JonaszPotoniec commited on
Commit
d360833
1 Parent(s): 5b709ae

Upload 11 files

Browse files
lora/adapter_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "mistralai/Mistral-7B-Instruct-v0.2",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 32,
13
+ "lora_dropout": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 32,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "k_proj",
23
+ "v_proj",
24
+ "q_proj",
25
+ "o_proj"
26
+ ],
27
+ "task_type": "CAUSAL_LM",
28
+ "use_rslora": false
29
+ }
lora/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:55d9d1451bdccb31f32cfc911e78164a142769326e45609b495b497055573696
3
+ size 109086416
lora/ggml-adapter-model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7b5d2b3f660095796895f335111b16762034309ac19a6c65e0094282371fbec0
3
+ size 109068288
lora/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eab7de143b6fd9dce17b8d63942edbd7330b32a1300ed35b9511063b2c13b3be
3
+ size 218255930
lora/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:83635e1e4a4d60e53c63282862f9f8697a33c0f01053ec0337496fd7356d5aed
3
+ size 14244
lora/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e7b7d9eda0a081ddcf93d2c46ff42aaacea03d93d3428c69b6a8e0e2e93eeed
3
+ size 1064
lora/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
lora/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
lora/tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "additional_special_tokens": [],
31
+ "bos_token": "<s>",
32
+ "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}",
33
+ "clean_up_tokenization_spaces": false,
34
+ "eos_token": "</s>",
35
+ "legacy": true,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": "</s>",
38
+ "sp_model_kwargs": {},
39
+ "spaces_between_special_tokens": false,
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
lora/trainer_state.json ADDED
@@ -0,0 +1,915 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.3047177107501935,
5
+ "eval_steps": 500,
6
+ "global_step": 1490,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.02,
13
+ "learning_rate": 0.000199739323151795,
14
+ "loss": 1.5905,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.03,
19
+ "learning_rate": 0.00019682229406025635,
20
+ "loss": 1.3988,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.05,
25
+ "learning_rate": 0.00019075754196709572,
26
+ "loss": 1.4513,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.06,
31
+ "learning_rate": 0.00018174223385588917,
32
+ "loss": 1.3878,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.08,
37
+ "learning_rate": 0.00016949152542372882,
38
+ "loss": 1.2662,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.09,
43
+ "learning_rate": 0.0001999998602293167,
44
+ "loss": 1.0731,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.11,
49
+ "learning_rate": 0.0001999830882200816,
50
+ "loss": 1.0733,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.12,
55
+ "learning_rate": 0.0001999383674462943,
56
+ "loss": 1.0432,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.14,
