Hugging Face's logo

ShashiVish
/
llama-7-int4-cover-letter

PEFT
TensorBoard
Model card Files Metrics Community
ShashiVish commited on
Commit
14ff18d
•
1 Parent(s): 87c787e

Upload folder using huggingface_hub

Browse files
Files changed (42) hide show
  1. README.md +21 -0
  2. adapter_config.json +21 -0
  3. adapter_model.bin +3 -0
  4. checkpoint-144/README.md +34 -0
  5. checkpoint-144/adapter_config.json +21 -0
  6. checkpoint-144/adapter_model.bin +3 -0
  7. checkpoint-144/optimizer.pt +3 -0
  8. checkpoint-144/rng_state.pth +3 -0
  9. checkpoint-144/scheduler.pt +3 -0
  10. checkpoint-144/special_tokens_map.json +24 -0
  11. checkpoint-144/tokenizer.json +0 -0
  12. checkpoint-144/tokenizer_config.json +38 -0
  13. checkpoint-144/trainer_state.json +451 -0
  14. checkpoint-144/training_args.bin +3 -0
  15. checkpoint-216/README.md +34 -0
  16. checkpoint-216/adapter_config.json +21 -0
  17. checkpoint-216/adapter_model.bin +3 -0
  18. checkpoint-216/optimizer.pt +3 -0
  19. checkpoint-216/rng_state.pth +3 -0
  20. checkpoint-216/scheduler.pt +3 -0
  21. checkpoint-216/special_tokens_map.json +24 -0
  22. checkpoint-216/tokenizer.json +0 -0
  23. checkpoint-216/tokenizer_config.json +38 -0
  24. checkpoint-216/trainer_state.json +667 -0
  25. checkpoint-216/training_args.bin +3 -0
  26. checkpoint-72/README.md +34 -0
  27. checkpoint-72/adapter_config.json +21 -0
  28. checkpoint-72/adapter_model.bin +3 -0
  29. checkpoint-72/optimizer.pt +3 -0
  30. checkpoint-72/rng_state.pth +3 -0
  31. checkpoint-72/scheduler.pt +3 -0
  32. checkpoint-72/special_tokens_map.json +24 -0
  33. checkpoint-72/tokenizer.json +0 -0
  34. checkpoint-72/tokenizer_config.json +38 -0
  35. checkpoint-72/trainer_state.json +235 -0
  36. checkpoint-72/training_args.bin +3 -0
  37. runs/Oct20_06-43-19_9f4572067104/events.out.tfevents.1697784278.9f4572067104.1069.0 +3 -0
  38. runs/Oct20_06-51-27_9f4572067104/events.out.tfevents.1697784698.9f4572067104.1069.1 +3 -0
  39. special_tokens_map.json +30 -0
  40. tokenizer.json +0 -0
  41. tokenizer_config.json +38 -0
  42. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ ### Framework versions
19
+
20
+
21
+ - PEFT 0.5.0
adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "NousResearch/Llama-2-7b-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": 64,
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 sha256:3170e0824a70c5c112e08f9a2401dddeddd935030118bdcda9940e0349dd09a7
3
+ size 134264202
checkpoint-144/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 0.5.0
33
+
34
+ - PEFT 0.5.0
checkpoint-144/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "NousResearch/Llama-2-7b-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": 64,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-144/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54a2a06bd23521c57e1eece72ed1715be0c9d021e8436a6e863f31d867c0ee0a
3
+ size 134264202
checkpoint-144/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87e24f2fba521d204120fac968a88374e1abe896454f8ac458ee588ec3c75dd4
3
+ size 268514874
checkpoint-144/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3d5b7bcfa63b582bdeec3e34f144e34978f9f461ca37f4f44a9d436ab99dab1c
3
+ size 14244
checkpoint-144/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:393139604afddfb6660b2e3c4a2918802f1119180d7aaa0f39ebae93c8bd7d80
3
+ size 1064
checkpoint-144/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
+ }
checkpoint-144/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-144/tokenizer_config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<unk>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<s>",
13
+ "lstrip": false,
14
+ "normalized": true,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ }
27
+ },
28
+ "bos_token": "<s>",
29
+ "clean_up_tokenization_spaces": false,
30
+ "eos_token": "</s>",
31
+ "legacy": false,
32
+ "model_max_length": 1000000000000000019884624838656,
33
+ "pad_token": "</s>",
34
+ "sp_model_kwargs": {},
35
+ "tokenizer_class": "LlamaTokenizer",
36
+ "unk_token": "<unk>",
37
+ "use_default_system_prompt": true
38
+ }
checkpoint-144/trainer_state.json ADDED
