_second": 6.504,
334
+ "step": 502
335
+ },
336
+ {
337
+ "epoch": 0.63,
338
+ "learning_rate": 2.4323984930933444e-05,
339
+ "loss": 0.6526,
340
+ "step": 510
341
+ },
342
+ {
343
+ "epoch": 0.64,
344
+ "learning_rate": 2.4198409376308078e-05,
345
+ "loss": 0.6391,
346
+ "step": 520
347
+ },
348
+ {
349
+ "epoch": 0.65,
350
+ "learning_rate": 2.4072833821682713e-05,
351
+ "loss": 0.6083,
352
+ "step": 530
353
+ },
354
+ {
355
+ "epoch": 0.66,
356
+ "learning_rate": 2.3947258267057347e-05,
357
+ "loss": 0.6283,
358
+ "step": 540
359
+ },
360
+ {
361
+ "epoch": 0.68,
362
+ "learning_rate": 2.382168271243198e-05,
363
+ "loss": 0.6115,
364
+ "step": 550
365
+ },
366
+ {
367
+ "epoch": 0.69,
368
+ "learning_rate": 2.3696107157806613e-05,
369
+ "loss": 0.6069,
370
+ "step": 560
371
+ },
372
+ {
373
+ "epoch": 0.7,
374
+ "learning_rate": 2.3570531603181248e-05,
375
+ "loss": 0.6462,
376
+ "step": 570
377
+ },
378
+ {
379
+ "epoch": 0.71,
380
+ "learning_rate": 2.3444956048555882e-05,
381
+ "loss": 0.6353,
382
+ "step": 580
383
+ },
384
+ {
385
+ "epoch": 0.73,
386
+ "learning_rate": 2.3319380493930517e-05,
387
+ "loss": 0.6194,
388
+ "step": 590
389
+ },
390
+ {
391
+ "epoch": 0.74,
392
+ "learning_rate": 2.3193804939305148e-05,
393
+ "loss": 0.618,
394
+ "step": 600
395
+ },
396
+ {
397
+ "epoch": 0.75,
398
+ "learning_rate": 2.3068229384679782e-05,
399
+ "loss": 0.5972,
400
+ "step": 610
401
+ },
402
+ {
403
+ "epoch": 0.76,
404
+ "learning_rate": 2.2942653830054417e-05,
405
+ "loss": 0.6001,
406
+ "step": 620
407
+ },
408
+ {
409
+ "epoch": 0.77,
410
+ "learning_rate": 2.281707827542905e-05,
411
+ "loss": 0.6341,
412
+ "step": 630
413
+ },
414
+ {
415
+ "epoch": 0.79,
416
+ "learning_rate": 2.2691502720803686e-05,
417
+ "loss": 0.6168,
418
+ "step": 640
419
+ },
420
+ {
421
+ "epoch": 0.8,
422
+ "learning_rate": 2.256592716617832e-05,
423
+ "loss": 0.6507,
424
+ "step": 650
425
+ },
426
+ {
427
+ "epoch": 0.81,
428
+ "learning_rate": 2.2440351611552952e-05,
429
+ "loss": 0.6237,
430
+ "step": 660
431
+ },
432
+ {
433
+ "epoch": 0.82,
434
+ "learning_rate": 2.2314776056927586e-05,
435
+ "loss": 0.6004,
436
+ "step": 670
437
+ },
438
+ {
439
+ "epoch": 0.84,
440
+ "learning_rate": 2.218920050230222e-05,
441
+ "loss": 0.6198,
442
+ "step": 680
443
+ },
444
+ {
445
+ "epoch": 0.85,
446
+ "learning_rate": 2.2063624947676852e-05,
447
+ "loss": 0.6166,
448
+ "step": 690
449
+ },
450
+ {
451
+ "epoch": 0.86,
452
+ "learning_rate": 2.1938049393051487e-05,
453
+ "loss": 0.6492,
454
+ "step": 700
455
+ },
456
+ {
457
+ "epoch": 0.87,
458
+ "learning_rate": 2.1812473838426118e-05,
459
+ "loss": 0.6194,
460
+ "step": 710
461
+ },
462
+ {
463
+ "epoch": 0.89,
464
+ "learning_rate": 2.1686898283800752e-05,
465
+ "loss": 0.5555,
466
+ "step": 720
467
+ },
468
+ {
469
+ "epoch": 0.9,
470
+ "learning_rate": 2.1561322729175387e-05,
471
+ "loss": 0.6118,
472