61
+ "learning_rate": 0.00019986571040897272,
62
+ "loss": 1.0459,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.15,
67
+ "learning_rate": 0.00019976513741829603,
68
+ "loss": 1.0564,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.17,
73
+ "learning_rate": 0.00019963667658792704,
74
+ "loss": 1.006,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.19,
79
+ "learning_rate": 0.00019948036382715371,
80
+ "loss": 1.0124,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.2,
85
+ "learning_rate": 0.0001992962428308511,
86
+ "loss": 1.0495,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.22,
91
+ "learning_rate": 0.00019908436506726714,
92
+ "loss": 1.0061,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.23,
97
+ "learning_rate": 0.00019884478976363548,
98
+ "loss": 1.0166,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.25,
103
+ "learning_rate": 0.00019857758388961943,
104
+ "loss": 0.986,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.26,
109
+ "learning_rate": 0.0001982828221385916,
110
+ "loss": 1.0052,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.28,
115
+ "learning_rate": 0.00019796058690675435,
116
+ "loss": 0.9986,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.29,
121
+ "learning_rate": 0.0001976109682701075,
122
+ "loss": 0.989,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.31,
127
+ "learning_rate": 0.00019723406395926856,
128
+ "loss": 1.0184,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.32,
133
+ "learning_rate": 0.00019682997933215385,
134
+ "loss": 0.9896,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.34,
139
+ "learning_rate": 0.00019639882734452722,
140
+ "loss": 1.0046,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 0.36,
145
+ "learning_rate": 0.000195940728518425,
146
+ "loss": 0.9922,
147
+ "step": 230
148
+ },
149
+ {
150
+ "epoch": 0.37,
151
+ "learning_rate": 0.00019545581090846584,
152
+ "loss": 0.9665,
153
+ "step": 240
154
+ },
155
+ {
156
+ "epoch": 0.39,
157
+ "learning_rate": 0.00019494421006605492,
158
+ "loss": 1.0012,
159
+ "step": 250
160
+ },
161
+ {
162
+ "epoch": 0.4,
163
+ "learning_rate": 0.00019440606900149275,
164
+ "loss": 0.9937,
165
+ "step": 260
166
+ },
167
+ {
168
+ "epoch": 0.42,
169
+ "learning_rate": 0.0001938415381439987,
170
+ "loss": 0.984,
171
+ "step": 270
172
+ },
173
+ {
174
+ "epoch": 0.43,
175
+ "learning_rate": 0.00019325077529966077,
176
+ "loss": 0.9651,
177
+ "step": 280
178
+ },
179
+ {
180
+ "epoch": 0.45,
181
+ "learning_rate": 0.00019263394560732326,
182
+ "loss": 0.9621,
183
+ "step": 290
184
+ },
185
+ {
186
+ "epoch": 0.46,
187
+ "learning_rate": 0.00019199122149242485,
188
+ "loss": 0.9761,
189
+ "step": 300
190
+ },
191
+ {
192
+ "epoch": 0.48,
193
+ "learning_rate": 0.00019132278261879945,
194
+ "loss": 0.9792,
195
+ "step": 310
196
+ },
197
+ {
198
+ "epoch": 0.49,
199
+ "learning_rate": 0.00019062881583845398,
200
+ "loss": 0.9918,
201
+ "step": 320
202
+ },
203
+ {
204
