Hugging Face's logo

sallywww
/
tot_llama_update

PEFT
Safetensors
Model card Files Community
sallywww commited on
Commit
84733df
•
1 Parent(s): b280d5c

addd

Browse files

add stuff

Files changed (27) hide show
  1. README.md +24 -0
  2. adapter_config.json +21 -0
  3. adapter_model.bin +3 -0
  4. checkpoint-1000/README.md +24 -0
  5. checkpoint-1000/adapter_config.json +21 -0
  6. checkpoint-1000/adapter_model.safetensors +3 -0
  7. checkpoint-1000/optimizer.pt +3 -0
  8. checkpoint-1000/rng_state.pth +3 -0
  9. checkpoint-1000/scheduler.pt +3 -0
  10. checkpoint-1000/trainer_state.json +371 -0
  11. checkpoint-1000/training_args.bin +3 -0
  12. checkpoint-1500/README.md +24 -0
  13. checkpoint-1500/adapter_config.json +21 -0
  14. checkpoint-1500/adapter_model.safetensors +3 -0
  15. checkpoint-1500/optimizer.pt +3 -0
  16. checkpoint-1500/rng_state.pth +3 -0
  17. checkpoint-1500/scheduler.pt +3 -0
  18. checkpoint-1500/trainer_state.json +546 -0
  19. checkpoint-1500/training_args.bin +3 -0
  20. checkpoint-500/README.md +24 -0
  21. checkpoint-500/adapter_config.json +21 -0
  22. checkpoint-500/adapter_model.safetensors +3 -0
  23. checkpoint-500/optimizer.pt +3 -0
  24. checkpoint-500/rng_state.pth +3 -0
  25. checkpoint-500/scheduler.pt +3 -0
  26. checkpoint-500/trainer_state.json +196 -0
  27. checkpoint-500/training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ - bnb_4bit_quant_storage: uint8
19
+ - load_in_4bit: False
20
+ - load_in_8bit: True
21
+ ### Framework versions
22
+
23
+
24
+ - PEFT 0.5.0
adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "sallywww/Llama-7B",
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
+ "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:f43e3c1d22d40de08a686d8f2eca862fe32522c110314f0996cbad068441d64f
3
+ size 16823434
checkpoint-1000/README.md ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ - bnb_4bit_quant_storage: uint8
19
+ - load_in_4bit: False
20
+ - load_in_8bit: True
21
+ ### Framework versions
22
+
23
+
24
+ - PEFT 0.5.0
checkpoint-1000/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "sallywww/Llama-7B",
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
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-1000/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e42b50769ce7113b870c21414ac485b2cfcb3290d07d8405d46df3155ce1c000
3
+ size 16794200
checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6540ebbdfc9b7e36d3f8f6bc9999bcd611d07c11f36de333c418e5ec32ac0626
3
+ size 33662074
checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:45776e70f7048c8554b90ed5d716cb0cad5954e1b5cc4952225b3f6075d24ef3
3
+ size 14244
checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ed2f32f660a723773ce8e9c92db5c6aff6e2a63e18d1c2dd5d1f984fb98d633
3
+ size 1064
checkpoint-1000/trainer_state.json ADDED
@@ -0,0 +1,371 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 15.57935735150925,
5
+ "eval_steps": 500,
6
+ "global_step": 1000,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.31,
13
+ "grad_norm": 0.48931118845939636,
14
+ "learning_rate": 1.98125e-05,
15
+ "loss": 1.7417,
16
+ "step": 20
17
+ },
18
+ {
19
+ "epoch": 0.62,
20
+ "grad_norm": 0.56327223777771,
21
+ "learning_rate": 1.9604166666666668e-05,
22
+ "loss": 1.7626,
23
+ "step": 40
24
+ },
25
+ {
26
+ "epoch": 0.93,
27
+ "grad_norm": 0.6054126620292664,
28
+ "learning_rate": 1.9395833333333335e-05,
29
+ "loss": 1.5774,
30
+ "step": 60
31
+ },
32
+ {
33
+ "epoch": 1.25,
34
+ "grad_norm": 0.8035449385643005,
35
+ "learning_rate": 1.9187500000000002e-05,
36
+ "loss": 1.4772,
37
+ "step": 80
38
+ },
39
+ {
40
+ "epoch": 1.56,
41
+ "grad_norm": 0.9972027540206909,
42
+ "learning_rate": 1.897916666666667e-05,
43
+ "loss": 1.409,
44
+ "step": 100
45
+ },
46
+ {
47
+ "epoch": 1.87,