@@ -0,0 +1,451 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.7058823529411766,
5
+ "eval_steps": 500,
6
+ "global_step": 144,
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": 0.8355,
15
+ "step": 2
16
+ },
17
+ {
18
+ "epoch": 0.04,
19
+ "learning_rate": 0.0002,
20
+ "loss": 0.8713,
21
+ "step": 4
22
+ },
23
+ {
24
+ "epoch": 0.06,
25
+ "learning_rate": 0.0002,
26
+ "loss": 0.8682,
27
+ "step": 6
28
+ },
29
+ {
30
+ "epoch": 0.08,
31
+ "learning_rate": 0.0002,
32
+ "loss": 0.9393,
33
+ "step": 8
34
+ },
35
+ {
36
+ "epoch": 0.1,
37
+ "learning_rate": 0.0002,
38
+ "loss": 0.7962,
39
+ "step": 10
40
+ },
41
+ {
42
+ "epoch": 0.12,
43
+ "learning_rate": 0.0002,
44
+ "loss": 0.812,
45
+ "step": 12
46
+ },
47
+ {
48
+ "epoch": 0.14,
49
+ "learning_rate": 0.0002,
50
+ "loss": 0.7402,
51
+ "step": 14
52
+ },
53
+ {
54
+ "epoch": 0.16,
55
+ "learning_rate": 0.0002,
56
+ "loss": 0.8024,
57
+ "step": 16
58
+ },
59
+ {
60
+ "epoch": 0.18,
61
+ "learning_rate": 0.0002,
62
+ "loss": 0.7745,
63
+ "step": 18
64
+ },
65
+ {
66
+ "epoch": 0.2,
67
+ "learning_rate": 0.0002,
68
+ "loss": 0.7962,
69
+ "step": 20
70
+ },
71
+ {
72
+ "epoch": 0.22,
73
+ "learning_rate": 0.0002,
74
+ "loss": 0.7168,
75
+ "step": 22
76
+ },
77
+ {
78
+ "epoch": 0.24,
79
+ "learning_rate": 0.0002,
80
+ "loss": 0.7441,
81
+ "step": 24
82
+ },
83
+ {
84
+ "epoch": 0.25,
85
+ "learning_rate": 0.0002,
86
+ "loss": 0.7646,
87
+ "step": 26
88
+ },
89
+ {
90
+ "epoch": 0.27,
91
+ "learning_rate": 0.0002,
92
+ "loss": 0.76,
93
+ "step": 28
94
+ },
95
+ {
96
+ "epoch": 0.29,
97
+ "learning_rate": 0.0002,
98
+ "loss": 0.8013,
99
+ "step": 30
100
+ },
101
+ {
102
+ "epoch": 0.31,
103
+ "learning_rate": 0.0002,
104
+ "loss": 0.7374,
105
+ "step": 32
106
+ },
107
+ {
108
+ "epoch": 0.33,
109
+ "learning_rate": 0.0002,
110
+ "loss": 0.8013,
111
+ "step": 34
112
+ },
113
+ {
114
+ "epoch": 0.35,
115
+ "learning_rate": 0.0002,
116
+ "loss": 0.7418,
117
+ "step": 36
118
+ },
119
+ {
120
+ "epoch": 0.37,
121
+ "learning_rate": 0.0002,
122
+ "loss": 0.7707,
123
+ "step": 38
124
+ },
125
+ {
126
+ "epoch": 0.39,
127
+ "learning_rate": 0.0002,
128
+ "loss": 0.8057,
129
+ "step": 40
130
+ },
131
+ {
132
+ "epoch": 0.41,
133
+ "learning_rate": 0.0002,
134
+ "loss": 0.7133,
135
+ "step": 42
136
+ },
137
+ {
138
+ "epoch": 0.43,
139
+ "learning_rate": 0.0002,
140
+ "loss": 0.7545,
141
+ "step": 44
142
+ },
143
+ {
144
+ "epoch": 0.45,
145
+ "learning_rate": 0.0002,
146
+ "loss": 0.6649,
147
+ "step": 46
148
+ },
149
+ {
150
+ "epoch": 0.47,
151
+ "learning_rate": 0.0002,
152
+ "loss": 0.6521,
153
+ "step": 48
154
+ },
155
+ {
156
+ "epoch": 0.49,
157
+ "learning_rate": 0.0002,
158
+ "loss": 0.7292,
159
+ "step": 50
160
+ },
161
+ {
162
+ "epoch": 0.51,
163
+ "learning_rate": 0.0002,
164
+ "loss": 0.6916,
165
+ "step": 52
166
+ },
167
+ {
168
+ "epoch": 0.53,
169
+ "learning_rate": 0.0002,
170
+ "loss": 0.729,
171
+ "step": 54
172
+ },
173
+ {
174
+ "epoch": 0.55,
175
+ "learning_rate": 0.0002,
176
+ "loss": 0.7188,
177
+ "step": 56
178
+ },
179
+ {
180
+ "epoch": 0.57,
181
+ "learning_rate": 0.0002,
182
+ "loss": 0.8067,
183
+ "step": 58
184
+ },
185
+ {
186
+ "epoch": 0.59,
187
+ "learning_rate": 0.0002,
188
+ "loss": 0.7742,
189
+ "step": 60
190
+ },
191
+ {
192
+ "epoch": 0.61,
193
+ "learning_rate": 0.0002,
194
+ "loss": 0.6993,
195
+ "step": 62
196
+ },
197
+ {
198
+ "epoch": 0.63,
199
+ "learning_rate": 0.0002,
200
+ "loss": 0.6729,
201
+ "step": 64
202
+ },
203
+ {
204
+ "epoch": 0.65,
205
+ "learning_rate": 0.0002,
206
+ "loss": 0.7793,
207
+ "step": 66
208
+ },
209
+ {
210
+ "epoch": 0.67,
211
+ "learning_rate": 0.0002,
212
+ "loss": 0.6627,
213
+ "step": 68
214
+ },
215
+ {
216
+ "epoch": 0.69,
217
+ "learning_rate": 0.0002,
218
+ "loss": 0.7312,
219
+ "step": 70
220
+ },
221
+ {
222
+ "epoch": 0.71,
223
+ "learning_rate": 0.0002,
224
+ "loss": 0.6635,
225
+ "step": 72
226
+ },
227
+ {
228
+ "epoch": 1.02,
229
+ "learning_rate": 0.0002,
230
+ "loss": 0.6613,
231
+ "step": 74
232
+ },
233
+ {
234
+ "epoch": 1.04,
235
+ "learning_rate": 0.0002,
236
+ "loss": 0.746,
237
+ "step": 76
238
+ },
239
+ {
240
+ "epoch": 1.06,
241
+ "learning_rate": 0.0002,
242
+ "loss": 0.6225,
243
+ "step": 78