+ "epoch": 0.51,
205
+ "learning_rate": 0.00018990951513933683,
206
+ "loss": 0.969,
207
+ "step": 330
208
+ },
209
+ {
210
+ "epoch": 0.53,
211
+ "learning_rate": 0.00018916508159111126,
212
+ "loss": 0.9513,
213
+ "step": 340
214
+ },
215
+ {
216
+ "epoch": 0.54,
217
+ "learning_rate": 0.0001883957232889495,
218
+ "loss": 0.9987,
219
+ "step": 350
220
+ },
221
+ {
222
+ "epoch": 0.56,
223
+ "learning_rate": 0.00018760165529536286,
224
+ "loss": 0.9687,
225
+ "step": 360
226
+ },
227
+ {
228
+ "epoch": 0.57,
229
+ "learning_rate": 0.00018678309958008432,
230
+ "loss": 0.9734,
231
+ "step": 370
232
+ },
233
+ {
234
+ "epoch": 0.59,
235
+ "learning_rate": 0.00018594028495802015,
236
+ "loss": 0.9705,
237
+ "step": 380
238
+ },
239
+ {
240
+ "epoch": 0.6,
241
+ "learning_rate": 0.00018507344702528816,
242
+ "loss": 0.949,
243
+ "step": 390
244
+ },
245
+ {
246
+ "epoch": 0.62,
247
+ "learning_rate": 0.0001841828280933604,
248
+ "loss": 0.9709,
249
+ "step": 400
250
+ },
251
+ {
252
+ "epoch": 0.63,
253
+ "learning_rate": 0.00018326867712132865,
254
+ "loss": 0.9484,
255
+ "step": 410
256
+ },
257
+ {
258
+ "epoch": 0.65,
259
+ "learning_rate": 0.00018233124964631156,
260
+ "loss": 0.9677,
261
+ "step": 420
262
+ },
263
+ {
264
+ "epoch": 0.67,
265
+ "learning_rate": 0.00018137080771202325,
266
+ "loss": 0.9597,
267
+ "step": 430
268
+ },
269
+ {
270
+ "epoch": 0.68,
271
+ "learning_rate": 0.00018038761979552285,
272
+ "loss": 0.9289,
273
+ "step": 440
274
+ },
275
+ {
276
+ "epoch": 0.7,
277
+ "learning_rate": 0.00017938196073216586,
278
+ "loss": 0.9749,
279
+ "step": 450
280
+ },
281
+ {
282
+ "epoch": 0.71,
283
+ "learning_rate": 0.00017835411163877782,
284
+ "loss": 0.9445,
285
+ "step": 460
286
+ },
287
+ {
288
+ "epoch": 0.73,
289
+ "learning_rate": 0.00017730435983507277,
290
+ "loss": 0.9609,
291
+ "step": 470
292
+ },
293
+ {
294
+ "epoch": 0.74,
295
+ "learning_rate": 0.00017623299876333697,
296
+ "loss": 0.9652,
297
+ "step": 480
298
+ },
299
+ {
300
+ "epoch": 0.76,
301
+ "learning_rate": 0.0001751403279064018,
302
+ "loss": 0.9415,
303
+ "step": 490
304
+ },
305
+ {
306
+ "epoch": 0.77,
307
+ "learning_rate": 0.0001740266527039276,
308
+ "loss": 0.9769,
309
+ "step": 500
310
+ },
311
+ {
312
+ "epoch": 0.79,
313
+ "learning_rate": 0.00017289228446702292,
314
+ "loss": 0.9477,
315
+ "step": 510
316
+ },
317
+ {
318
+ "epoch": 0.8,
319
+ "learning_rate": 0.00017173754029122188,
320
+ "loss": 0.9622,
321
+ "step": 520
322
+ },
323
+ {
324
+ "epoch": 0.82,
325
+ "learning_rate": 0.00017056274296784518,
326
+ "loss": 0.9631,
327
+ "step": 530
328
+ },
329
+ {
330
+ "epoch": 0.84,
331
+ "learning_rate": 0.00016936822089376837,
332
+ "loss": 0.9482,
333
+ "step": 540
334
+ },
335
+ {
336
+ "epoch": 0.85,
337
+ "learning_rate": 0.00016815430797962374,
338
+ "loss": 0.9666,
339
+ "step": 550
340
+ },
341
+ {
342
+ "epoch": 0.87,
343
+ "learning_rate": 0.00016692134355646046,
344
+ "loss": 0.9331,
345
+ "step": 560
346
+ },
347
+ {
348
+ "epoch": 0.88,
