Hugging Face's logo

YOZ1
/
RadAI4

PEFT
Model card Files Community
1
YOZ1 commited on
Commit
75b6b37
•
1 Parent(s): e83c0ce

Upload 11 files

Browse files
Files changed (11) hide show
  1. README.md +32 -0
  2. adapter_config.json +21 -0
  3. adapter_model.bin +3 -0
  4. optimizer.pt +3 -0
  5. rng_state.pth +3 -0
  6. scheduler.pt +3 -0
  7. special_tokens_map.json +24 -0
  8. tokenizer.json +0 -0
  9. tokenizer_config.json +31 -0
  10. trainer_state.json +400 -0
  11. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,32 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - load_in_8bit: False
9
+ - load_in_4bit: True
10
+ - llm_int8_threshold: 6.0
11
+ - llm_int8_skip_modules: None
12
+ - llm_int8_enable_fp32_cpu_offload: False
13
+ - llm_int8_has_fp16_weight: False
14
+ - bnb_4bit_quant_type: nf4
15
+ - bnb_4bit_use_double_quant: False
16
+ - bnb_4bit_compute_dtype: float16
17
+
18
+ The following `bitsandbytes` quantization config was used during training:
19
+ - load_in_8bit: False
20
+ - load_in_4bit: True
21
+ - llm_int8_threshold: 6.0
22
+ - llm_int8_skip_modules: None
23
+ - llm_int8_enable_fp32_cpu_offload: False
24
+ - llm_int8_has_fp16_weight: False
25
+ - bnb_4bit_quant_type: nf4
26
+ - bnb_4bit_use_double_quant: False
27
+ - bnb_4bit_compute_dtype: float16
28
+ ### Framework versions
29
+
30
+ - PEFT 0.4.0
31
+
32
+ - PEFT 0.4.0
adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "daryl149/llama-2-13b-chat-hf",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 16,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 128,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a256c21f00d9b22ef0d07feac5b3bd68e89118ea7e394043c7d09d2e3f26ba1
3
+ size 419488077
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58a9fc85f27784960be1052c692b7f818376ca18fa0fb86065b592f8b799597b
3
+ size 838959685
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0c2e5c20d54aee9a1a5b17f4e49a28c4ab72940bb481741067f290257ceae8a9
3
+ size 14575
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:50b57a34df83b700e2c13775ff734b4569b74ce7e20da3479db76577bb4e906e
3
+ size 627
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": true,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "__type": "AddedToken",
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "clean_up_tokenization_spaces": false,
11
+ "eos_token": {
12
+ "__type": "AddedToken",
13
+ "content": "</s>",
14
+ "lstrip": false,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ "model_max_length": 1000000000000000019884624838656,
20
+ "pad_token": null,
21
+ "sp_model_kwargs": {},
22
+ "tokenizer_class": "LlamaTokenizer",
23
+ "unk_token": {
24
+ "__type": "AddedToken",
25
+ "content": "<unk>",
26
+ "lstrip": false,
27
+ "normalized": true,
28
+ "rstrip": false,
29
+ "single_word": false
30
+ }
31
+ }
trainer_state.json ADDED
@@ -0,0 +1,400 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 14.349775784753364,
5
+ "global_step": 3200,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.22,
12
+ "learning_rate": 0.0002,
13
+ "loss": 0.6202,
14
+ "step": 50
15
+ },
16
+ {
17
+ "epoch": 0.45,
18
+ "learning_rate": 0.0002,
19
+ "loss": 0.4487,
20
+ "step": 100
21
+ },
22
+ {
23
+ "epoch": 0.67,
24
+ "learning_rate": 0.0002,
25
+ "loss": 0.419,
26
+ "step": 150
27
+ },
28
+ {
29
+ "epoch": 0.9,
30
+ "learning_rate": 0.0002,
31
+ "loss": 0.4151,
32
+ "step": 200
33
+ },
34
+ {
35