+ "step": 730
473
+ },
474
+ {
475
+ "epoch": 0.91,
476
+ "learning_rate": 2.143574717455002e-05,
477
+ "loss": 0.6099,
478
+ "step": 740
479
+ },
480
+ {
481
+ "epoch": 0.92,
482
+ "learning_rate": 2.1310171619924656e-05,
483
+ "loss": 0.5656,
484
+ "step": 750
485
+ },
486
+ {
487
+ "epoch": 0.93,
488
+ "eval_oasst_export_w_label_accuracy": 0.7307692307692307,
489
+ "eval_oasst_export_w_label_kendalltau": 0.3831868131868137,
490
+ "eval_oasst_export_w_label_loss": 0.63671875,
491
+ "eval_oasst_export_w_label_neg_score": -0.3466796875,
492
+ "eval_oasst_export_w_label_pos_score": 0.58740234375,
493
+ "eval_oasst_export_w_label_runtime": 70.0188,
494
+ "eval_oasst_export_w_label_samples_per_second": 25.993,
495
+ "eval_oasst_export_w_label_score_diff": 0.93408203125,
496
+ "eval_oasst_export_w_label_steps_per_second": 6.498,
497
+ "step": 753
498
+ },
499
+ {
500
+ "epoch": 0.93,
501
+ "learning_rate": 2.1184596065299287e-05,
502
+ "loss": 0.6576,
503
+ "step": 760
504
+ },
505
+ {
506
+ "epoch": 0.95,
507
+ "learning_rate": 2.1059020510673922e-05,
508
+ "loss": 0.59,
509
+ "step": 770
510
+ },
511
+ {
512
+ "epoch": 0.96,
513
+ "learning_rate": 2.0933444956048556e-05,
514
+ "loss": 0.6391,
515
+ "step": 780
516
+ },
517
+ {
518
+ "epoch": 0.97,
519
+ "learning_rate": 2.080786940142319e-05,
520
+ "loss": 0.6061,
521
+ "step": 790
522
+ },
523
+ {
524
+ "epoch": 0.98,
525
+ "learning_rate": 2.0682293846797825e-05,
526
+ "loss": 0.5858,
527
+ "step": 800
528
+ },
529
+ {
530
+ "epoch": 1.0,
531
+ "learning_rate": 2.0556718292172457e-05,
532
+ "loss": 0.6246,
533
+ "step": 810
534
+ },
535
+ {
536
+ "epoch": 1.01,
537
+ "learning_rate": 2.043114273754709e-05,
538
+ "loss": 0.5878,
539
+ "step": 820
540
+ },
541
+ {
542
+ "epoch": 1.02,
543
+ "learning_rate": 2.0305567182921726e-05,
544
+ "loss": 0.578,
545
+ "step": 830
546
+ },
547
+ {
548
+ "epoch": 1.03,
549
+ "learning_rate": 2.017999162829636e-05,
550
+ "loss": 0.6247,
551
+ "step": 840
552
+ },
553
+ {
554
+ "epoch": 1.05,
555
+ "learning_rate": 2.0054416073670995e-05,
556
+ "loss": 0.5841,
557
+ "step": 850
558
+ },
559
+ {
560
+ "epoch": 1.06,
561
+ "learning_rate": 1.9928840519045626e-05,
562
+ "loss": 0.6293,
563
+ "step": 860
564
+ },
565
+ {
566
+ "epoch": 1.07,
567
+ "learning_rate": 1.980326496442026e-05,
568
+ "loss": 0.5858,
569
+ "step": 870
570
+ },
571
+ {
572
+ "epoch": 1.08,
573
+ "learning_rate": 1.9677689409794892e-05,
574
+ "loss": 0.597,
575
+ "step": 880
576
+ },
577
+ {
578
+ "epoch": 1.09,
579
+ "learning_rate": 1.9552113855169526e-05,
580
+ "loss": 0.5869,
581
+ "step": 890
582
+ },
583
+ {
584
+ "epoch": 1.11,
585
+ "learning_rate": 1.942653830054416e-05,
586
+ "loss": 0.5672,
587
+ "step": 900
588
+ },
589
+ {
590
+ "epoch": 1.12,
591
+ "learning_rate": 1.9300962745918795e-05,
592
+ "loss": 0.6281,
593
+ "step": 910
594
+ },
595
+ {
596
+ "epoch": 1.13,
597