48
+ "grad_norm": 0.7526600360870361,
49
+ "learning_rate": 1.8781250000000003e-05,
50
+ "loss": 1.3088,
51
+ "step": 120
52
+ },
53
+ {
54
+ "epoch": 2.18,
55
+ "grad_norm": 0.8370587825775146,
56
+ "learning_rate": 1.8572916666666666e-05,
57
+ "loss": 1.2086,
58
+ "step": 140
59
+ },
60
+ {
61
+ "epoch": 2.49,
62
+ "grad_norm": 0.8337924480438232,
63
+ "learning_rate": 1.8364583333333334e-05,
64
+ "loss": 1.1321,
65
+ "step": 160
66
+ },
67
+ {
68
+ "epoch": 2.8,
69
+ "grad_norm": 1.0028570890426636,
70
+ "learning_rate": 1.815625e-05,
71
+ "loss": 1.0439,
72
+ "step": 180
73
+ },
74
+ {
75
+ "epoch": 3.12,
76
+ "grad_norm": 0.7663702368736267,
77
+ "learning_rate": 1.7947916666666668e-05,
78
+ "loss": 1.026,
79
+ "step": 200
80
+ },
81
+ {
82
+ "epoch": 3.43,
83
+ "grad_norm": 0.8835870623588562,
84
+ "learning_rate": 1.7739583333333335e-05,
85
+ "loss": 0.9994,
86
+ "step": 220
87
+ },
88
+ {
89
+ "epoch": 3.74,
90
+ "grad_norm": 1.010624885559082,
91
+ "learning_rate": 1.7531250000000003e-05,
92
+ "loss": 0.9381,
93
+ "step": 240
94
+ },
95
+ {
96
+ "epoch": 4.05,
97
+ "grad_norm": 0.9645354151725769,
98
+ "learning_rate": 1.7322916666666666e-05,
99
+ "loss": 0.8667,
100
+ "step": 260
101
+ },
102
+ {
103
+ "epoch": 4.36,
104
+ "grad_norm": 0.9926008582115173,
105
+ "learning_rate": 1.7114583333333334e-05,
106
+ "loss": 0.8729,
107
+ "step": 280
108
+ },
109
+ {
110
+ "epoch": 4.67,
111
+ "grad_norm": 1.0364482402801514,
112
+ "learning_rate": 1.690625e-05,
113
+ "loss": 0.836,
114
+ "step": 300
115
+ },
116
+ {
117
+ "epoch": 4.99,
118
+ "grad_norm": 1.3519001007080078,
119
+ "learning_rate": 1.6697916666666668e-05,
120
+ "loss": 0.8126,
121
+ "step": 320
122
+ },
123
+ {
124
+ "epoch": 5.3,
125
+ "grad_norm": 1.2280508279800415,
126
+ "learning_rate": 1.6489583333333335e-05,
127
+ "loss": 0.7953,
128
+ "step": 340
129
+ },
130
+ {
131
+ "epoch": 5.61,
132
+ "grad_norm": 0.8931779265403748,
133
+ "learning_rate": 1.6281250000000003e-05,
134
+ "loss": 0.7813,
135
+ "step": 360
136
+ },
137
+ {
138
+ "epoch": 5.92,
139
+ "grad_norm": 2.7053027153015137,
140
+ "learning_rate": 1.6072916666666667e-05,
141
+ "loss": 0.7256,
142
+ "step": 380
143
+ },
144
+ {
145
+ "epoch": 6.23,
146
+ "grad_norm": 1.411024808883667,
147
+ "learning_rate": 1.5864583333333334e-05,
148
+ "loss": 0.7336,
149
+ "step": 400
150
+ },
151
+ {
152
+ "epoch": 6.54,
153
+ "grad_norm": 1.0111807584762573,
154
+ "learning_rate": 1.565625e-05,
155
+ "loss": 0.7207,
156
+ "step": 420
157
+ },
158
+ {
159
+ "epoch": 6.85,
160
+ "grad_norm": 2.213623523712158,
161
+ "learning_rate": 1.544791666666667e-05,
162
+ "loss": 0.664,
163
+ "step": 440
164
+ },
165
+ {
166
+ "epoch": 7.17,
167
+ "grad_norm": 1.3642323017120361,
168
+ "learning_rate": 1.5239583333333334e-05,
169
+ "loss": 0.6848,
170
+ "step": 460
171
+ },
172
+ {
173
+ "epoch": 7.48,
174
+ "grad_norm": 1.3692028522491455,
175
+ "learning_rate": 1.5031250000000001e-05,
176
+ "loss": 0.6663,
177
+ "step": 480
178
+ },
179
+ {
180
+ "epoch": 7.79,
181
+ "grad_norm": 1.9850131273269653,
182
+ "learning_rate": 1.4822916666666667e-05,
183
+ "loss": 0.6199,
184
+ "step": 500
185
+ },
186
+ {
187
+ "epoch": 8.1,
188
+ "grad_norm": 1.6070563793182373,
189
+ "learning_rate": 1.4614583333333334e-05,
190
+ "loss": 0.6295,
191
+ "step": 520
192
+ },
193
+ {
194
+ "epoch": 8.41,
195
+ "grad_norm": 3.226116418838501,
196
+ "learning_rate": 1.4406250000000001e-05,
197
+ "loss": 0.6227,
198
+ "step": 540
199
+ },
200
+ {
201
+ "epoch": 8.72,
202
+ "grad_norm": 1.5464348793029785,
203
+ "learning_rate": 1.4197916666666667e-05,
204
+ "loss": 0.589,
205
+ "step": 560
206
+ },
207
+ {
208
+ "epoch": 9.04,
209
+ "grad_norm": 1.4290672540664673,
210
+ "learning_rate": 1.3989583333333334e-05,
211