244
+ },
245
+ {
246
+ "epoch": 1.08,
247
+ "learning_rate": 0.0002,
248
+ "loss": 0.6959,
249
+ "step": 80
250
+ },
251
+ {
252
+ "epoch": 1.1,
253
+ "learning_rate": 0.0002,
254
+ "loss": 0.7332,
255
+ "step": 82
256
+ },
257
+ {
258
+ "epoch": 1.12,
259
+ "learning_rate": 0.0002,
260
+ "loss": 0.7556,
261
+ "step": 84
262
+ },
263
+ {
264
+ "epoch": 1.14,
265
+ "learning_rate": 0.0002,
266
+ "loss": 0.6855,
267
+ "step": 86
268
+ },
269
+ {
270
+ "epoch": 1.16,
271
+ "learning_rate": 0.0002,
272
+ "loss": 0.6031,
273
+ "step": 88
274
+ },
275
+ {
276
+ "epoch": 1.18,
277
+ "learning_rate": 0.0002,
278
+ "loss": 0.7095,
279
+ "step": 90
280
+ },
281
+ {
282
+ "epoch": 1.2,
283
+ "learning_rate": 0.0002,
284
+ "loss": 0.7242,
285
+ "step": 92
286
+ },
287
+ {
288
+ "epoch": 1.22,
289
+ "learning_rate": 0.0002,
290
+ "loss": 0.6949,
291
+ "step": 94
292
+ },
293
+ {
294
+ "epoch": 1.24,
295
+ "learning_rate": 0.0002,
296
+ "loss": 0.6036,
297
+ "step": 96
298
+ },
299
+ {
300
+ "epoch": 1.25,
301
+ "learning_rate": 0.0002,
302
+ "loss": 0.5898,
303
+ "step": 98
304
+ },
305
+ {
306
+ "epoch": 1.27,
307
+ "learning_rate": 0.0002,
308
+ "loss": 0.6601,
309
+ "step": 100
310
+ },
311
+ {
312
+ "epoch": 1.29,
313
+ "learning_rate": 0.0002,
314
+ "loss": 0.7339,
315
+ "step": 102
316
+ },
317
+ {
318
+ "epoch": 1.31,
319
+ "learning_rate": 0.0002,
320
+ "loss": 0.6019,
321
+ "step": 104
322
+ },
323
+ {
324
+ "epoch": 1.33,
325
+ "learning_rate": 0.0002,
326
+ "loss": 0.564,
327
+ "step": 106
328
+ },
329
+ {
330
+ "epoch": 1.35,
331
+ "learning_rate": 0.0002,
332
+ "loss": 0.6074,
333
+ "step": 108
334
+ },
335
+ {
336
+ "epoch": 1.37,
337
+ "learning_rate": 0.0002,
338
+ "loss": 0.5943,
339
+ "step": 110
340
+ },
341
+ {
342
+ "epoch": 1.39,
343
+ "learning_rate": 0.0002,
344
+ "loss": 0.6432,
345
+ "step": 112
346
+ },
347
+ {
348
+ "epoch": 1.41,
349
+ "learning_rate": 0.0002,
350
+ "loss": 0.7072,
351
+ "step": 114
352
+ },
353
+ {
354
+ "epoch": 1.43,
355
+ "learning_rate": 0.0002,
356
+ "loss": 0.6307,
357
+ "step": 116
358
+ },
359
+ {
360
+ "epoch": 1.45,
361
+ "learning_rate": 0.0002,
362
+ "loss": 0.6776,
363
+ "step": 118
364
+ },
365
+ {
366
+ "epoch": 1.47,
367
+ "learning_rate": 0.0002,
368
+ "loss": 0.6353,
369
+ "step": 120
370
+ },
371
+ {
372
+ "epoch": 1.49,
373
+ "learning_rate": 0.0002,
374
+ "loss": 0.5932,
375
+ "step": 122
376
+ },
377
+ {
378
+ "epoch": 1.51,
379
+ "learning_rate": 0.0002,
380
+ "loss": 0.5897,
381
+ "step": 124
382
+ },
383
+ {
384
+ "epoch": 1.53,
385
+ "learning_rate": 0.0002,
386
+ "loss": 0.688,
387
+ "step": 126
388
+ },
389
+ {
390
+ "epoch": 1.55,
391
+ "learning_rate": 0.0002,
392
+ "loss": 0.6442,
393
+ "step": 128
394
+ },
395
+ {
396
+ "epoch": 1.57,
397
+ "learning_rate": 0.0002,
398
+ "loss": 0.5699,
399
+ "step": 130
400
+ },
401
+ {
402
+ "epoch": 1.59,
403
+ "learning_rate": 0.0002,
404
+ "loss": 0.7366,
405
+ "step": 132
406
+ },
407
+ {
408
+ "epoch": 1.61,
409
+ "learning_rate": 0.0002,
410
+ "loss": 0.6914,
411
+ "step": 134
412
+ },
413
+ {
414
+ "epoch": 1.63,
415
+ "learning_rate": 0.0002,
416
+ "loss": 0.7067,
417
+ "step": 136
418
+ },
419
+ {
420
+ "epoch": 1.65,
421
+ "learning_rate": 0.0002,
422
+ "loss": 0.6865,
423
+ "step": 138
424
+ },
425
+ {
426
+ "epoch": 1.67,
427
+ "learning_rate": 0.0002,
428
+ "loss": 0.6918,
429
+ "step": 140
430
+ },
431
+ {
432
+ "epoch": 1.69,
433
+ "learning_rate": 0.0002,
434
+ "loss": 0.665,
435
+ "step": 142
436
+ },
437
+ {
438
+ "epoch": 1.71,
439
+ "learning_rate": 0.0002,
440
+ "loss": 0.5812,
441
+ "step": 144
442
+ }
443
+ ],
444
+ "logging_steps": 2,
445
+ "max_steps": 306,
446
+ "num_train_epochs": 3,
447
+ "save_steps": 500,
448
+ "total_flos": 2.34609646239744e+16,
449
+ "trial_name": null,
450
+ "trial_params": null
451
+ }
checkpoint-144/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33cb0f3135999cf4fd3c3a923b8477711cd34d4ba0be2ef19a722ffd92958a37
3
+ size 4536
checkpoint-216/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 0.5.0
33
+
34
+ - PEFT 0.5.0
