Hugging Face's logo

0xk1h0
/
pythia-6.9b-deduped-py150k-r20-LoRA

PEFT
Model card Files Community
1
0xk1h0 commited on
Commit
01436c9
·
1 Parent(s): 37ce8c1

Upload folder using huggingface_hub

Browse files
Files changed (11) hide show
  1. README.md +32 -1
  2. adapter_config.json +20 -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 +6 -0
  8. tokenizer.json +0 -0
  9. tokenizer_config.json +9 -0
  10. trainer_state.json +319 -0
  11. training_args.bin +3 -0
README.md CHANGED
@@ -1,3 +1,34 @@
1
  ---
2
- license: mit
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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: True
10
+ - load_in_4bit: False
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: fp4
16
+ - bnb_4bit_use_double_quant: False
17
+ - bnb_4bit_compute_dtype: float32
18
+
19
+ The following `bitsandbytes` quantization config was used during training:
20
+ - quant_method: bitsandbytes
21
+ - load_in_8bit: True
22
+ - load_in_4bit: False
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: fp4
28
+ - bnb_4bit_use_double_quant: False
29
+ - bnb_4bit_compute_dtype: float32
30
+ ### Framework versions
31
+
32
+ - PEFT 0.5.0
33
+
34
+ - PEFT 0.5.0
adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "EleutherAI/pythia-6.9b-deduped",
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": 32,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "query_key_value"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c08c71d7aeaca1f591acb580282ac95f1256354ca44124199afd6af5a17632bc
3
+ size 16800753
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a1c859bb000d3b9445567d47f5af81d32bd4a4eabdd7421916646b9317c99be
3
+ size 883
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7a02d96ae40550845789ce32d0c457ee664c6a248c82f22c5c5d6072c37059fc
3
+ size 14575
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa9ee96276bca7f1100f76a621751f7c1da8c741871fef4e85c55116d1e8fcb6
3
+ size 627
special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "pad_token": "<|endoftext|>",
5
+ "unk_token": "<|endoftext|>"
6
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": "<|endoftext|>",
4
+ "clean_up_tokenization_spaces": true,
5
+ "eos_token": "<|endoftext|>",
6
+ "model_max_length": 1000000000000000019884624838656,
7
+ "tokenizer_class": "GPTNeoXTokenizer",
8
+ "unk_token": "<|endoftext|>"
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,319 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0660980810234542,
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": 0.02,
13
+ "learning_rate": 0.0002,
14
+ "loss": 0.0,
15
+ "step": 10
16
+ },
17
+ {
18
+ "epoch": 0.04,
19
+ "learning_rate": 0.0002,
20
+ "loss": 0.0,
21
+ "step": 20
22
+ },
23
+ {
24
+ "epoch": 0.06,
25
+ "learning_rate": 0.0002,
26
+ "loss": 0.0,
27
+ "step": 30
28
+ },
29
+ {
30
+ "epoch": 0.09,
31
+ "learning_rate": 0.0002,
32
+ "loss": 0.0,
33
+ "step": 40
34
+ },
35
+ {
36
+ "epoch": 0.11,
37
+ "learning_rate": 0.0002,
38
+ "loss": 0.0,
39
+ "step": 50
40
+ },
41
+ {
42
+ "epoch": 0.13,
43
+ "learning_rate": 0.0002,
44
+ "loss": 0.0,
45
+ "step": 60
46
+ },
47
+ {
48
+ "epoch": 0.15,
49
+ "learning_rate": 0.0002,
50
+ "loss": 0.0,
51
+ "step": 70
52
+ },
53
+ {
54
+ "epoch": 0.17,
55
+ "learning_rate": 0.0002,
56
+ "loss": 0.0,
57
+ "step": 80
58
+ },
59
+ {
60
+ "epoch": 0.19,
61
+ "learning_rate": 0.0002,
62
+ "loss": 0.0,
63
+ "step": 90
64
+ },
65
+ {
66
+ "epoch": 0.21,
67
+ "learning_rate": 0.0002,
68
+ "loss": 0.0,
69
+ "step": 100
70
+ },
71
+ {
72
+ "epoch": 0.23,
73
+ "learning_rate": 0.0002,
74
+ "loss": 0.0,
75
+ "step": 110
76
+ },
77
+ {
78
+ "epoch": 0.26,
79
+ "learning_rate": 0.0002,
80
+ "loss": 0.0,
81
+ "step": 120
82
+ },
83
+ {
84
+ "epoch": 0.28,
85
+ "learning_rate": 0.0002,
86
+ "loss": 0.0,
87
+ "step": 130
88
+ },
89
+ {
90
+ "epoch": 0.3,
91
+ "learning_rate": 0.0002,
92
+ "loss": 0.0,
93
+ "step": 140
94
+ },
95
+ {
96
+ "epoch": 0.32,
97
+ "learning_rate": 0.0002,
98
+ "loss": 0.0,
99
+ "step": 150
100