+ "loss": 0.6263,
212
+ "step": 580
213
+ },
214
+ {
215
+ "epoch": 9.35,
216
+ "grad_norm": 1.682243824005127,
217
+ "learning_rate": 1.3781250000000001e-05,
218
+ "loss": 0.5747,
219
+ "step": 600
220
+ },
221
+ {
222
+ "epoch": 9.66,
223
+ "grad_norm": 1.6785274744033813,
224
+ "learning_rate": 1.3572916666666667e-05,
225
+ "loss": 0.5809,
226
+ "step": 620
227
+ },
228
+ {
229
+ "epoch": 9.97,
230
+ "grad_norm": 2.156558036804199,
231
+ "learning_rate": 1.3364583333333334e-05,
232
+ "loss": 0.5725,
233
+ "step": 640
234
+ },
235
+ {
236
+ "epoch": 10.28,
237
+ "grad_norm": 1.739134430885315,
238
+ "learning_rate": 1.3156250000000001e-05,
239
+ "loss": 0.549,
240
+ "step": 660
241
+ },
242
+ {
243
+ "epoch": 10.59,
244
+ "grad_norm": 1.2140896320343018,
245
+ "learning_rate": 1.2947916666666667e-05,
246
+ "loss": 0.5493,
247
+ "step": 680
248
+ },
249
+ {
250
+ "epoch": 10.91,
251
+ "grad_norm": 1.517707109451294,
252
+ "learning_rate": 1.2739583333333334e-05,
253
+ "loss": 0.5589,
254
+ "step": 700
255
+ },
256
+ {
257
+ "epoch": 11.22,
258
+ "grad_norm": 1.8752249479293823,
259
+ "learning_rate": 1.2531250000000001e-05,
260
+ "loss": 0.5001,
261
+ "step": 720
262
+ },
263
+ {
264
+ "epoch": 11.53,
265
+ "grad_norm": 1.8684614896774292,
266
+ "learning_rate": 1.2322916666666667e-05,
267
+ "loss": 0.52,
268
+ "step": 740
269
+ },
270
+ {
271
+ "epoch": 11.84,
272
+ "grad_norm": 2.3552052974700928,
273
+ "learning_rate": 1.2114583333333334e-05,
274
+ "loss": 0.5479,
275
+ "step": 760
276
+ },
277
+ {
278
+ "epoch": 12.15,
279
+ "grad_norm": 1.2074153423309326,
280
+ "learning_rate": 1.1906250000000001e-05,
281
+ "loss": 0.4977,
282
+ "step": 780
283
+ },
284
+ {
285
+ "epoch": 12.46,
286
+ "grad_norm": 3.0007381439208984,
287
+ "learning_rate": 1.1697916666666667e-05,
288
+ "loss": 0.5252,
289
+ "step": 800
290
+ },
291
+ {
292
+ "epoch": 12.78,
293
+ "grad_norm": 1.8956348896026611,
294
+ "learning_rate": 1.1489583333333334e-05,
295
+ "loss": 0.503,
296
+ "step": 820
297
+ },
298
+ {
299
+ "epoch": 13.09,
300
+ "grad_norm": 1.3779544830322266,
301
+ "learning_rate": 1.1281250000000001e-05,
302
+ "loss": 0.4991,
303
+ "step": 840
304
+ },
305
+ {
306
+ "epoch": 13.4,
307
+ "grad_norm": 1.5404250621795654,
308
+ "learning_rate": 1.1072916666666667e-05,
309
+ "loss": 0.5026,
310
+ "step": 860
311
+ },
312
+ {
313
+ "epoch": 13.71,
314
+ "grad_norm": 2.149167060852051,
315
+ "learning_rate": 1.0864583333333334e-05,
316
+ "loss": 0.4788,
317
+ "step": 880
318
+ },
319
+ {
320
+ "epoch": 14.02,
321
+ "grad_norm": 1.4978464841842651,
322
+ "learning_rate": 1.0656250000000002e-05,
323
+ "loss": 0.4605,
324
+ "step": 900
325
+ },
326
+ {
327
+ "epoch": 14.33,
328
+ "grad_norm": 1.5203664302825928,
329
+ "learning_rate": 1.0447916666666667e-05,
330
+ "loss": 0.4858,
331
+ "step": 920
332
+ },
333
+ {
334
+ "epoch": 14.64,
335
+ "grad_norm": 1.850074291229248,
336
+ "learning_rate": 1.0239583333333334e-05,
337
+ "loss": 0.4797,
338
+ "step": 940
339
+ },
340
+ {
341
+ "epoch": 14.96,
342
+ "grad_norm": 1.7591063976287842,
343
+ "learning_rate": 1.0031250000000002e-05,
344
+ "loss": 0.4619,
345
+ "step": 960
346
+ },
347
+ {
348
+ "epoch": 15.27,
349
+ "grad_norm": 1.4994142055511475,
350
+ "learning_rate": 9.822916666666667e-06,
351
+ "loss": 0.4627,
352
+ "step": 980
353
+ },
354
+ {
355
+ "epoch": 15.58,
356
+ "grad_norm": 1.888311743736267,
357
+ "learning_rate": 9.614583333333334e-06,
358
+ "loss": 0.4469,
359
+ "step": 1000
360
+ }
361
+ ],
362
+ "logging_steps": 20,
363
+ "max_steps": 1920,
364
+ "num_input_tokens_seen": 0,
365
+ "num_train_epochs": 30,
366
+ "save_steps": 500,
367
+ "total_flos": 2.5991277871104e+18,
368
+ "train_batch_size": 1,
369
+ "trial_name": null,
370
+ "trial_params": null
371
+ }
checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf3bd8d0ae9745c03ca60e38c25c6801ef1b8e53b972f64cac4e257777c6a59b
3
+ size 4920
checkpoint-1500/README.md ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ - bnb_4bit_quant_storage: uint8
19
+ - load_in_4bit: False
20
+ - load_in_8bit: True
21
+ ### Framework versions
22
+
23
+
24
+ - PEFT 0.5.0
checkpoint-1500/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "sallywww/Llama-7B",
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
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-1500/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1d5214a96cc502424822c75c533933a4fb83728a30f80c1cefe6eacbdd7d072c
3
+ size 16794200
checkpoint-1500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:110271a4c7328208e3d46bb8c27b9463355919c67cf51105f81ff64912036124
3
+ size 33662074
checkpoint-1500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e132493465221d0b51d395548812c95d19fe43496e3c420d7155a0d54496a203
3
+ size 14244
checkpoint-1500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:96348156a623838a1fbfad1b35cb88b84848be403effa3bb42455999f713f36e
3
+ size 1064
checkpoint-1500/trainer_state.json ADDED
@@ -0,0 +1,546 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 23.369036027263874,
5
+ "eval_steps": 500,
6
+ "global_step": 1500,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.31,
13
+ "grad_norm": 0.48931118845939636,
14
+ "learning_rate": 1.98125e-05,
15
+ "loss": 1.7417,
16
+ "step": 20
17
+ },
18
+ {
19
+ "epoch": 0.62,
20
+ "grad_norm": 0.56327223777771,
21
+ "learning_rate": 1.9604166666666668e-05,
22
+ "loss": 1.7626,
23
+ "step": 40
24
+ },
25
+ {
26
+ "epoch": 0.93,
27
+ "grad_norm": 0.6054126620292664,
28
+ "learning_rate": 1.9395833333333335e-05,
29
+ "loss": 1.5774,
30
+ "step": 60
31
+ },
32
+ {
33
+ "epoch": 1.25,
34
+ "grad_norm": 0.8035449385643005,
35
+ "learning_rate": 1.9187500000000002e-05,
36
+ "loss": 1.4772,
37
+ "step": 80
38
+ },
39
+ {
40
+ "epoch": 1.56,
41
+ "grad_norm": 0.9972027540206909,
42
+ "learning_rate": 1.897916666666667e-05,
43
+ "loss": 1.409,
44
+ "step": 100
45
+ },
46
+ {
47
+ "epoch": 1.87,
48
+ "grad_norm": 0.7526600360870361,
49
+ "learning_rate": 1.8781250000000003e-05,
50
+ "loss": 1.3088,
51
+ "step": 120
52
+ },
53
+ {
54
+ "epoch": 2.18,
55
+ "grad_norm": 0.8370587825775146,
56
+ "learning_rate": 1.8572916666666666e-05,
57
+ "loss": 1.2086,
58
+ "step": 140
59
+ },
60
+ {
61
+ "epoch": 2.49,
62
+ "grad_norm": 0.8337924480438232,
63
+ "learning_rate": 1.8364583333333334e-05,
64
+ "loss": 1.1321,
65
+ "step": 160
66
+ },
67
+ {
68
+ "epoch": 2.8,
69
+ "grad_norm": 1.0028570890426636,
70
+ "learning_rate": 1.815625e-05,
71
+ "loss": 1.0439,
72
+ "step": 180
73
+ },
74
+ {
75
+ "epoch": 3.12,
76
+ "grad_norm": 0.7663702368736267,
77
+ "learning_rate": 1.7947916666666668e-05,
78
+ "loss": 1.026,
79
+ "step": 200
80
+ },
81
+ {
82
+ "epoch": 3.43,
83
+ "grad_norm": 0.8835870623588562,
84
+ "learning_rate": 1.7739583333333335e-05,
85
+ "loss": 0.9994,
86
+ "step": 220
87
+ },
88
+ {
89
+ "epoch": 3.74,
90
+ "grad_norm": 1.010624885559082,
91
+ "learning_rate": 1.7531250000000003e-05,
92
+ "loss": 0.9381,
93
+ "step": 240
94
+ },
95
+ {
96
+ "epoch": 4.05,
97
+ "grad_norm": 0.9645354151725769,
98
+ "learning_rate": 1.7322916666666666e-05,
99
+ "loss": 0.8667,
100
+ "step": 260
101
+ },
102
+ {
103
+ "epoch": 4.36,
104
+ "grad_norm": 0.9926008582115173,
105
+ "learning_rate": 1.7114583333333334e-05,
106
+ "loss": 0.8729,
107
+ "step": 280
108
+ },
109
+ {
110
+ "epoch": 4.67,
111
+ "grad_norm": 1.0364482402801514,
112
+ "learning_rate": 1.690625e-05,
113
+ "loss": 0.836,
114
+ "step": 300
115
+ },
116
+ {
117
+ "epoch": 4.99,
118
+ "grad_norm": 1.3519001007080078,
119
+ "learning_rate": 1.6697916666666668e-05,
