Hugging Face's logo

Glazastik
/
GlazastikChat

Text Generation
PEFT
Russian
English
RefinedWebModel
falcon
custom_code
Model card Files Community
1
Daniil-plotnikov commited on
Commit
7b1ddfc
•
1 Parent(s): 4227e52

Upload 11 files

Browse files
Files changed (11) hide show
  1. README.md +34 -0
  2. adapter_config.json +23 -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 +17 -0
  8. tokenizer.json +0 -0
  9. tokenizer_config.json +7 -0
  10. trainer_state.json +319 -0
  11. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+
7
+ The following `bitsandbytes` quantization config was used during training:
8
+ - quant_method: bitsandbytes
9
+ - load_in_8bit: False
10
+ - load_in_4bit: True
11
+ - llm_int8_threshold: 6.0
12
+ - llm_int8_skip_modules: None
13
+ - llm_int8_enable_fp32_cpu_offload: False
14
+ - llm_int8_has_fp16_weight: False
15
+ - bnb_4bit_quant_type: nf4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float16
18
+
19
+ The following `bitsandbytes` quantization config was used during training:
20
+ - quant_method: bitsandbytes
21
+ - load_in_8bit: False
22
+ - load_in_4bit: True
23
+ - llm_int8_threshold: 6.0
24
+ - llm_int8_skip_modules: None
25
+ - llm_int8_enable_fp32_cpu_offload: False
26
+ - llm_int8_has_fp16_weight: False
27
+ - bnb_4bit_quant_type: nf4
28
+ - bnb_4bit_use_double_quant: False
29
+ - bnb_4bit_compute_dtype: float16
30
+ ### Framework versions
31
+
32
+ - PEFT 0.6.0.dev0
33
+
34
+ - PEFT 0.6.0.dev0
adapter_config.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "ybelkada/falcon-7b-sharded-bf16",
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": 64,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "query_key_value",
18
+ "dense",
19
+ "dense_h_to_4h",
20
+ "dense_4h_to_h"
21
+ ],
22
+ "task_type": "CAUSAL_LM"
23
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:41c01eb3e004f3da2ac8e7a79c240720641fd2b09ff5003e70b216e45b9819ba
3
+ size 522284877
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74732fbfaa07354aa2db1b0925f8ab472258fe719139037b1e8cfaf4cbecdbc5
3
+ size 1044539909
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd03062f7a31e65a1e0ef0313490c31eb38905bf7c54f0409667a8d9019ec1ad
3
+ size 14575
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5efc416a6883409dd7ab6f5c779e107c7c2baa7af6e12ed9fbd9dd73b8b20784
3
+ size 627
special_tokens_map.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ ">>TITLE<<",
4
+ ">>ABSTRACT<<",
5
+ ">>INTRODUCTION<<",
6
+ ">>SUMMARY<<",
7
+ ">>COMMENT<<",
8
+ ">>ANSWER<<",
9
+ ">>QUESTION<<",
10
+ ">>DOMAIN<<",
11
+ ">>PREFIX<<",
12
+ ">>SUFFIX<<",
13
+ ">>MIDDLE<<"
14
+ ],
15
+ "eos_token": "<|endoftext|>",
16
+ "pad_token": "<|endoftext|>"
17
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "clean_up_tokenization_spaces": true,
4
+ "eos_token": "<|endoftext|>",
5
+ "model_max_length": 2048,
6
+ "tokenizer_class": "PreTrainedTokenizerFast"
7
+ }
trainer_state.json ADDED
@@ -0,0 +1,319 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.14938751120406335,
5
+ "eval_steps": 500,
6
+ "global_step": 500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.0,
13
+ "learning_rate": 0.0002,
14
+ "loss": 1.0618,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.01,
19
+ "learning_rate": 0.0002,
20
+ "loss": 0.9484,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.01,
25
+ "learning_rate": 0.0002,
26
+ "loss": 0.882,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.01,
31
+ "learning_rate": 0.0002,
32
+ "loss": 0.7933,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.01,
37
+ "learning_rate": 0.0002,
38
+ "loss": 0.6232,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.02,
43
+ "learning_rate": 0.0002,
44
+ "loss": 0.9928,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.02,
49
+ "learning_rate": 0.0002,
50
+ "loss": 0.8874,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.02,
55
+ "learning_rate": 0.0002,
56
+ "loss": 0.8473,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.03,
61
+ "learning_rate": 0.0002,
62
+ "loss": 0.7204,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.03,
67
+ "learning_rate": 0.0002,
68
+ "loss": 0.5444,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.03,
73
+ "learning_rate": 0.0002,
74
+ "loss": 0.9667,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.04,
79
+ "learning_rate": 0.0002,
80
+ "loss": 0.8497,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.04,
85
+ "learning_rate": 0.0002,