checkpoint-216/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "NousResearch/Llama-2-7b-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": 64,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-216/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3170e0824a70c5c112e08f9a2401dddeddd935030118bdcda9940e0349dd09a7
3
+ size 134264202
checkpoint-216/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0dd22f8921c95a222b39cff9293c8019a7b9a49fb3a4afa5ad6de829826cb70
3
+ size 268514874
checkpoint-216/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:42cfe6f292512f2102d03c56aa6e87c8d03f2911e1308a79240cd80e9dee6f93
3
+ size 14244
checkpoint-216/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc51f9a8a3ca31cb31a3a29644a4e7f5e255fc732f24248b1fee305d561e9d55
3
+ size 1064
checkpoint-216/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
+ }
checkpoint-216/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-216/tokenizer_config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<unk>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<s>",
13
+ "lstrip": false,
14
+ "normalized": true,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ }
27
+ },
28
+ "bos_token": "<s>",
29
+ "clean_up_tokenization_spaces": false,
30
+ "eos_token": "</s>",
31
+ "legacy": false,
32
+ "model_max_length": 1000000000000000019884624838656,
33
+ "pad_token": "</s>",
34
+ "sp_model_kwargs": {},
35
+ "tokenizer_class": "LlamaTokenizer",
36
+ "unk_token": "<unk>",
37
+ "use_default_system_prompt": true
38
+ }
checkpoint-216/trainer_state.json ADDED
@@ -0,0 +1,667 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.7058823529411766,
5
+ "eval_steps": 500,
6
+ "global_step": 216,
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": 0.8355,
15
+ "step": 2
16
+ },
17
+ {
18
+ "epoch": 0.04,
19
+ "learning_rate": 0.0002,
20
+ "loss": 0.8713,
21
+ "step": 4
22
+ },
23
+ {
24
+ "epoch": 0.06,
25
+ "learning_rate": 0.0002,
26
+ "loss": 0.8682,
27
+ "step": 6
28
+ },
29
+ {
30
+ "epoch": 0.08,
31
+ "learning_rate": 0.0002,
32
+ "loss": 0.9393,
33
+ "step": 8
34
+ },
35
+ {
36
+ "epoch": 0.1,
37
+ "learning_rate": 0.0002,
38
+ "loss": 0.7962,
39
+ "step": 10
40
+ },
41
+ {
42
+ "epoch": 0.12,
43
+ "learning_rate": 0.0002,
44
+ "loss": 0.812,
45
+ "step": 12
46
+ },
47
+ {
48
+ "epoch": 0.14,
49
+ "learning_rate": 0.0002,
50
+ "loss": 0.7402,
51
+ "step": 14
52
+ },
53
+ {
54
+ "epoch": 0.16,
55
+ "learning_rate": 0.0002,
56
+ "loss": 0.8024,
57
+ "step": 16
58
+ },
59
+ {
60
+ "epoch": 0.18,
61
+ "learning_rate": 0.0002,
62
+ "loss": 0.7745,
63
+ "step": 18
64
+ },
65
+ {
66
+ "epoch": 0.2,
67
+ "learning_rate": 0.0002,
68
+ "loss": 0.7962,
69
+ "step": 20
70
+ },
71
+ {
72
+ "epoch": 0.22,
73
+ "learning_rate": 0.0002,
74
+ "loss": 0.7168,
75
+ "step": 22
76
+ },
77
+ {
78
+ "epoch": 0.24,
79
+ "learning_rate": 0.0002,
80
+ "loss": 0.7441,
81
+ "step": 24
82
+ },
83
+ {
84
+ "epoch": 0.25,
85
+ "learning_rate": 0.0002,
86
+ "loss": 0.7646,
87
+ "step": 26
88
+ },
89
+ {
90
+ "epoch": 0.27,
91
+ "learning_rate": 0.0002,
92
+ "loss": 0.76,
93
+ "step": 28
94
+ },
95
+ {
96
+ "epoch": 0.29,
97
+ "learning_rate": 0.0002,
98
+ "loss": 0.8013,
99
+ "step": 30
100
+ },
101
+ {
102
+ "epoch": 0.31,
103
+ "learning_rate": 0.0002,
104
+ "loss": 0.7374,
105
+ "step": 32
106
+ },
107
+ {
108
+ "epoch": 0.33,
109
+ "learning_rate": 0.0002,
110
+ "loss": 0.8013,
111
+ "step": 34
112
+ },
113
+ {
114
+ "epoch": 0.35,
115
+ "learning_rate": 0.0002,
116
+ "loss": 0.7418,
117
+ "step": 36
118
+ },
119
+ {
120
+ "epoch": 0.37,
121
+ "learning_rate": 0.0002,
122
+ "loss": 0.7707,
123
+ "step": 38
124
+ },
125
+ {
126
+ "epoch": 0.39,
127
+ "learning_rate": 0.0002,
128
+ "loss": 0.8057,
129
+ "step": 40
130
+ },
131
+ {
132
+ "epoch": 0.41,
133
+ "learning_rate": 0.0002,
134
+ "loss": 0.7133,
135
+ "step": 42
136
+ },
137
+ {
138
+ "epoch": 0.43,
139
+ "learning_rate": 0.0002,
140
+ "loss": 0.7545,
141
+ "step": 44
142
+ },
143
+ {
144
+ "epoch": 0.45,
145
+ "learning_rate": 0.0002,
146
+ "loss": 0.6649,
147
+ "step": 46
148
+ },
149
+ {
150
+ "epoch": 0.47,
151
+ "learning_rate": 0.0002,
152
+ "loss": 0.6521,
153
+ "step": 48
154
+ },
155
+ {
156
+ "epoch": 0.49,
157
+ "learning_rate": 0.0002,
158
+ "loss": 0.7292,
159
+ "step": 50
160
+ },
161
+ {
162