349
+ "learning_rate": 0.00016566967228088972,
350
+ "loss": 0.9395,
351
+ "step": 570
352
+ },
353
+ {
354
+ "epoch": 0.9,
355
+ "learning_rate": 0.0001643996440387412,
356
+ "loss": 0.9432,
357
+ "step": 580
358
+ },
359
+ {
360
+ "epoch": 0.91,
361
+ "learning_rate": 0.0001631116138472578,
362
+ "loss": 0.9244,
363
+ "step": 590
364
+ },
365
+ {
366
+ "epoch": 0.93,
367
+ "learning_rate": 0.00016180594175585586,
368
+ "loss": 0.9617,
369
+ "step": 600
370
+ },
371
+ {
372
+ "epoch": 0.94,
373
+ "learning_rate": 0.00016048299274547883,
374
+ "loss": 0.9213,
375
+ "step": 610
376
+ },
377
+ {
378
+ "epoch": 0.96,
379
+ "learning_rate": 0.00015914313662657224,
380
+ "loss": 0.9497,
381
+ "step": 620
382
+ },
383
+ {
384
+ "epoch": 0.97,
385
+ "learning_rate": 0.00015778674793570896,
386
+ "loss": 0.9377,
387
+ "step": 630
388
+ },
389
+ {
390
+ "epoch": 0.99,
391
+ "learning_rate": 0.00015641420583089295,
392
+ "loss": 0.9205,
393
+ "step": 640
394
+ },
395
+ {
396
+ "epoch": 1.01,
397
+ "learning_rate": 0.00015502589398557146,
398
+ "loss": 0.9664,
399
+ "step": 650
400
+ },
401
+ {
402
+ "epoch": 1.02,
403
+ "learning_rate": 0.0001536222004813849,
404
+ "loss": 0.86,
405
+ "step": 660
406
+ },
407
+ {
408
+ "epoch": 1.04,
409
+ "learning_rate": 0.0001522035176996845,
410
+ "loss": 0.8729,
411
+ "step": 670
412
+ },
413
+ {
414
+ "epoch": 1.05,
415
+ "learning_rate": 0.00015077024221184793,
416
+ "loss": 0.8386,
417
+ "step": 680
418
+ },
419
+ {
420
+ "epoch": 1.07,
421
+ "learning_rate": 0.00014932277466842377,
422
+ "loss": 0.836,
423
+ "step": 690
424
+ },
425
+ {
426
+ "epoch": 1.08,
427
+ "learning_rate": 0.0001478615196871358,
428
+ "loss": 0.8874,
429
+ "step": 700
430
+ },
431
+ {
432
+ "epoch": 1.1,
433
+ "learning_rate": 0.00014638688573977805,
434
+ "loss": 0.8546,
435
+ "step": 710
436
+ },
437
+ {
438
+ "epoch": 1.11,
439
+ "learning_rate": 0.00014489928503803285,
440
+ "loss": 0.8682,
441
+ "step": 720
442
+ },
443
+ {
444
+ "epoch": 1.13,
445
+ "learning_rate": 0.00014339913341824314,
446
+ "loss": 0.8407,
447
+ "step": 730
448
+ },
449
+ {
450
+ "epoch": 1.14,
451
+ "learning_rate": 0.0001418868502251717,
452
+ "loss": 0.8476,
453
+ "step": 740
454
+ },
455
+ {
456
+ "epoch": 1.16,
457
+ "learning_rate": 0.00014036285819477967,
458
+ "loss": 0.8962,
459
+ "step": 750
460
+ },
461
+ {
462
+ "epoch": 1.18,
463
+ "learning_rate": 0.0001388275833360572,
464
+ "loss": 0.8799,
465
+ "step": 760
466
+ },
467
+ {
468
+ "epoch": 1.19,
469
+ "learning_rate": 0.00013728145481193882,
470
+ "loss": 0.8585,
471
+ "step": 770
472
+ },
473
+ {
474
+ "epoch": 1.21,
475
+ "learning_rate": 0.00013572490481933758,
476
+ "loss": 0.8541,
477
+ "step": 780
478
+ },
479
+ {
480
+ "epoch": 1.22,
481
+ "learning_rate": 0.000134158368468331,
482
+ "loss": 0.8239,
483
+ "step": 790
484
+ },
485
+ {
486
+ "epoch": 1.24,
487
+ "learning_rate": 0.0001325822836605323,
488
+ "loss": 0.8758,
489
+ "step": 800
490
+ },
491
+ {
492
+ "epoch": 1.25,
493