+ "epoch": 1.12,
36
+ "learning_rate": 0.0002,
37
+ "loss": 0.3933,
38
+ "step": 250
39
+ },
40
+ {
41
+ "epoch": 1.35,
42
+ "learning_rate": 0.0002,
43
+ "loss": 0.3797,
44
+ "step": 300
45
+ },
46
+ {
47
+ "epoch": 1.57,
48
+ "learning_rate": 0.0002,
49
+ "loss": 0.3688,
50
+ "step": 350
51
+ },
52
+ {
53
+ "epoch": 1.79,
54
+ "learning_rate": 0.0002,
55
+ "loss": 0.3749,
56
+ "step": 400
57
+ },
58
+ {
59
+ "epoch": 2.02,
60
+ "learning_rate": 0.0002,
61
+ "loss": 0.3624,
62
+ "step": 450
63
+ },
64
+ {
65
+ "epoch": 2.24,
66
+ "learning_rate": 0.0002,
67
+ "loss": 0.3331,
68
+ "step": 500
69
+ },
70
+ {
71
+ "epoch": 2.47,
72
+ "learning_rate": 0.0002,
73
+ "loss": 0.3347,
74
+ "step": 550
75
+ },
76
+ {
77
+ "epoch": 2.69,
78
+ "learning_rate": 0.0002,
79
+ "loss": 0.3533,
80
+ "step": 600
81
+ },
82
+ {
83
+ "epoch": 2.91,
84
+ "learning_rate": 0.0002,
85
+ "loss": 0.3413,
86
+ "step": 650
87
+ },
88
+ {
89
+ "epoch": 3.14,
90
+ "learning_rate": 0.0002,
91
+ "loss": 0.3242,
92
+ "step": 700
93
+ },
94
+ {
95
+ "epoch": 3.36,
96
+ "learning_rate": 0.0002,
97
+ "loss": 0.3091,
98
+ "step": 750
99
+ },
100
+ {
101
+ "epoch": 3.59,
102
+ "learning_rate": 0.0002,
103
+ "loss": 0.3166,
104
+ "step": 800
105
+ },
106
+ {
107
+ "epoch": 3.81,
108
+ "learning_rate": 0.0002,
109
+ "loss": 0.3121,
110
+ "step": 850
111
+ },
112
+ {
113
+ "epoch": 4.04,
114
+ "learning_rate": 0.0002,
115
+ "loss": 0.304,
116
+ "step": 900
117
+ },
118
+ {
119
+ "epoch": 4.26,
120
+ "learning_rate": 0.0002,
121
+ "loss": 0.2776,
122
+ "step": 950
123
+ },
124
+ {
125
+ "epoch": 4.48,
126
+ "learning_rate": 0.0002,
127
+ "loss": 0.2783,
128
+ "step": 1000
129
+ },
130
+ {
131
+ "epoch": 4.71,
132
+ "learning_rate": 0.0002,
133
+ "loss": 0.2836,
134
+ "step": 1050
135
+ },
136
+ {
137
+ "epoch": 4.93,
138
+ "learning_rate": 0.0002,
139
+ "loss": 0.2931,
140
+ "step": 1100
141
+ },
142
+ {
143
+ "epoch": 5.16,
144
+ "learning_rate": 0.0002,
145
+ "loss": 0.2589,
146
+ "step": 1150
147
+ },
148
+ {
149
+ "epoch": 5.38,
150
+ "learning_rate": 0.0002,
151
+ "loss": 0.2525,
152
+ "step": 1200
153
+ },
154
+ {
155
+ "epoch": 5.61,
156
+ "learning_rate": 0.0002,
157
+ "loss": 0.255,
158
+ "step": 1250
159
+ },
160
+ {
161
+ "epoch": 5.83,
162
+ "learning_rate": 0.0002,
163
+ "loss": 0.2564,
164
+ "step": 1300
165
+ },
166
+ {
167
+ "epoch": 6.05,
168
+ "learning_rate": 0.0002,
169
+ "loss": 0.242,
170
+ "step": 1350
171
+ },
172
+ {
173
+ "epoch": 6.28,
174
+ "learning_rate": 0.0002,
175
+ "loss": 0.2092,
176
+ "step": 1400
177
+ },
178
+ {
179
+ "epoch": 6.5,
180
+ "learning_rate": 0.0002,
181
+ "loss": 0.2172,
182
+ "step": 1450
183
+ },
184
+ {
185
+ "epoch": 6.73,
186
+ "learning_rate": 0.0002,
187
+ "loss": 0.2233,
188
+ "step": 1500
189
+ },
190
+ {
191
+ "epoch": 6.95,
192
+ "learning_rate": 0.0002,
193
+ "loss": 0.2261,
194
+ "step": 1550
195
+ },
196
+ {
197
+ "epoch": 7.17,
198
+ "learning_rate": 0.0002,
199
+ "loss": 0.1944,
200
+ "step": 1600
201
+ },
202
+ {
203
+ "epoch": 7.4,
204
+ "learning_rate": 0.0002,
205
+ "loss": 0.1803,
206
+ "step": 1650
207
+ },
208
+ {
209
+ "epoch": 7.62,
210
+ "learning_rate": 0.0002,
211
+ "loss": 0.1852,
212
+ "step": 1700
213
+ },
214
+ {
215
+ "epoch": 7.85,
216
+ "learning_rate": 0.0002,
217
+ "loss": 0.1931,
218
+ "step": 1750
219
+ },
220
+ {
221
+ "epoch": 8.07,
222
+ "learning_rate": 0.0002,
223
+ "loss": 0.1755,
224
+ "step": 1800