+ "epoch": 0.51,
163
+ "learning_rate": 0.0002,
164
+ "loss": 0.6916,
165
+ "step": 52
166
+ },
167
+ {
168
+ "epoch": 0.53,
169
+ "learning_rate": 0.0002,
170
+ "loss": 0.729,
171
+ "step": 54
172
+ },
173
+ {
174
+ "epoch": 0.55,
175
+ "learning_rate": 0.0002,
176
+ "loss": 0.7188,
177
+ "step": 56
178
+ },
179
+ {
180
+ "epoch": 0.57,
181
+ "learning_rate": 0.0002,
182
+ "loss": 0.8067,
183
+ "step": 58
184
+ },
185
+ {
186
+ "epoch": 0.59,
187
+ "learning_rate": 0.0002,
188
+ "loss": 0.7742,
189
+ "step": 60
190
+ },
191
+ {
192
+ "epoch": 0.61,
193
+ "learning_rate": 0.0002,
194
+ "loss": 0.6993,
195
+ "step": 62
196
+ },
197
+ {
198
+ "epoch": 0.63,
199
+ "learning_rate": 0.0002,
200
+ "loss": 0.6729,
201
+ "step": 64
202
+ },
203
+ {
204
+ "epoch": 0.65,
205
+ "learning_rate": 0.0002,
206
+ "loss": 0.7793,
207
+ "step": 66
208
+ },
209
+ {
210
+ "epoch": 0.67,
211
+ "learning_rate": 0.0002,
212
+ "loss": 0.6627,
213
+ "step": 68
214
+ },
215
+ {
216
+ "epoch": 0.69,
217
+ "learning_rate": 0.0002,
218
+ "loss": 0.7312,
219
+ "step": 70
220
+ },
221
+ {
222
+ "epoch": 0.71,
223
+ "learning_rate": 0.0002,
224
+ "loss": 0.6635,
225
+ "step": 72
226
+ },
227
+ {
228
+ "epoch": 1.02,
229
+ "learning_rate": 0.0002,
230
+ "loss": 0.6613,
231
+ "step": 74
232
+ },
233
+ {
234
+ "epoch": 1.04,
235
+ "learning_rate": 0.0002,
236
+ "loss": 0.746,
237
+ "step": 76
238
+ },
239
+ {
240
+ "epoch": 1.06,
241
+ "learning_rate": 0.0002,
242
+ "loss": 0.6225,
243
+ "step": 78
244
+ },
245
+ {
246
+ "epoch": 1.08,
247
+ "learning_rate": 0.0002,
248
+ "loss": 0.6959,
249
+ "step": 80
250
+ },
251
+ {
252
+ "epoch": 1.1,
253
+ "learning_rate": 0.0002,
254
+ "loss": 0.7332,
255
+ "step": 82
256
+ },
257
+ {
258
+ "epoch": 1.12,
259
+ "learning_rate": 0.0002,
260
+ "loss": 0.7556,
261
+ "step": 84
262
+ },
263
+ {
264
+ "epoch": 1.14,
265
+ "learning_rate": 0.0002,
266
+ "loss": 0.6855,
267
+ "step": 86
268
+ },
269
+ {
270
+ "epoch": 1.16,
271
+ "learning_rate": 0.0002,
272
+ "loss": 0.6031,
273
+ "step": 88
274
+ },
275
+ {
276
+ "epoch": 1.18,
277
+ "learning_rate": 0.0002,
278
+ "loss": 0.7095,
279
+ "step": 90
280
+ },
281
+ {
282
+ "epoch": 1.2,
283
+ "learning_rate": 0.0002,
284
+ "loss": 0.7242,
285
+ "step": 92
286
+ },
287
+ {
288
+ "epoch": 1.22,
289
+ "learning_rate": 0.0002,
290
+ "loss": 0.6949,
291
+ "step": 94
292
+ },
293
+ {
294
+ "epoch": 1.24,
295
+ "learning_rate": 0.0002,
296
+ "loss": 0.6036,
297
+ "step": 96
298
+ },
299
+ {
300
+ "epoch": 1.25,
301
+ "learning_rate": 0.0002,
302
+ "loss": 0.5898,
303
+ "step": 98
304
+ },
305
+ {
306
+ "epoch": 1.27,
307
+ "learning_rate": 0.0002,
308
+ "loss": 0.6601,
309
+ "step": 100
310
+ },
311
+ {
312
+ "epoch": 1.29,
313
+ "learning_rate": 0.0002,
314
+ "loss": 0.7339,
315
+ "step": 102
316
+ },
317
+ {
318
+ "epoch": 1.31,
319
+ "learning_rate": 0.0002,
320
+ "loss": 0.6019,
321
+ "step": 104
322
+ },
323
+ {
324
+ "epoch": 1.33,
325
+ "learning_rate": 0.0002,
326
+ "loss": 0.564,
327
+ "step": 106
328
+ },
329
+ {
330
+ "epoch": 1.35,
331
+ "learning_rate": 0.0002,
332
+ "loss": 0.6074,
333
+ "step": 108
334
+ },
335
+ {
336
+ "epoch": 1.37,
337
+ "learning_rate": 0.0002,
338
+ "loss": 0.5943,
339
+ "step": 110
340
+ },
341
+ {
342
+ "epoch": 1.39,
343
+ "learning_rate": 0.0002,
344
+ "loss": 0.6432,
345
+ "step": 112
346
+ },
347
+ {
348
+ "epoch": 1.41,
349
+ "learning_rate": 0.0002,
350
+ "loss": 0.7072,
351
+ "step": 114
352
+ },
353
+ {
354
+ "epoch": 1.43,
355
+ "learning_rate": 0.0002,
356
+ "loss": 0.6307,
357
+ "step": 116
358
+ },
359
+ {
360
+ "epoch": 1.45,
361
+ "learning_rate": 0.0002,
362
+ "loss": 0.6776,
363
+ "step": 118
364
+ },
365
+ {
366
+ "epoch": 1.47,
367
+ "learning_rate": 0.0002,
368
+ "loss": 0.6353,
369
+ "step": 120
370
+ },
371
+ {
372
+ "epoch": 1.49,
373
+ "learning_rate": 0.0002,
374
+ "loss": 0.5932,
375
+ "step": 122
376
+ },
377
+ {
378
+ "epoch": 1.51,
379
+ "learning_rate": 0.0002,
380
+ "loss": 0.5897,
381
+ "step": 124
382
+ },
383
+ {
384
+ "epoch": 1.53,
385
+ "learning_rate": 0.0002,
386
+ "loss": 0.688,
387
+ "step": 126
388
+ },
389
+ {
390
+ "epoch": 1.55,
391
+ "learning_rate": 0.0002,
392
+ "loss": 0.6442,