+ "learning_rate": 0.00013099709096668193,
494
+ "loss": 0.8682,
495
+ "step": 810
496
+ },
497
+ {
498
+ "epoch": 1.27,
499
+ "learning_rate": 0.00012940323350349258,
500
+ "loss": 0.8588,
501
+ "step": 820
502
+ },
503
+ {
504
+ "epoch": 1.28,
505
+ "learning_rate": 0.0001278011568097824,
506
+ "loss": 0.8455,
507
+ "step": 830
508
+ },
509
+ {
510
+ "epoch": 1.3,
511
+ "learning_rate": 0.00012619130872193162,
512
+ "loss": 0.8485,
513
+ "step": 840
514
+ },
515
+ {
516
+ "epoch": 1.31,
517
+ "learning_rate": 0.00012457413924869644,
518
+ "loss": 0.8995,
519
+ "step": 850
520
+ },
521
+ {
522
+ "epoch": 1.33,
523
+ "learning_rate": 0.0001229501004454159,
524
+ "loss": 0.8736,
525
+ "step": 860
526
+ },
527
+ {
528
+ "epoch": 1.35,
529
+ "learning_rate": 0.00012131964628764678,
530
+ "loss": 0.8638,
531
+ "step": 870
532
+ },
533
+ {
534
+ "epoch": 1.36,
535
+ "learning_rate": 0.00011968323254426135,
536
+ "loss": 0.8412,
537
+ "step": 880
538
+ },
539
+ {
540
+ "epoch": 1.38,
541
+ "learning_rate": 0.00011804131665004423,
542
+ "loss": 0.815,
543
+ "step": 890
544
+ },
545
+ {
546
+ "epoch": 1.39,
547
+ "learning_rate": 0.00011639435757782336,
548
+ "loss": 0.8969,
549
+ "step": 900
550
+ },
551
+ {
552
+ "epoch": 1.41,
553
+ "learning_rate": 0.0001147428157101709,
554
+ "loss": 0.8737,
555
+ "step": 910
556
+ },
557
+ {
558
+ "epoch": 1.42,
559
+ "learning_rate": 0.00011308715271071049,
560
+ "loss": 0.8604,
561
+ "step": 920
562
+ },
563
+ {
564
+ "epoch": 1.44,
565
+ "learning_rate": 0.00011142783139506601,
566
+ "loss": 0.8425,
567
+ "step": 930
568
+ },
569
+ {
570
+ "epoch": 1.45,
571
+ "learning_rate": 0.00010976531560148841,
572
+ "loss": 0.8262,
573
+ "step": 940
574
+ },
575
+ {
576
+ "epoch": 1.47,
577
+ "learning_rate": 0.00010810007006119685,
578
+ "loss": 0.9044,
579
+ "step": 950
580
+ },
581
+ {
582
+ "epoch": 1.48,
583
+ "learning_rate": 0.00010643256026846992,
584
+ "loss": 0.8575,
585
+ "step": 960
586
+ },
587
+ {
588
+ "epoch": 1.5,
589
+ "learning_rate": 0.00010476325235052389,
590
+ "loss": 0.8602,
591
+ "step": 970
592
+ },
593
+ {
594
+ "epoch": 1.52,
595
+ "learning_rate": 0.00010309261293721384,
596
+ "loss": 0.8507,
597
+ "step": 980
598
+ },
599
+ {
600
+ "epoch": 1.53,
601
+ "learning_rate": 0.00010142110903059424,
602
+ "loss": 0.8279,
603
+ "step": 990
604
+ },
605
+ {
606
+ "epoch": 1.55,
607
+ "learning_rate": 9.974920787437567e-05,
608
+ "loss": 0.8845,
609
+ "step": 1000
610
+ },
611
+ {
612
+ "epoch": 1.56,
613
+ "learning_rate": 9.807737682331383e-05,
614
+ "loss": 0.8515,
615
+ "step": 1010
616
+ },
617
+ {
618
+ "epoch": 1.58,
619
+ "learning_rate": 9.640608321256761e-05,
620
+ "loss": 0.8467,
621
+ "step": 1020
622
+ },
623
+ {
624
+ "epoch": 1.59,
625
+ "learning_rate": 9.47357942270625e-05,
626
+ "loss": 0.8406,
627
+ "step": 1030
628
+ },
629
+ {
630
+ "epoch": 1.61,
631
+ "learning_rate": 9.306697677089621e-05,
632
+ "loss": 0.8435,
633
+ "step": 1040
634
+ },
635
+ {
636
+ "epoch": 1.62,
637