120
+ "loss": 0.8126,
121
+ "step": 320
122
+ },
123
+ {
124
+ "epoch": 5.3,
125
+ "grad_norm": 1.2280508279800415,
126
+ "learning_rate": 1.6489583333333335e-05,
127
+ "loss": 0.7953,
128
+ "step": 340
129
+ },
130
+ {
131
+ "epoch": 5.61,
132
+ "grad_norm": 0.8931779265403748,
133
+ "learning_rate": 1.6281250000000003e-05,
134
+ "loss": 0.7813,
135
+ "step": 360
136
+ },
137
+ {
138
+ "epoch": 5.92,
139
+ "grad_norm": 2.7053027153015137,
140
+ "learning_rate": 1.6072916666666667e-05,
141
+ "loss": 0.7256,
142
+ "step": 380
143
+ },
144
+ {
145
+ "epoch": 6.23,
146
+ "grad_norm": 1.411024808883667,
147
+ "learning_rate": 1.5864583333333334e-05,
148
+ "loss": 0.7336,
149
+ "step": 400
150
+ },
151
+ {
152
+ "epoch": 6.54,
153
+ "grad_norm": 1.0111807584762573,
154
+ "learning_rate": 1.565625e-05,
155
+ "loss": 0.7207,
156
+ "step": 420
157
+ },
158
+ {
159
+ "epoch": 6.85,
160
+ "grad_norm": 2.213623523712158,
161
+ "learning_rate": 1.544791666666667e-05,
162
+ "loss": 0.664,
163
+ "step": 440
164
+ },
165
+ {
166
+ "epoch": 7.17,
167
+ "grad_norm": 1.3642323017120361,
168
+ "learning_rate": 1.5239583333333334e-05,
169
+ "loss": 0.6848,
170
+ "step": 460
171
+ },
172
+ {
173
+ "epoch": 7.48,
174
+ "grad_norm": 1.3692028522491455,
175
+ "learning_rate": 1.5031250000000001e-05,
176
+ "loss": 0.6663,
177
+ "step": 480
178
+ },
179
+ {
180
+ "epoch": 7.79,
181
+ "grad_norm": 1.9850131273269653,
182
+ "learning_rate": 1.4822916666666667e-05,
183
+ "loss": 0.6199,
184
+ "step": 500
185
+ },
186
+ {
187
+ "epoch": 8.1,
188
+ "grad_norm": 1.6070563793182373,
189
+ "learning_rate": 1.4614583333333334e-05,
190
+ "loss": 0.6295,
191
+ "step": 520
192
+ },
193
+ {
194
+ "epoch": 8.41,
195
+ "grad_norm": 3.226116418838501,
196
+ "learning_rate": 1.4406250000000001e-05,
197
+ "loss": 0.6227,
198
+ "step": 540
199
+ },
200
+ {
201
+ "epoch": 8.72,
202
+ "grad_norm": 1.5464348793029785,
203
+ "learning_rate": 1.4197916666666667e-05,
204
+ "loss": 0.589,
205
+ "step": 560
206
+ },
207
+ {
208
+ "epoch": 9.04,
209
+ "grad_norm": 1.4290672540664673,
210
+ "learning_rate": 1.3989583333333334e-05,
211
+ "loss": 0.6263,
212
+ "step": 580
213
+ },
214
+ {
215
+ "epoch": 9.35,
216
+ "grad_norm": 1.682243824005127,
217
+ "learning_rate": 1.3781250000000001e-05,
218
+ "loss": 0.5747,
219
+ "step": 600
220
+ },
221
+ {
222
+ "epoch": 9.66,
223
+ "grad_norm": 1.6785274744033813,
224
+ "learning_rate": 1.3572916666666667e-05,
225
+ "loss": 0.5809,
226
+ "step": 620
227
+ },
228
+ {
229
+ "epoch": 9.97,
230
+ "grad_norm": 2.156558036804199,
231
+ "learning_rate": 1.3364583333333334e-05,
232
+ "loss": 0.5725,
233
+ "step": 640
234
+ },
235
+ {
236
+ "epoch": 10.28,
237
+ "grad_norm": 1.739134430885315,
238
+ "learning_rate": 1.3156250000000001e-05,
239
+ "loss": 0.549,
240
+ "step": 660
241
+ },
242
+ {
243
+ "epoch": 10.59,
244
+ "grad_norm": 1.2140896320343018,
245
+ "learning_rate": 1.2947916666666667e-05,
246
+ "loss": 0.5493,
247
+ "step": 680
248
+ },
249
+ {
250
+ "epoch": 10.91,
251
+ "grad_norm": 1.517707109451294,
252
+ "learning_rate": 1.2739583333333334e-05,
253
+ "loss": 0.5589,
254
+ "step": 700
255
+ },
256
+ {
257
+ "epoch": 11.22,
258
+ "grad_norm": 1.8752249479293823,
259
+ "learning_rate": 1.2531250000000001e-05,
260
+ "loss": 0.5001,
261
+ "step": 720
262
+ },
263
+ {
264
+ "epoch": 11.53,
265
+ "grad_norm": 1.8684614896774292,
266
+ "learning_rate": 1.2322916666666667e-05,
267
+ "loss": 0.52,
268
+ "step": 740
269
+ },
270
+ {
271
+ "epoch": 11.84,
272
+ "grad_norm": 2.3552052974700928,
273
+ "learning_rate": 1.2114583333333334e-05,
274
+ "loss": 0.5479,
275
+ "step": 760
276
+ },
277
+ {
278
+ "epoch": 12.15,
279
+ "grad_norm": 1.2074153423309326,
280
+ "learning_rate": 1.1906250000000001e-05,
281
+ "loss": 0.4977,
282