225
+ },
226
+ {
227
+ "epoch": 8.3,
228
+ "learning_rate": 0.0002,
229
+ "loss": 0.1465,
230
+ "step": 1850
231
+ },
232
+ {
233
+ "epoch": 8.52,
234
+ "learning_rate": 0.0002,
235
+ "loss": 0.1535,
236
+ "step": 1900
237
+ },
238
+ {
239
+ "epoch": 8.74,
240
+ "learning_rate": 0.0002,
241
+ "loss": 0.1542,
242
+ "step": 1950
243
+ },
244
+ {
245
+ "epoch": 8.97,
246
+ "learning_rate": 0.0002,
247
+ "loss": 0.161,
248
+ "step": 2000
249
+ },
250
+ {
251
+ "epoch": 9.19,
252
+ "learning_rate": 0.0002,
253
+ "loss": 0.1205,
254
+ "step": 2050
255
+ },
256
+ {
257
+ "epoch": 9.42,
258
+ "learning_rate": 0.0002,
259
+ "loss": 0.1197,
260
+ "step": 2100
261
+ },
262
+ {
263
+ "epoch": 9.64,
264
+ "learning_rate": 0.0002,
265
+ "loss": 0.1242,
266
+ "step": 2150
267
+ },
268
+ {
269
+ "epoch": 9.87,
270
+ "learning_rate": 0.0002,
271
+ "loss": 0.1277,
272
+ "step": 2200
273
+ },
274
+ {
275
+ "epoch": 10.09,
276
+ "learning_rate": 0.0002,
277
+ "loss": 0.1151,
278
+ "step": 2250
279
+ },
280
+ {
281
+ "epoch": 10.31,
282
+ "learning_rate": 0.0002,
283
+ "loss": 0.0925,
284
+ "step": 2300
285
+ },
286
+ {
287
+ "epoch": 10.54,
288
+ "learning_rate": 0.0002,
289
+ "loss": 0.0969,
290
+ "step": 2350
291
+ },
292
+ {
293
+ "epoch": 10.76,
294
+ "learning_rate": 0.0002,
295
+ "loss": 0.1023,
296
+ "step": 2400
297
+ },
298
+ {
299
+ "epoch": 10.99,
300
+ "learning_rate": 0.0002,
301
+ "loss": 0.1016,
302
+ "step": 2450
303
+ },
304
+ {
305
+ "epoch": 11.21,
306
+ "learning_rate": 0.0002,
307
+ "loss": 0.0708,
308
+ "step": 2500
309
+ },
310
+ {
311
+ "epoch": 11.43,
312
+ "learning_rate": 0.0002,
313
+ "loss": 0.0735,
314
+ "step": 2550
315
+ },
316
+ {
317
+ "epoch": 11.66,
318
+ "learning_rate": 0.0002,
319
+ "loss": 0.0804,
320
+ "step": 2600
321
+ },
322
+ {
323
+ "epoch": 11.88,
324
+ "learning_rate": 0.0002,
325
+ "loss": 0.0833,
326
+ "step": 2650
327
+ },
328
+ {
329
+ "epoch": 12.11,
330
+ "learning_rate": 0.0002,
331
+ "loss": 0.072,
332
+ "step": 2700
333
+ },
334
+ {
335
+ "epoch": 12.33,
336
+ "learning_rate": 0.0002,
337
+ "loss": 0.0592,
338
+ "step": 2750
339
+ },
340
+ {
341
+ "epoch": 12.56,
342
+ "learning_rate": 0.0002,
343
+ "loss": 0.0585,
344
+ "step": 2800
345
+ },
346
+ {
347
+ "epoch": 12.78,
348
+ "learning_rate": 0.0002,
349
+ "loss": 0.0647,
350
+ "step": 2850
351
+ },
352
+ {
353
+ "epoch": 13.0,
354
+ "learning_rate": 0.0002,
355
+ "loss": 0.0656,
356
+ "step": 2900
357
+ },
358
+ {
359
+ "epoch": 13.23,
360
+ "learning_rate": 0.0002,
361
+ "loss": 0.0473,
362
+ "step": 2950
363
+ },
364
+ {
365
+ "epoch": 13.45,
366
+ "learning_rate": 0.0002,
367
+ "loss": 0.0509,
368
+ "step": 3000
369
+ },
370
+ {
371
+ "epoch": 13.68,
372
+ "learning_rate": 0.0002,
373
+ "loss": 0.0523,
374
+ "step": 3050
375
+ },
376
+ {
377
+ "epoch": 13.9,
378
+ "learning_rate": 0.0002,
379
+ "loss": 0.0531,
380
+ "step": 3100
381
+ },
382
+ {
383
+ "epoch": 14.13,
384
+ "learning_rate": 0.0002,
385
+ "loss": 0.0455,
386
+ "step": 3150
387
+ },
388
+ {
389
+ "epoch": 14.35,
390
+ "learning_rate": 0.0002,
391
+ "loss": 0.0407,
392
+ "step": 3200
393
+ }
394
+ ],
395
+ "max_steps": 3345,
396
+ "num_train_epochs": 15,
397
+ "total_flos": 5.0781774924831744e+17,
398
+ "trial_name": null,
399
+ "trial_params": null
400
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b09b1caf24a0186a166f53e562937b0d693d105be7ba739039a5644a93038f31
3
+ size 3963