+ "learning_rate": 9.140009733682262e-05,
638
+ "loss": 0.8836,
639
+ "step": 1050
640
+ },
641
+ {
642
+ "epoch": 1.64,
643
+ "learning_rate": 8.973562187585071e-05,
644
+ "loss": 0.866,
645
+ "step": 1060
646
+ },
647
+ {
648
+ "epoch": 1.66,
649
+ "learning_rate": 8.80740156669951e-05,
650
+ "loss": 0.8585,
651
+ "step": 1070
652
+ },
653
+ {
654
+ "epoch": 1.67,
655
+ "learning_rate": 8.641574318721413e-05,
656
+ "loss": 0.847,
657
+ "step": 1080
658
+ },
659
+ {
660
+ "epoch": 1.69,
661
+ "learning_rate": 8.476126798157258e-05,
662
+ "loss": 0.8221,
663
+ "step": 1090
664
+ },
665
+ {
666
+ "epoch": 1.7,
667
+ "learning_rate": 8.311105253366448e-05,
668
+ "loss": 0.8873,
669
+ "step": 1100
670
+ },
671
+ {
672
+ "epoch": 1.72,
673
+ "learning_rate": 8.146555813633274e-05,
674
+ "loss": 0.8415,
675
+ "step": 1110
676
+ },
677
+ {
678
+ "epoch": 1.73,
679
+ "learning_rate": 7.982524476272188e-05,
680
+ "loss": 0.8533,
681
+ "step": 1120
682
+ },
683
+ {
684
+ "epoch": 1.75,
685
+ "learning_rate": 7.819057093769931e-05,
686
+ "loss": 0.8281,
687
+ "step": 1130
688
+ },
689
+ {
690
+ "epoch": 1.76,
691
+ "learning_rate": 7.656199360968167e-05,
692
+ "loss": 0.8319,
693
+ "step": 1140
694
+ },
695
+ {
696
+ "epoch": 1.78,
697
+ "learning_rate": 7.493996802290186e-05,
698
+ "loss": 0.8593,
699
+ "step": 1150
700
+ },
701
+ {
702
+ "epoch": 1.79,
703
+ "learning_rate": 7.332494759015226e-05,
704
+ "loss": 0.8361,
705
+ "step": 1160
706
+ },
707
+ {
708
+ "epoch": 1.81,
709
+ "learning_rate": 7.171738376604012e-05,
710
+ "loss": 0.8536,
711
+ "step": 1170
712
+ },
713
+ {
714
+ "epoch": 1.83,
715
+ "learning_rate": 7.011772592079013e-05,
716
+ "loss": 0.8478,
717
+ "step": 1180
718
+ },
719
+ {
720
+ "epoch": 1.84,
721
+ "learning_rate": 6.85264212146299e-05,
722
+ "loss": 0.8276,
723
+ "step": 1190
724
+ },
725
+ {
726
+ "epoch": 1.86,
727
+ "learning_rate": 6.694391447279287e-05,
728
+ "loss": 0.8934,
729
+ "step": 1200
730
+ },
731
+ {
732
+ "epoch": 1.87,
733
+ "learning_rate": 6.537064806117435e-05,
734
+ "loss": 0.8359,
735
+ "step": 1210
736
+ },
737
+ {
738
+ "epoch": 1.89,
739
+ "learning_rate": 6.380706176267455e-05,
740
+ "loss": 0.8477,
741
+ "step": 1220
742
+ },
743
+ {
744
+ "epoch": 1.9,
745
+ "learning_rate": 6.225359265426414e-05,
746
+ "loss": 0.8293,
747
+ "step": 1230
748
+ },
749
+ {
750
+ "epoch": 1.92,
751
+ "learning_rate": 6.071067498480583e-05,
752
+ "loss": 0.828,
753
+ "step": 1240
754
+ },
755
+ {
756
+ "epoch": 1.93,
757
+ "learning_rate": 5.9178740053666706e-05,
758
+ "loss": 0.8773,
759
+ "step": 1250
760
+ },
761
+ {
762
+ "epoch": 1.95,
763
+ "learning_rate": 5.765821609015495e-05,
764
+ "loss": 0.8376,
765
+ "step": 1260
766
+ },
767
+ {
768
+ "epoch": 1.96,
769
+ "learning_rate": 5.61495281338148e-05,
770
+ "loss": 0.8384,
771
+ "step": 1270
772
+ },
773
+ {
774
+ "epoch": 1.98,
775
+ "learning_rate": 5.465309791561328e-05,
776
+ "loss": 0.8403,
777
+ "step": 1280
778
+ },
779
+ {
780
+ "epoch": 2.0,
781