+ "step": 780
283
+ },
284
+ {
285
+ "epoch": 12.46,
286
+ "grad_norm": 3.0007381439208984,
287
+ "learning_rate": 1.1697916666666667e-05,
288
+ "loss": 0.5252,
289
+ "step": 800
290
+ },
291
+ {
292
+ "epoch": 12.78,
293
+ "grad_norm": 1.8956348896026611,
294
+ "learning_rate": 1.1489583333333334e-05,
295
+ "loss": 0.503,
296
+ "step": 820
297
+ },
298
+ {
299
+ "epoch": 13.09,
300
+ "grad_norm": 1.3779544830322266,
301
+ "learning_rate": 1.1281250000000001e-05,
302
+ "loss": 0.4991,
303
+ "step": 840
304
+ },
305
+ {
306
+ "epoch": 13.4,
307
+ "grad_norm": 1.5404250621795654,
308
+ "learning_rate": 1.1072916666666667e-05,
309
+ "loss": 0.5026,
310
+ "step": 860
311
+ },
312
+ {
313
+ "epoch": 13.71,
314
+ "grad_norm": 2.149167060852051,
315
+ "learning_rate": 1.0864583333333334e-05,
316
+ "loss": 0.4788,
317
+ "step": 880
318
+ },
319
+ {
320
+ "epoch": 14.02,
321
+ "grad_norm": 1.4978464841842651,
322
+ "learning_rate": 1.0656250000000002e-05,
323
+ "loss": 0.4605,
324
+ "step": 900
325
+ },
326
+ {
327
+ "epoch": 14.33,
328
+ "grad_norm": 1.5203664302825928,
329
+ "learning_rate": 1.0447916666666667e-05,
330
+ "loss": 0.4858,
331
+ "step": 920
332
+ },
333
+ {
334
+ "epoch": 14.64,
335
+ "grad_norm": 1.850074291229248,
336
+ "learning_rate": 1.0239583333333334e-05,
337
+ "loss": 0.4797,
338
+ "step": 940
339
+ },
340
+ {
341
+ "epoch": 14.96,
342
+ "grad_norm": 1.7591063976287842,
343
+ "learning_rate": 1.0031250000000002e-05,
344
+ "loss": 0.4619,
345
+ "step": 960
346
+ },
347
+ {
348
+ "epoch": 15.27,
349
+ "grad_norm": 1.4994142055511475,
350
+ "learning_rate": 9.822916666666667e-06,
351
+ "loss": 0.4627,
352
+ "step": 980
353
+ },
354
+ {
355
+ "epoch": 15.58,
356
+ "grad_norm": 1.888311743736267,
357
+ "learning_rate": 9.614583333333334e-06,
358
+ "loss": 0.4469,
359
+ "step": 1000
360
+ },
361
+ {
362
+ "epoch": 15.89,
363
+ "grad_norm": 1.7547377347946167,
364
+ "learning_rate": 9.406250000000002e-06,
365
+ "loss": 0.4379,
366
+ "step": 1020
367
+ },
368
+ {
369
+ "epoch": 16.2,
370
+ "grad_norm": 1.353371262550354,
371
+ "learning_rate": 9.197916666666667e-06,
372
+ "loss": 0.4464,
373
+ "step": 1040
374
+ },
375
+ {
376
+ "epoch": 16.51,
377
+ "grad_norm": 1.8960446119308472,
378
+ "learning_rate": 8.989583333333334e-06,
379
+ "loss": 0.4181,
380
+ "step": 1060
381
+ },
382
+ {
383
+ "epoch": 16.83,
384
+ "grad_norm": 1.644089937210083,
385
+ "learning_rate": 8.781250000000002e-06,
386
+ "loss": 0.481,
387
+ "step": 1080
388
+ },
389
+ {
390
+ "epoch": 17.14,
391
+ "grad_norm": 2.3077402114868164,
392
+ "learning_rate": 8.572916666666667e-06,
393
+ "loss": 0.4199,
394
+ "step": 1100
395
+ },
396
+ {
397
+ "epoch": 17.45,
398
+ "grad_norm": 2.0317394733428955,
399
+ "learning_rate": 8.364583333333334e-06,
400
+ "loss": 0.4178,
401
+ "step": 1120
402
+ },
403
+ {
404
+ "epoch": 17.76,
405
+ "grad_norm": 1.8424252271652222,
406
+ "learning_rate": 8.156250000000002e-06,
407
+ "loss": 0.4355,
408
+ "step": 1140
409
+ },
410
+ {
411
+ "epoch": 18.07,
412
+ "grad_norm": 2.730163335800171,
413
+ "learning_rate": 7.947916666666667e-06,
414
+ "loss": 0.4449,
415
+ "step": 1160
416
+ },
417
+ {
418
+ "epoch": 18.38,
419
+ "grad_norm": 1.7422763109207153,
420
+ "learning_rate": 7.739583333333333e-06,
421
+ "loss": 0.437,
422
+ "step": 1180
423
+ },
424
+ {
425
+ "epoch": 18.7,
426
+ "grad_norm": 1.8822110891342163,
427
+ "learning_rate": 7.531250000000001e-06,
428
+ "loss": 0.4313,
429
+ "step": 1200
430
+ },
431
+ {
432
+ "epoch": 19.01,
433
+ "grad_norm": 1.8389759063720703,
434
+ "learning_rate": 7.322916666666667e-06,
435
+ "loss": 0.4112,
436
+ "step": 1220
437
+ },
438
+ {
439
+ "epoch": 19.32,
440
+ "grad_norm": 1.2449133396148682,
441
+ "learning_rate": 7.114583333333334e-06,
442
+ "loss": 0.3969,
443