393
+ "step": 128
394
+ },
395
+ {
396
+ "epoch": 1.57,
397
+ "learning_rate": 0.0002,
398
+ "loss": 0.5699,
399
+ "step": 130
400
+ },
401
+ {
402
+ "epoch": 1.59,
403
+ "learning_rate": 0.0002,
404
+ "loss": 0.7366,
405
+ "step": 132
406
+ },
407
+ {
408
+ "epoch": 1.61,
409
+ "learning_rate": 0.0002,
410
+ "loss": 0.6914,
411
+ "step": 134
412
+ },
413
+ {
414
+ "epoch": 1.63,
415
+ "learning_rate": 0.0002,
416
+ "loss": 0.7067,
417
+ "step": 136
418
+ },
419
+ {
420
+ "epoch": 1.65,
421
+ "learning_rate": 0.0002,
422
+ "loss": 0.6865,
423
+ "step": 138
424
+ },
425
+ {
426
+ "epoch": 1.67,
427
+ "learning_rate": 0.0002,
428
+ "loss": 0.6918,
429
+ "step": 140
430
+ },
431
+ {
432
+ "epoch": 1.69,
433
+ "learning_rate": 0.0002,
434
+ "loss": 0.665,
435
+ "step": 142
436
+ },
437
+ {
438
+ "epoch": 1.71,
439
+ "learning_rate": 0.0002,
440
+ "loss": 0.5812,
441
+ "step": 144
442
+ },
443
+ {
444
+ "epoch": 2.02,
445
+ "learning_rate": 0.0002,
446
+ "loss": 0.6092,
447
+ "step": 146
448
+ },
449
+ {
450
+ "epoch": 2.04,
451
+ "learning_rate": 0.0002,
452
+ "loss": 0.6446,
453
+ "step": 148
454
+ },
455
+ {
456
+ "epoch": 2.06,
457
+ "learning_rate": 0.0002,
458
+ "loss": 0.6022,
459
+ "step": 150
460
+ },
461
+ {
462
+ "epoch": 2.08,
463
+ "learning_rate": 0.0002,
464
+ "loss": 0.6215,
465
+ "step": 152
466
+ },
467
+ {
468
+ "epoch": 2.1,
469
+ "learning_rate": 0.0002,
470
+ "loss": 0.705,
471
+ "step": 154
472
+ },
473
+ {
474
+ "epoch": 2.12,
475
+ "learning_rate": 0.0002,
476
+ "loss": 0.5592,
477
+ "step": 156
478
+ },
479
+ {
480
+ "epoch": 2.14,
481
+ "learning_rate": 0.0002,
482
+ "loss": 0.6238,
483
+ "step": 158
484
+ },
485
+ {
486
+ "epoch": 2.16,
487
+ "learning_rate": 0.0002,
488
+ "loss": 0.4967,
489
+ "step": 160
490
+ },
491
+ {
492
+ "epoch": 2.18,
493
+ "learning_rate": 0.0002,
494
+ "loss": 0.6922,
495
+ "step": 162
496
+ },
497
+ {
498
+ "epoch": 2.2,
499
+ "learning_rate": 0.0002,
500
+ "loss": 0.6024,
501
+ "step": 164
502
+ },
503
+ {
504
+ "epoch": 2.22,
505
+ "learning_rate": 0.0002,
506
+ "loss": 0.6448,
507
+ "step": 166
508
+ },
509
+ {
510
+ "epoch": 2.24,
511
+ "learning_rate": 0.0002,
512
+ "loss": 0.5405,
513
+ "step": 168
514
+ },
515
+ {
516
+ "epoch": 2.25,
517
+ "learning_rate": 0.0002,
518
+ "loss": 0.6757,
519
+ "step": 170
520
+ },
521
+ {
522
+ "epoch": 2.27,
523
+ "learning_rate": 0.0002,
524
+ "loss": 0.6052,
525
+ "step": 172
526
+ },
527
+ {
528
+ "epoch": 2.29,
529
+ "learning_rate": 0.0002,
530
+ "loss": 0.6523,
531
+ "step": 174
532
+ },
533
+ {
534
+ "epoch": 2.31,
535
+ "learning_rate": 0.0002,
536
+ "loss": 0.5521,
537
+ "step": 176
538
+ },
539
+ {
540
+ "epoch": 2.33,
541
+ "learning_rate": 0.0002,
542
+ "loss": 0.6658,
543
+ "step": 178
544
+ },
545
+ {
546
+ "epoch": 2.35,
547
+ "learning_rate": 0.0002,
548
+ "loss": 0.5512,
549
+ "step": 180
550
+ },
551
+ {
552
+ "epoch": 2.37,
553
+ "learning_rate": 0.0002,
554
+ "loss": 0.6127,
555
+ "step": 182
556
+ },
557
+ {
558
+ "epoch": 2.39,
559
+ "learning_rate": 0.0002,
560
+ "loss": 0.6525,
561
+ "step": 184
562
+ },
563
+ {
564
+ "epoch": 2.41,
565
+ "learning_rate": 0.0002,
566
+ "loss": 0.6247,
567
+ "step": 186
568
+ },
569
+ {
570
+ "epoch": 2.43,
571
+ "learning_rate": 0.0002,
572
+ "loss": 0.557,
573
+ "step": 188
574
+ },
575
+ {
576
+ "epoch": 2.45,
577
+ "learning_rate": 0.0002,
578
+ "loss": 0.6364,
579
+ "step": 190
580
+ },
581
+ {
582
+ "epoch": 2.47,
583
+ "learning_rate": 0.0002,
584
+ "loss": 0.5925,
585
+ "step": 192
586
+ },
587
+ {
588
+ "epoch": 2.49,
589
+ "learning_rate": 0.0002,
590
+ "loss": 0.6482,
591
+ "step": 194
592
+ },
593
+ {
594
+ "epoch": 2.51,
595
+ "learning_rate": 0.0002,
596
+ "loss": 0.5387,
597
+ "step": 196
598
+ },
599
+ {
600
+ "epoch": 2.53,
601
+ "learning_rate": 0.0002,
602
+ "loss": 0.6056,
603
+ "step": 198
604
+ },
605
+ {
606
+ "epoch": 2.55,
607
+ "learning_rate": 0.0002,
608
+ "loss": 0.5665,
609
+ "step": 200
610
+ },
611
+ {
612
+ "epoch": 2.57,
613
+ "learning_rate": 0.0002,
614
+ "loss": 0.6772,
615
+ "step": 202
616
+ },
617
+ {
618
+ "epoch": 2.59,
619
+ "learning_rate": 0.0002,
620
+ "loss": 0.6314,
621
+ "step": 204
622
+ },
623
+ {
624