86
+ "loss": 0.8284,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.04,
91
+ "learning_rate": 0.0002,
92
+ "loss": 0.6822,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.04,
97
+ "learning_rate": 0.0002,
98
+ "loss": 0.5556,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.05,
103
+ "learning_rate": 0.0002,
104
+ "loss": 0.9517,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.05,
109
+ "learning_rate": 0.0002,
110
+ "loss": 0.8587,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.05,
115
+ "learning_rate": 0.0002,
116
+ "loss": 0.8218,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.06,
121
+ "learning_rate": 0.0002,
122
+ "loss": 0.7056,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.06,
127
+ "learning_rate": 0.0002,
128
+ "loss": 0.5523,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.06,
133
+ "learning_rate": 0.0002,
134
+ "loss": 0.9718,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.07,
139
+ "learning_rate": 0.0002,
140
+ "loss": 0.8399,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 0.07,
145
+ "learning_rate": 0.0002,
146
+ "loss": 0.7947,
147
+ "step": 230
148
+ },
149
+ {
150
+ "epoch": 0.07,
151
+ "learning_rate": 0.0002,
152
+ "loss": 0.6984,
153
+ "step": 240
154
+ },
155
+ {
156
+ "epoch": 0.07,
157
+ "learning_rate": 0.0002,
158
+ "loss": 0.548,
159
+ "step": 250
160
+ },
161
+ {
162
+ "epoch": 0.08,
163
+ "learning_rate": 0.0002,
164
+ "loss": 0.9692,
165
+ "step": 260
166
+ },
167
+ {
168
+ "epoch": 0.08,
169
+ "learning_rate": 0.0002,
170
+ "loss": 0.8413,
171
+ "step": 270
172
+ },
173
+ {
174
+ "epoch": 0.08,
175
+ "learning_rate": 0.0002,
176
+ "loss": 0.816,
177
+ "step": 280
178
+ },
179
+ {
180
+ "epoch": 0.09,
181
+ "learning_rate": 0.0002,
182
+ "loss": 0.6845,
183
+ "step": 290
184
+ },
185
+ {
186
+ "epoch": 0.09,
187
+ "learning_rate": 0.0002,
188
+ "loss": 0.4823,
189
+ "step": 300
190
+ },
191
+ {
192
+ "epoch": 0.09,
193
+ "learning_rate": 0.0002,
194
+ "loss": 0.9822,
195
+ "step": 310
196
+ },
197
+ {
198
+ "epoch": 0.1,
199
+ "learning_rate": 0.0002,
200
+ "loss": 0.8399,
201
+ "step": 320
202
+ },
203
+ {
204
+ "epoch": 0.1,
205
+ "learning_rate": 0.0002,
206
+ "loss": 0.8271,
207
+ "step": 330
208
+ },
209
+ {
210
+ "epoch": 0.1,
211
+ "learning_rate": 0.0002,
212
+ "loss": 0.6497,
213
+ "step": 340
214
+ },
215
+ {
216
+ "epoch": 0.1,
217
+ "learning_rate": 0.0002,
218
+ "loss": 0.5488,
219
+ "step": 350
220
+ },
221
+ {
222
+ "epoch": 0.11,
223
+ "learning_rate": 0.0002,
224
+ "loss": 0.9632,
225
+ "step": 360
226
+ },
227
+ {
228
+ "epoch": 0.11,
229
+ "learning_rate": 0.0002,
230
+ "loss": 0.8875,
231
+ "step": 370
232
+ },
233
+ {
234
+ "epoch": 0.11,
235
+ "learning_rate": 0.0002,
236
+ "loss": 0.7813,
237
+ "step": 380
238
+ },
239
+ {
240
+ "epoch": 0.12,
241
+ "learning_rate": 0.0002,
242
+ "loss": 0.6623,
243
+ "step": 390
244
+ },
245
+ {
246
+ "epoch": 0.12,
247
+ "learning_rate": 0.0002,
248
+ "loss": 0.5352,
249
+ "step": 400
250
+ },
251
+ {
252
+ "epoch": 0.12,
253
+ "learning_rate": 0.0002,
254
+ "loss": 0.9625,
255
+ "step": 410
256
+ },
257
+ {
258
+ "epoch": 0.13,
259
+ "learning_rate": 0.0002,
260
+ "loss": 0.8437,
261
+ "step": 420
262
+ },
263
+ {
264
+ "epoch": 0.13,
265
+ "learning_rate": 0.0002,
266
+ "loss": 0.765,
267
+ "step": 430
268
+ },
269
+ {
270
+ "epoch": 0.13,
271
+ "learning_rate": 0.0002,
272
+ "loss": 0.6525,
273
+ "step": 440
274
+ },
275
+ {
276
+ "epoch": 0.13,
277
+ "learning_rate": 0.0002,
278
+ "loss": 0.5245,
279
+ "step": 450
280
+ },
281
+ {
282
+ "epoch": 0.14,
283
+ "learning_rate": 0.0002,
284
+ "loss": 0.9327,
285
+ "step": 460
286
+ },
287
+ {
288
+ "epoch": 0.14,
289
+ "learning_rate": 0.0002,
290
+ "loss": 0.7908,
291
+ "step": 470
292
+ },
293
+ {
294
+ "epoch": 0.14,
295
+ "learning_rate": 0.0002,
296
+ "loss": 0.7853,
297
+ "step": 480
298
+ },
299
+ {
300
+ "epoch": 0.15,
301
+ "learning_rate": 0.0002,
302
+ "loss": 0.6868,
303
+ "step": 490
304
+ },
305
+ {
306
+ "epoch": 0.15,
307
+ "learning_rate": 0.0002,
308
+ "loss": 0.5026,
309
+ "step": 500
310
+ }
311
+ ],
312
+ "logging_steps": 10,
313
+ "max_steps": 500,
314
+ "num_train_epochs": 1,
315
+ "save_steps": 50,
316
+ "total_flos": 4.892025651291341e+16,
317
+ "trial_name": null,
318
+ "trial_params": null
319
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6b74102a3c8f6ba6522f4dbec11494e1d6ea2e3fe07b7486f1dbc3329740f6a6
3
+ size 4027