+ "learning_rate": 1.9175387191293427e-05,
598
+ "loss": 0.5356,
599
+ "step": 920
600
+ },
601
+ {
602
+ "epoch": 1.14,
603
+ "learning_rate": 1.904981163666806e-05,
604
+ "loss": 0.5277,
605
+ "step": 930
606
+ },
607
+ {
608
+ "epoch": 1.16,
609
+ "learning_rate": 1.8924236082042696e-05,
610
+ "loss": 0.5968,
611
+ "step": 940
612
+ },
613
+ {
614
+ "epoch": 1.17,
615
+ "learning_rate": 1.879866052741733e-05,
616
+ "loss": 0.6129,
617
+ "step": 950
618
+ },
619
+ {
620
+ "epoch": 1.18,
621
+ "learning_rate": 1.8673084972791965e-05,
622
+ "loss": 0.5914,
623
+ "step": 960
624
+ },
625
+ {
626
+ "epoch": 1.19,
627
+ "learning_rate": 1.8547509418166596e-05,
628
+ "loss": 0.5648,
629
+ "step": 970
630
+ },
631
+ {
632
+ "epoch": 1.21,
633
+ "learning_rate": 1.842193386354123e-05,
634
+ "loss": 0.5783,
635
+ "step": 980
636
+ },
637
+ {
638
+ "epoch": 1.22,
639
+ "learning_rate": 1.8296358308915865e-05,
640
+ "loss": 0.5895,
641
+ "step": 990
642
+ },
643
+ {
644
+ "epoch": 1.23,
645
+ "learning_rate": 1.81707827542905e-05,
646
+ "loss": 0.589,
647
+ "step": 1000
648
+ },
649
+ {
650
+ "epoch": 1.23,
651
+ "eval_oasst_export_w_label_accuracy": 0.7329670329670329,
652
+ "eval_oasst_export_w_label_kendalltau": 0.36846153846153845,
653
+ "eval_oasst_export_w_label_loss": 0.63330078125,
654
+ "eval_oasst_export_w_label_neg_score": -0.88427734375,
655
+ "eval_oasst_export_w_label_pos_score": 0.0426025390625,
656
+ "eval_oasst_export_w_label_runtime": 70.038,
657
+ "eval_oasst_export_w_label_samples_per_second": 25.986,
658
+ "eval_oasst_export_w_label_score_diff": 0.9267578125,
659
+ "eval_oasst_export_w_label_steps_per_second": 6.496,
660
+ "step": 1004
661
+ },
662
+ {
663
+ "epoch": 1.24,
664
+ "learning_rate": 1.8045207199665134e-05,
665
+ "loss": 0.5239,
666
+ "step": 1010
667
+ },
668
+ {
669
+ "epoch": 1.25,
670
+ "learning_rate": 1.7919631645039765e-05,
671
+ "loss": 0.5674,
672
+ "step": 1020
673
+ },
674
+ {
675
+ "epoch": 1.27,
676
+ "learning_rate": 1.77940560904144e-05,
677
+ "loss": 0.5643,
678
+ "step": 1030
679
+ },
680
+ {
681
+ "epoch": 1.28,
682
+ "learning_rate": 1.7668480535789035e-05,
683
+ "loss": 0.5631,
684
+ "step": 1040
685
+ },
686
+ {
687
+ "epoch": 1.29,
688
+ "learning_rate": 1.754290498116367e-05,
689
+ "loss": 0.5808,
690
+ "step": 1050
691
+ },
692
+ {
693
+ "epoch": 1.3,
694
+ "learning_rate": 1.7417329426538304e-05,
695
+ "loss": 0.5936,
696
+ "step": 1060
697
+ },
698
+ {
699
+ "epoch": 1.32,
700
+ "learning_rate": 1.729175387191293e-05,
701
+ "loss": 0.597,
702
+ "step": 1070
703
+ },
704
+ {
705
+ "epoch": 1.33,
706
+ "learning_rate": 1.7166178317287566e-05,
707
+ "loss": 0.5744,
708
+ "step": 1080
709
+ },
710
+ {
711
+ "epoch": 1.34,
712
+ "learning_rate": 1.70406027626622e-05,
713
+ "loss": 0.6338,
714
+ "step": 1090
715
+ },
716
+ {
717
+ "epoch": 1.35,
718
+ "learning_rate": 1.6915027208036835e-05,
719