+ "epoch": 2.61,
625
+ "learning_rate": 0.0002,
626
+ "loss": 0.5302,
627
+ "step": 206
628
+ },
629
+ {
630
+ "epoch": 2.63,
631
+ "learning_rate": 0.0002,
632
+ "loss": 0.6412,
633
+ "step": 208
634
+ },
635
+ {
636
+ "epoch": 2.65,
637
+ "learning_rate": 0.0002,
638
+ "loss": 0.5737,
639
+ "step": 210
640
+ },
641
+ {
642
+ "epoch": 2.67,
643
+ "learning_rate": 0.0002,
644
+ "loss": 0.5281,
645
+ "step": 212
646
+ },
647
+ {
648
+ "epoch": 2.69,
649
+ "learning_rate": 0.0002,
650
+ "loss": 0.7309,
651
+ "step": 214
652
+ },
653
+ {
654
+ "epoch": 2.71,
655
+ "learning_rate": 0.0002,
656
+ "loss": 0.5749,
657
+ "step": 216
658
+ }
659
+ ],
660
+ "logging_steps": 2,
661
+ "max_steps": 306,
662
+ "num_train_epochs": 3,
663
+ "save_steps": 500,
664
+ "total_flos": 3.51914469359616e+16,
665
+ "trial_name": null,
666
+ "trial_params": null
667
+ }
checkpoint-216/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33cb0f3135999cf4fd3c3a923b8477711cd34d4ba0be2ef19a722ffd92958a37
3
+ size 4536
checkpoint-72/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 0.5.0
33
+
34
+ - PEFT 0.5.0
checkpoint-72/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "NousResearch/Llama-2-7b-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": 64,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-72/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37e3d75852820a414fe4a1818b00ace38bda89e61cb0222933d0ab2ba39634ad
3
+ size 134264202
checkpoint-72/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0d9164def6dd9f378ca748da28b6f00d816eea20de4756793129b096922254a0
3
+ size 268514874
checkpoint-72/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cae4ff739f309438d1817ec739375d53ca8a5d3841e74553bd9555582dd6652
3
+ size 14244
checkpoint-72/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1f493fcf3ea59bb0222a869aea50870e4b13bbb98997cc63d0369db89f348a58
3
+ size 1064
checkpoint-72/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
+ }
checkpoint-72/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-72/tokenizer_config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<unk>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<s>",
13
+ "lstrip": false,
14
+ "normalized": true,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ }
27
+ },
28
+ "bos_token": "<s>",
29
+ "clean_up_tokenization_spaces": false,
30
+ "eos_token": "</s>",
31
+ "legacy": false,
32
+ "model_max_length": 1000000000000000019884624838656,
33
+ "pad_token": "</s>",
34
+ "sp_model_kwargs": {},
35
+ "tokenizer_class": "LlamaTokenizer",
36
+ "unk_token": "<unk>",
37
+ "use_default_system_prompt": true
38
+ }
checkpoint-72/trainer_state.json ADDED
@@ -0,0 +1,235 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.7058823529411765,
5
+ "eval_steps": 500,
6
+ "global_step": 72,
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": 0.8355,
15
+ "step": 2
16
+ },
17
+ {
18
+ "epoch": 0.04,
19
+ "learning_rate": 0.0002,
20
+ "loss": 0.8713,
21
+ "step": 4
22
+ },
23
+ {
24
+ "epoch": 0.06,
25
+ "learning_rate": 0.0002,
26
+ "loss": 0.8682,
27
+ "step": 6
28
+ },
29
+ {
30
+ "epoch": 0.08,
31
+ "learning_rate": 0.0002,
32
+ "loss": 0.9393,
33
+ "step": 8
34
+ },
35
+ {
36
+ "epoch": 0.1,
37
+ "learning_rate": 0.0002,
38
+ "loss": 0.7962,
39
+ "step": 10
40
+ },
41
+ {
42
+ "epoch": 0.12,
43
+ "learning_rate": 0.0002,
44
+ "loss": 0.812,
45
+ "step": 12
46
+ },
47
+ {
48
+ "epoch": 0.14,
49
+ "learning_rate": 0.0002,
50
+ "loss": 0.7402,
51
+ "step": 14
52
+ },
53
+ {
54
+ "epoch": 0.16,
55
+ "learning_rate": 0.0002,
56
+ "loss": 0.8024,
57
+ "step": 16
58
+ },
59
+ {
60
+ "epoch": 0.18,
61
+ "learning_rate": 0.0002,
62
+ "loss": 0.7745,
63
+ "step": 18
64
+ },
65
+ {
66
+ "epoch": 0.2,
67
+ "learning_rate": 0.0002,
68
+ "loss": 0.7962,
69
+ "step": 20
70
+ },
71
+ {
72
+ "epoch": 0.22,
73
+ "learning_rate": 0.0002,
74
+ "loss": 0.7168,
75
+ "step": 22
76
+ },
77
+ {
78
+ "epoch": 0.24,
79
+ "learning_rate": 0.0002,
80
+ "loss": 0.7441,
81
+ "step": 24
82
+ },
83
+ {
84
+ "epoch": 0.25,
85
+ "learning_rate": 0.0002,
86
+ "loss": 0.7646,
87
+ "step": 26
88
+ },
89
+ {
90
+ "epoch": 0.27,
91
+ "learning_rate": 0.0002,
92
+ "loss": 0.76,
93