+ "learning_rate": 5.316934374005142e-05,
782
+ "loss": 0.8464,
783
+ "step": 1290
784
+ },
785
+ {
786
+ "epoch": 2.01,
787
+ "learning_rate": 5.1698680368233864e-05,
788
+ "loss": 0.8334,
789
+ "step": 1300
790
+ },
791
+ {
792
+ "epoch": 2.03,
793
+ "learning_rate": 5.02415189019283e-05,
794
+ "loss": 0.7771,
795
+ "step": 1310
796
+ },
797
+ {
798
+ "epoch": 2.04,
799
+ "learning_rate": 4.879826666864852e-05,
800
+ "loss": 0.7607,
801
+ "step": 1320
802
+ },
803
+ {
804
+ "epoch": 2.06,
805
+ "learning_rate": 4.736932710779173e-05,
806
+ "loss": 0.7367,
807
+ "step": 1330
808
+ },
809
+ {
810
+ "epoch": 2.07,
811
+ "learning_rate": 4.595509965786368e-05,
812
+ "loss": 0.763,
813
+ "step": 1340
814
+ },
815
+ {
816
+ "epoch": 2.09,
817
+ "learning_rate": 4.4555979644821145e-05,
818
+ "loss": 0.7878,
819
+ "step": 1350
820
+ },
821
+ {
822
+ "epoch": 2.1,
823
+ "learning_rate": 4.3172358171564916e-05,
824
+ "loss": 0.763,
825
+ "step": 1360
826
+ },
827
+ {
828
+ "epoch": 2.12,
829
+ "learning_rate": 4.180462200861256e-05,
830
+ "loss": 0.7487,
831
+ "step": 1370
832
+ },
833
+ {
834
+ "epoch": 2.13,
835
+ "learning_rate": 4.045315348598283e-05,
836
+ "loss": 0.738,
837
+ "step": 1380
838
+ },
839
+ {
840
+ "epoch": 2.15,
841
+ "learning_rate": 3.9118330386320814e-05,
842
+ "loss": 0.7439,
843
+ "step": 1390
844
+ },
845
+ {
846
+ "epoch": 2.17,
847
+ "learning_rate": 3.780052583929475e-05,
848
+ "loss": 0.7903,
849
+ "step": 1400
850
+ },
851
+ {
852
+ "epoch": 2.18,
853
+ "learning_rate": 3.650010821729303e-05,
854
+ "loss": 0.7653,
855
+ "step": 1410
856
+ },
857
+ {
858
+ "epoch": 2.2,
859
+ "learning_rate": 3.521744103245145e-05,
860
+ "loss": 0.7447,
861
+ "step": 1420
862
+ },
863
+ {
864
+ "epoch": 2.21,
865
+ "learning_rate": 3.395288283503867e-05,
866
+ "loss": 0.7186,
867
+ "step": 1430
868
+ },
869
+ {
870
+ "epoch": 2.23,
871
+ "learning_rate": 3.2706787113229075e-05,
872
+ "loss": 0.7259,
873
+ "step": 1440
874
+ },
875
+ {
876
+ "epoch": 2.24,
877
+ "learning_rate": 3.147950219429036e-05,
878
+ "loss": 0.8041,
879
+ "step": 1450
880
+ },
881
+ {
882
+ "epoch": 2.26,
883
+ "learning_rate": 3.027137114721401e-05,
884
+ "loss": 0.7545,
885
+ "step": 1460
886
+ },
887
+ {
888
+ "epoch": 2.27,
889
+ "learning_rate": 2.9082731686815412e-05,
890
+ "loss": 0.7363,
891
+ "step": 1470
892
+ },
893
+ {
894
+ "epoch": 2.29,
895
+ "learning_rate": 2.7913916079330905e-05,
896
+ "loss": 0.7364,
897
+ "step": 1480
898
+ },
899
+ {
900
+ "epoch": 2.3,
901
+ "learning_rate": 2.676525104953759e-05,
902
+ "loss": 0.7509,
903
+ "step": 1490
904
+ }
905
+ ],
906
+ "logging_steps": 10,
907
+ "max_steps": 1938,
908
+ "num_input_tokens_seen": 0,
909
+ "num_train_epochs": 3,
910
+ "save_steps": 10,
911
+ "total_flos": 4.7276257552131686e+17,
912
+ "train_batch_size": 16,
913
+ "trial_name": null,
914
+ "trial_params": null
915
+ }
lora/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c500541aa668ea5363dd7ebb8b8c75b1643e44ba2af49808565c3de188ce16d
3
+ size 4728