+ "loss": 0.5946,
720
+ "step": 1100
721
+ },
722
+ {
723
+ "epoch": 1.36,
724
+ "learning_rate": 1.678945165341147e-05,
725
+ "loss": 0.5915,
726
+ "step": 1110
727
+ },
728
+ {
729
+ "epoch": 1.38,
730
+ "learning_rate": 1.6663876098786104e-05,
731
+ "loss": 0.6319,
732
+ "step": 1120
733
+ },
734
+ {
735
+ "epoch": 1.39,
736
+ "learning_rate": 1.6538300544160735e-05,
737
+ "loss": 0.5423,
738
+ "step": 1130
739
+ },
740
+ {
741
+ "epoch": 1.4,
742
+ "learning_rate": 1.641272498953537e-05,
743
+ "loss": 0.5818,
744
+ "step": 1140
745
+ },
746
+ {
747
+ "epoch": 1.41,
748
+ "learning_rate": 1.6287149434910005e-05,
749
+ "loss": 0.5576,
750
+ "step": 1150
751
+ },
752
+ {
753
+ "epoch": 1.43,
754
+ "learning_rate": 1.616157388028464e-05,
755
+ "loss": 0.5835,
756
+ "step": 1160
757
+ },
758
+ {
759
+ "epoch": 1.44,
760
+ "learning_rate": 1.6035998325659274e-05,
761
+ "loss": 0.6021,
762
+ "step": 1170
763
+ },
764
+ {
765
+ "epoch": 1.45,
766
+ "learning_rate": 1.5910422771033905e-05,
767
+ "loss": 0.6185,
768
+ "step": 1180
769
+ },
770
+ {
771
+ "epoch": 1.46,
772
+ "learning_rate": 1.578484721640854e-05,
773
+ "loss": 0.5637,
774
+ "step": 1190
775
+ },
776
+ {
777
+ "epoch": 1.48,
778
+ "learning_rate": 1.5659271661783174e-05,
779
+ "loss": 0.5624,
780
+ "step": 1200
781
+ },
782
+ {
783
+ "epoch": 1.49,
784
+ "learning_rate": 1.553369610715781e-05,
785
+ "loss": 0.5981,
786
+ "step": 1210
787
+ },
788
+ {
789
+ "epoch": 1.5,
790
+ "learning_rate": 1.5408120552532443e-05,
791
+ "loss": 0.5924,
792
+ "step": 1220
793
+ },
794
+ {
795
+ "epoch": 1.51,
796
+ "learning_rate": 1.5282544997907074e-05,
797
+ "loss": 0.5456,
798
+ "step": 1230
799
+ },
800
+ {
801
+ "epoch": 1.52,
802
+ "learning_rate": 1.5156969443281707e-05,
803
+ "loss": 0.6087,
804
+ "step": 1240
805
+ },
806
+ {
807
+ "epoch": 1.54,
808
+ "learning_rate": 1.5031393888656342e-05,
809
+ "loss": 0.5329,
810
+ "step": 1250
811
+ },
812
+ {
813
+ "epoch": 1.54,
814
+ "eval_oasst_export_w_label_accuracy": 0.7263736263736263,
815
+ "eval_oasst_export_w_label_kendalltau": 0.37285714285714255,
816
+ "eval_oasst_export_w_label_loss": 0.6298828125,
817
+ "eval_oasst_export_w_label_neg_score": -0.1357421875,
818
+ "eval_oasst_export_w_label_pos_score": 0.84716796875,
819
+ "eval_oasst_export_w_label_runtime": 70.012,
820
+ "eval_oasst_export_w_label_samples_per_second": 25.996,
821
+ "eval_oasst_export_w_label_score_diff": 0.98291015625,
822
+ "eval_oasst_export_w_label_steps_per_second": 6.499,
823
+ "step": 1255
824
+ },
825
+ {
826
+ "epoch": 1.55,
827
+ "learning_rate": 1.4905818334030976e-05,
828
+ "loss": 0.5531,
829
+ "step": 1260
830
+ },
831
+ {
832
+ "epoch": 1.56,
833
+ "learning_rate": 1.478024277940561e-05,
834
+ "loss": 0.5672,
835
+ "step": 1270
836
+ },
837
+ {
838
+ "epoch": 1.57,
839
+ "learning_rate": 1.4654667224780244e-05,
840
+ "loss": 0.5994,
841
+ "step": 1280