+ "step": 28
94
+ },
95
+ {
96
+ "epoch": 0.29,
97
+ "learning_rate": 0.0002,
98
+ "loss": 0.8013,
99
+ "step": 30
100
+ },
101
+ {
102
+ "epoch": 0.31,
103
+ "learning_rate": 0.0002,
104
+ "loss": 0.7374,
105
+ "step": 32
106
+ },
107
+ {
108
+ "epoch": 0.33,
109
+ "learning_rate": 0.0002,
110
+ "loss": 0.8013,
111
+ "step": 34
112
+ },
113
+ {
114
+ "epoch": 0.35,
115
+ "learning_rate": 0.0002,
116
+ "loss": 0.7418,
117
+ "step": 36
118
+ },
119
+ {
120
+ "epoch": 0.37,
121
+ "learning_rate": 0.0002,
122
+ "loss": 0.7707,
123
+ "step": 38
124
+ },
125
+ {
126
+ "epoch": 0.39,
127
+ "learning_rate": 0.0002,
128
+ "loss": 0.8057,
129
+ "step": 40
130
+ },
131
+ {
132
+ "epoch": 0.41,
133
+ "learning_rate": 0.0002,
134
+ "loss": 0.7133,
135
+ "step": 42
136
+ },
137
+ {
138
+ "epoch": 0.43,
139
+ "learning_rate": 0.0002,
140
+ "loss": 0.7545,
141
+ "step": 44
142
+ },
143
+ {
144
+ "epoch": 0.45,
145
+ "learning_rate": 0.0002,
146
+ "loss": 0.6649,
147
+ "step": 46
148
+ },
149
+ {
150
+ "epoch": 0.47,
151
+ "learning_rate": 0.0002,
152
+ "loss": 0.6521,
153
+ "step": 48
154
+ },
155
+ {
156
+ "epoch": 0.49,
157
+ "learning_rate": 0.0002,
158
+ "loss": 0.7292,
159
+ "step": 50
160
+ },
161
+ {
162
+ "epoch": 0.51,
163
+ "learning_rate": 0.0002,
164
+ "loss": 0.6916,
165
+ "step": 52
166
+ },
167
+ {
168
+ "epoch": 0.53,
169
+ "learning_rate": 0.0002,
170
+ "loss": 0.729,
171
+ "step": 54
172
+ },
173
+ {
174
+ "epoch": 0.55,
175
+ "learning_rate": 0.0002,
176
+ "loss": 0.7188,
177
+ "step": 56
178
+ },
179
+ {
180
+ "epoch": 0.57,
181
+ "learning_rate": 0.0002,
182
+ "loss": 0.8067,
183
+ "step": 58
184
+ },
185
+ {
186
+ "epoch": 0.59,
187
+ "learning_rate": 0.0002,
188
+ "loss": 0.7742,
189
+ "step": 60
190
+ },
191
+ {
192
+ "epoch": 0.61,
193
+ "learning_rate": 0.0002,
194
+ "loss": 0.6993,
195
+ "step": 62
196
+ },
197
+ {
198
+ "epoch": 0.63,
199
+ "learning_rate": 0.0002,
200
+ "loss": 0.6729,
201
+ "step": 64
202
+ },
203
+ {
204
+ "epoch": 0.65,
205
+ "learning_rate": 0.0002,
206
+ "loss": 0.7793,
207
+ "step": 66
208
+ },
209
+ {
210
+ "epoch": 0.67,
211
+ "learning_rate": 0.0002,
212
+ "loss": 0.6627,
213
+ "step": 68
214
+ },
215
+ {
216
+ "epoch": 0.69,
217
+ "learning_rate": 0.0002,
218
+ "loss": 0.7312,
219
+ "step": 70
220
+ },
221
+ {
222
+ "epoch": 0.71,
223
+ "learning_rate": 0.0002,
224
+ "loss": 0.6635,
225
+ "step": 72
226
+ }
227
+ ],
228
+ "logging_steps": 2,
229
+ "max_steps": 306,
230
+ "num_train_epochs": 3,
231
+ "save_steps": 500,
232
+ "total_flos": 1.17304823119872e+16,
233
+ "trial_name": null,
234
+ "trial_params": null
235
+ }
checkpoint-72/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33cb0f3135999cf4fd3c3a923b8477711cd34d4ba0be2ef19a722ffd92958a37
3
+ size 4536
runs/Oct20_06-43-19_9f4572067104/events.out.tfevents.1697784278.9f4572067104.1069.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3b0b6cc3cb995519b569d2d852b20dca3f5a2dd7751fe5a6fd12a5277d8c8959
3
+ size 4896
runs/Oct20_06-51-27_9f4572067104/events.out.tfevents.1697784698.9f4572067104.1069.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ba945a443d35e6ba7d5956f37e249267728dbeda64fa7a0e56797c56497501c
3
+ size 21709
special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": {
17
+ "content": "</s>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<unk>",
25
+ "lstrip": false,
26
+ "normalized": true,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<unk>",
5
+ "lstrip": false,
6
+ "normalized": true,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<s>",
13
+ "lstrip": false,
14
+ "normalized": true,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": true,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ }
27
+ },
28
+ "bos_token": "<s>",
29
+ "clean_up_tokenization_spaces": false,
30
+ "eos_token": "</s>",
31
+ "legacy": false,
32
+ "model_max_length": 1000000000000000019884624838656,
33
+ "pad_token": "</s>",
34
+ "sp_model_kwargs": {},
35
+ "tokenizer_class": "LlamaTokenizer",
36
+ "unk_token": "<unk>",
37
+ "use_default_system_prompt": true
38
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:33cb0f3135999cf4fd3c3a923b8477711cd34d4ba0be2ef19a722ffd92958a37
3
+ size 4536