+ "step": 1240
444
+ },
445
+ {
446
+ "epoch": 19.63,
447
+ "grad_norm": 1.4220046997070312,
448
+ "learning_rate": 6.906250000000001e-06,
449
+ "loss": 0.4306,
450
+ "step": 1260
451
+ },
452
+ {
453
+ "epoch": 19.94,
454
+ "grad_norm": 1.3844712972640991,
455
+ "learning_rate": 6.697916666666667e-06,
456
+ "loss": 0.4085,
457
+ "step": 1280
458
+ },
459
+ {
460
+ "epoch": 20.25,
461
+ "grad_norm": 1.5047410726547241,
462
+ "learning_rate": 6.489583333333334e-06,
463
+ "loss": 0.4421,
464
+ "step": 1300
465
+ },
466
+ {
467
+ "epoch": 20.56,
468
+ "grad_norm": 1.6129776239395142,
469
+ "learning_rate": 6.281250000000001e-06,
470
+ "loss": 0.42,
471
+ "step": 1320
472
+ },
473
+ {
474
+ "epoch": 20.88,
475
+ "grad_norm": 3.006173849105835,
476
+ "learning_rate": 6.0729166666666675e-06,
477
+ "loss": 0.3902,
478
+ "step": 1340
479
+ },
480
+ {
481
+ "epoch": 21.19,
482
+ "grad_norm": 1.5414382219314575,
483
+ "learning_rate": 5.864583333333334e-06,
484
+ "loss": 0.3927,
485
+ "step": 1360
486
+ },
487
+ {
488
+ "epoch": 21.5,
489
+ "grad_norm": 2.1584370136260986,
490
+ "learning_rate": 5.656250000000001e-06,
491
+ "loss": 0.4052,
492
+ "step": 1380
493
+ },
494
+ {
495
+ "epoch": 21.81,
496
+ "grad_norm": 1.9206900596618652,
497
+ "learning_rate": 5.4479166666666675e-06,
498
+ "loss": 0.3872,
499
+ "step": 1400
500
+ },
501
+ {
502
+ "epoch": 22.12,
503
+ "grad_norm": 1.6601423025131226,
504
+ "learning_rate": 5.239583333333333e-06,
505
+ "loss": 0.4221,
506
+ "step": 1420
507
+ },
508
+ {
509
+ "epoch": 22.43,
510
+ "grad_norm": 1.8656572103500366,
511
+ "learning_rate": 5.031250000000001e-06,
512
+ "loss": 0.4031,
513
+ "step": 1440
514
+ },
515
+ {
516
+ "epoch": 22.75,
517
+ "grad_norm": 1.9469555616378784,
518
+ "learning_rate": 4.822916666666667e-06,
519
+ "loss": 0.395,
520
+ "step": 1460
521
+ },
522
+ {
523
+ "epoch": 23.06,
524
+ "grad_norm": 1.801340103149414,
525
+ "learning_rate": 4.614583333333334e-06,
526
+ "loss": 0.3584,
527
+ "step": 1480
528
+ },
529
+ {
530
+ "epoch": 23.37,
531
+ "grad_norm": 2.101327657699585,
532
+ "learning_rate": 4.40625e-06,
533
+ "loss": 0.4129,
534
+ "step": 1500
535
+ }
536
+ ],
537
+ "logging_steps": 20,
538
+ "max_steps": 1920,
539
+ "num_input_tokens_seen": 0,
540
+ "num_train_epochs": 30,
541
+ "save_steps": 500,
542
+ "total_flos": 3.8986916806656e+18,
543
+ "train_batch_size": 1,
544
+ "trial_name": null,
545
+ "trial_params": null
546
+ }
checkpoint-1500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf3bd8d0ae9745c03ca60e38c25c6801ef1b8e53b972f64cac4e257777c6a59b
3
+ size 4920
checkpoint-500/README.md ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ - bnb_4bit_quant_storage: uint8
19
+ - load_in_4bit: False
20
+ - load_in_8bit: True
21
+ ### Framework versions
22
+
23
+
24
+ - PEFT 0.5.0
checkpoint-500/adapter_config.json ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "sallywww/Llama-7B",
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
+ "q_proj",
18
+ "v_proj"
19
+ ],
20
+ "task_type": "CAUSAL_LM"
21
+ }
checkpoint-500/adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:654f695d6e03024337c73cbe966ca72e2fee1b9e6ab5b9b1239e7c257c0a78f7
3
+ size 16794200
checkpoint-500/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f91499d6daaa5c0820c2b01015605bb44bf64e08f9fd2b0f29fdd4c35a1cf0cd
3
+ size 33662074
checkpoint-500/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f19e5dab01fca7eac74ccda881a891a69a6035d9ca2466c8864ab4197bb0bfd5
3
+ size 14244
checkpoint-500/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dc017cc9d795a6f91b7e052338890b819f0d888fdf983f6a197b661b094767f6
3
+ size 1064
checkpoint-500/trainer_state.json ADDED
@@ -0,0 +1,196 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 7.789678675754625,