+ },
101
+ {
102
+ "epoch": 0.34,
103
+ "learning_rate": 0.0002,
104
+ "loss": 0.0,
105
+ "step": 160
106
+ },
107
+ {
108
+ "epoch": 0.36,
109
+ "learning_rate": 0.0002,
110
+ "loss": 0.0,
111
+ "step": 170
112
+ },
113
+ {
114
+ "epoch": 0.38,
115
+ "learning_rate": 0.0002,
116
+ "loss": 0.0,
117
+ "step": 180
118
+ },
119
+ {
120
+ "epoch": 0.41,
121
+ "learning_rate": 0.0002,
122
+ "loss": 0.0,
123
+ "step": 190
124
+ },
125
+ {
126
+ "epoch": 0.43,
127
+ "learning_rate": 0.0002,
128
+ "loss": 0.0,
129
+ "step": 200
130
+ },
131
+ {
132
+ "epoch": 0.45,
133
+ "learning_rate": 0.0002,
134
+ "loss": 0.0,
135
+ "step": 210
136
+ },
137
+ {
138
+ "epoch": 0.47,
139
+ "learning_rate": 0.0002,
140
+ "loss": 0.0,
141
+ "step": 220
142
+ },
143
+ {
144
+ "epoch": 0.49,
145
+ "learning_rate": 0.0002,
146
+ "loss": 0.0,
147
+ "step": 230
148
+ },
149
+ {
150
+ "epoch": 0.51,
151
+ "learning_rate": 0.0002,
152
+ "loss": 0.0,
153
+ "step": 240
154
+ },
155
+ {
156
+ "epoch": 0.53,
157
+ "learning_rate": 0.0002,
158
+ "loss": 0.0,
159
+ "step": 250
160
+ },
161
+ {
162
+ "epoch": 0.55,
163
+ "learning_rate": 0.0002,
164
+ "loss": 0.0,
165
+ "step": 260
166
+ },
167
+ {
168
+ "epoch": 0.58,
169
+ "learning_rate": 0.0002,
170
+ "loss": 0.0,
171
+ "step": 270
172
+ },
173
+ {
174
+ "epoch": 0.6,
175
+ "learning_rate": 0.0002,
176
+ "loss": 0.0,
177
+ "step": 280
178
+ },
179
+ {
180
+ "epoch": 0.62,
181
+ "learning_rate": 0.0002,
182
+ "loss": 0.0,
183
+ "step": 290
184
+ },
185
+ {
186
+ "epoch": 0.64,
187
+ "learning_rate": 0.0002,
188
+ "loss": 0.0,
189
+ "step": 300
190
+ },
191
+ {
192
+ "epoch": 0.66,
193
+ "learning_rate": 0.0002,
194
+ "loss": 0.0,
195
+ "step": 310
196
+ },
197
+ {
198
+ "epoch": 0.68,
199
+ "learning_rate": 0.0002,
200
+ "loss": 0.0,
201
+ "step": 320
202
+ },
203
+ {
204
+ "epoch": 0.7,
205
+ "learning_rate": 0.0002,
206
+ "loss": 0.0,
207
+ "step": 330
208
+ },
209
+ {
210
+ "epoch": 0.72,
211
+ "learning_rate": 0.0002,
212
+ "loss": 0.0,
213
+ "step": 340
214
+ },
215
+ {
216
+ "epoch": 0.75,
217
+ "learning_rate": 0.0002,
218
+ "loss": 0.0,
219
+ "step": 350
220
+ },
221
+ {
222
+ "epoch": 0.77,
223
+ "learning_rate": 0.0002,
224
+ "loss": 0.0,
225
+ "step": 360
226
+ },
227
+ {
228
+ "epoch": 0.79,
229
+ "learning_rate": 0.0002,
230
+ "loss": 0.0,
231
+ "step": 370
232
+ },
233
+ {
234
+ "epoch": 0.81,
235
+ "learning_rate": 0.0002,
236
+ "loss": 0.0,
237
+ "step": 380
238
+ },
239
+ {
240
+ "epoch": 0.83,
241
+ "learning_rate": 0.0002,
242
+ "loss": 0.0,
243
+ "step": 390
244
+ },
245
+ {
246
+ "epoch": 0.85,
247
+ "learning_rate": 0.0002,
248
+ "loss": 0.0,
249
+ "step": 400
250
+ },
251
+ {
252
+ "epoch": 0.87,
253
+ "learning_rate": 0.0002,
254
+ "loss": 0.0,
255
+ "step": 410
256
+ },
257
+ {
258
+ "epoch": 0.9,
259
+ "learning_rate": 0.0002,
260
+ "loss": 0.0,
261
+ "step": 420
262
+ },
263
+ {
264
+ "epoch": 0.92,
265
+ "learning_rate": 0.0002,
266
+ "loss": 0.0,
267
+ "step": 430
268
+ },
269
+ {
270
+ "epoch": 0.94,
271
+ "learning_rate": 0.0002,
272
+ "loss": 0.0,
273
+ "step": 440
274
+ },
275
+ {
276
+ "epoch": 0.96,
277
+ "learning_rate": 0.0002,
278
+ "loss": 0.0,
279
+ "step": 450
280
+ },
281
+ {
282
+ "epoch": 0.98,
283
+ "learning_rate": 0.0002,
284
+ "loss": 0.0,
285
+ "step": 460
286
+ },
287
+ {
288
+ "epoch": 1.0,
289
+ "learning_rate": 0.0002,
290
+ "loss": 0.0,
291
+ "step": 470
292
+ },
293
+ {
294
+ "epoch": 1.02,
295
+ "learning_rate": 0.0002,
296
+ "loss": 0.0,
297
+ "step": 480
298
+ },
299
+ {
300
+ "epoch": 1.04,
301
+ "learning_rate": 0.0002,
302
+ "loss": 0.0,
303
+ "step": 490
304
+ },
305
+ {
306
+ "epoch": 1.07,
307
+ "learning_rate": 0.0002,
308
+ "loss": 0.0,
309
+ "step": 500
310
+ }
311
+ ],
312
+ "logging_steps": 10,
313
+ "max_steps": 500,
314
+ "num_train_epochs": 2,
315
+ "save_steps": 250,
316
+ "total_flos": 6.534086153899868e+17,
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:b6f305f6fbd087dbd5098e603278b8f9a6af8ddb8df6edf91ac185e9c1670d19
3
+ size 4091