842
+ },
843
+ {
844
+ "epoch": 1.59,
845
+ "learning_rate": 1.4529091670154878e-05,
846
+ "loss": 0.5801,
847
+ "step": 1290
848
+ },
849
+ {
850
+ "epoch": 1.6,
851
+ "learning_rate": 1.4403516115529511e-05,
852
+ "loss": 0.5672,
853
+ "step": 1300
854
+ },
855
+ {
856
+ "epoch": 1.61,
857
+ "learning_rate": 1.4277940560904144e-05,
858
+ "loss": 0.5731,
859
+ "step": 1310
860
+ },
861
+ {
862
+ "epoch": 1.62,
863
+ "learning_rate": 1.4152365006278777e-05,
864
+ "loss": 0.5448,
865
+ "step": 1320
866
+ },
867
+ {
868
+ "epoch": 1.64,
869
+ "learning_rate": 1.4026789451653411e-05,
870
+ "loss": 0.617,
871
+ "step": 1330
872
+ },
873
+ {
874
+ "epoch": 1.65,
875
+ "learning_rate": 1.3901213897028046e-05,
876
+ "loss": 0.6464,
877
+ "step": 1340
878
+ },
879
+ {
880
+ "epoch": 1.66,
881
+ "learning_rate": 1.3775638342402679e-05,
882
+ "loss": 0.568,
883
+ "step": 1350
884
+ },
885
+ {
886
+ "epoch": 1.67,
887
+ "learning_rate": 1.3650062787777313e-05,
888
+ "loss": 0.6232,
889
+ "step": 1360
890
+ },
891
+ {
892
+ "epoch": 1.68,
893
+ "learning_rate": 1.3524487233151948e-05,
894
+ "loss": 0.5646,
895
+ "step": 1370
896
+ },
897
+ {
898
+ "epoch": 1.7,
899
+ "learning_rate": 1.3398911678526581e-05,
900
+ "loss": 0.563,
901
+ "step": 1380
902
+ },
903
+ {
904
+ "epoch": 1.71,
905
+ "learning_rate": 1.3273336123901215e-05,
906
+ "loss": 0.5761,
907
+ "step": 1390
908
+ },
909
+ {
910
+ "epoch": 1.72,
911
+ "learning_rate": 1.3147760569275847e-05,
912
+ "loss": 0.5987,
913
+ "step": 1400
914
+ },
915
+ {
916
+ "epoch": 1.73,
917
+ "learning_rate": 1.3022185014650481e-05,
918
+ "loss": 0.5446,
919
+ "step": 1410
920
+ },
921
+ {
922
+ "epoch": 1.75,
923
+ "learning_rate": 1.2896609460025116e-05,
924
+ "loss": 0.5977,
925
+ "step": 1420
926
+ },
927
+ {
928
+ "epoch": 1.76,
929
+ "learning_rate": 1.2771033905399749e-05,
930
+ "loss": 0.5778,
931
+ "step": 1430
932
+ },
933
+ {
934
+ "epoch": 1.77,
935
+ "learning_rate": 1.2645458350774383e-05,
936
+ "loss": 0.5553,
937
+ "step": 1440
938
+ },
939
+ {
940
+ "epoch": 1.78,
941
+ "learning_rate": 1.2519882796149016e-05,
942
+ "loss": 0.5804,
943
+ "step": 1450
944
+ },
945
+ {
946
+ "epoch": 1.8,
947
+ "learning_rate": 1.239430724152365e-05,
948
+ "loss": 0.5993,
949
+ "step": 1460
950
+ },
951
+ {
952
+ "epoch": 1.81,
953
+ "learning_rate": 1.2268731686898285e-05,
954
+ "loss": 0.5553,
955
+ "step": 1470
956
+ },
957
+ {
958
+ "epoch": 1.82,
959
+ "learning_rate": 1.2143156132272918e-05,
960
+ "loss": 0.5741,
961
+ "step": 1480
962
+ },
963
+ {
964
+ "epoch": 1.83,
965
+ "learning_rate": 1.2017580577647551e-05,
966
+ "loss": 0.5665,
967
+ "step": 1490
968
+ },
969
+ {
970
+ "epoch": 1.84,
971
+ "learning_rate": 1.1892005023022185e-05,
972
+ "loss": 0.6043,
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 sha256:79ec22b1f70bae4d68852f8fcd41d8be6b667b75d66d6a5009fc8c6a38b78e9b
3
+ size 4603