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": 0.31,
13
+ "grad_norm": 0.48931118845939636,
14
+ "learning_rate": 1.98125e-05,
15
+ "loss": 1.7417,
16
+ "step": 20
17
+ },
18
+ {
19
+ "epoch": 0.62,
20
+ "grad_norm": 0.56327223777771,
21
+ "learning_rate": 1.9604166666666668e-05,
22
+ "loss": 1.7626,
23
+ "step": 40
24
+ },
25
+ {
26
+ "epoch": 0.93,
27
+ "grad_norm": 0.6054126620292664,
28
+ "learning_rate": 1.9395833333333335e-05,
29
+ "loss": 1.5774,
30
+ "step": 60
31
+ },
32
+ {
33
+ "epoch": 1.25,
34
+ "grad_norm": 0.8035449385643005,
35
+ "learning_rate": 1.9187500000000002e-05,
36
+ "loss": 1.4772,
37
+ "step": 80
38
+ },
39
+ {
40
+ "epoch": 1.56,
41
+ "grad_norm": 0.9972027540206909,
42
+ "learning_rate": 1.897916666666667e-05,
43
+ "loss": 1.409,
44
+ "step": 100
45
+ },
46
+ {
47
+ "epoch": 1.87,
48
+ "grad_norm": 0.7526600360870361,
49
+ "learning_rate": 1.8781250000000003e-05,
50
+ "loss": 1.3088,
51
+ "step": 120
52
+ },
53
+ {
54
+ "epoch": 2.18,
55
+ "grad_norm": 0.8370587825775146,
56
+ "learning_rate": 1.8572916666666666e-05,
57
+ "loss": 1.2086,
58
+ "step": 140
59
+ },
60
+ {
61
+ "epoch": 2.49,
62
+ "grad_norm": 0.8337924480438232,
63
+ "learning_rate": 1.8364583333333334e-05,
64
+ "loss": 1.1321,
65
+ "step": 160
66
+ },
67
+ {
68
+ "epoch": 2.8,
69
+ "grad_norm": 1.0028570890426636,
70
+ "learning_rate": 1.815625e-05,
71
+ "loss": 1.0439,
72
+ "step": 180
73
+ },
74
+ {
75
+ "epoch": 3.12,
76
+ "grad_norm": 0.7663702368736267,
77
+ "learning_rate": 1.7947916666666668e-05,
78
+ "loss": 1.026,
79
+ "step": 200
80
+ },
81
+ {
82
+ "epoch": 3.43,
83
+ "grad_norm": 0.8835870623588562,
84
+ "learning_rate": 1.7739583333333335e-05,
85
+ "loss": 0.9994,
86
+ "step": 220
87
+ },
88
+ {
89
+ "epoch": 3.74,
90
+ "grad_norm": 1.010624885559082,
91
+ "learning_rate": 1.7531250000000003e-05,
92
+ "loss": 0.9381,
93
+ "step": 240
94
+ },
95
+ {
96
+ "epoch": 4.05,
97
+ "grad_norm": 0.9645354151725769,
98
+ "learning_rate": 1.7322916666666666e-05,
99
+ "loss": 0.8667,
100
+ "step": 260
101
+ },
102
+ {
103
+ "epoch": 4.36,
104
+ "grad_norm": 0.9926008582115173,
105
+ "learning_rate": 1.7114583333333334e-05,
106
+ "loss": 0.8729,
107
+ "step": 280
108
+ },
109
+ {
110
+ "epoch": 4.67,
111
+ "grad_norm": 1.0364482402801514,
112
+ "learning_rate": 1.690625e-05,
113
+ "loss": 0.836,
114
+ "step": 300
115
+ },
116
+ {
117
+ "epoch": 4.99,
118
+ "grad_norm": 1.3519001007080078,
119
+ "learning_rate": 1.6697916666666668e-05,
120
+ "loss": 0.8126,
121
+ "step": 320
122
+ },
123
+ {
124
+ "epoch": 5.3,
125
+ "grad_norm": 1.2280508279800415,
126
+ "learning_rate": 1.6489583333333335e-05,
127
+ "loss": 0.7953,
128
+ "step": 340
129
+ },
130
+ {
131
+ "epoch": 5.61,
132
+ "grad_norm": 0.8931779265403748,
133
+ "learning_rate": 1.6281250000000003e-05,
134
+ "loss": 0.7813,
135
+ "step": 360
136
+ },
137
+ {
138
+ "epoch": 5.92,
139
+ "grad_norm": 2.7053027153015137,
140
+ "learning_rate": 1.6072916666666667e-05,
141
+ "loss": 0.7256,
142
+ "step": 380
143
+ },
144
+ {
145
+ "epoch": 6.23,
146
+ "grad_norm": 1.411024808883667,
147
+ "learning_rate": 1.5864583333333334e-05,
148
+ "loss": 0.7336,
149
+ "step": 400
150
+ },
151
+ {
152
+ "epoch": 6.54,
153
+ "grad_norm": 1.0111807584762573,
154
+ "learning_rate": 1.565625e-05,
155
+ "loss": 0.7207,
156
+ "step": 420
157
+ },
158
+ {
159
+ "epoch": 6.85,
160
+ "grad_norm": 2.213623523712158,
161
+ "learning_rate": 1.544791666666667e-05,
162
+ "loss": 0.664,
163
+ "step": 440
164
+ },
165
+ {
166
+ "epoch": 7.17,
167
+ "grad_norm": 1.3642323017120361,
168
+ "learning_rate": 1.5239583333333334e-05,
169
+ "loss": 0.6848,
170
+ "step": 460
171
+ },
172
+ {
173
+ "epoch": 7.48,
174
+ "grad_norm": 1.3692028522491455,
175
+ "learning_rate": 1.5031250000000001e-05,
176
+ "loss": 0.6663,
177
+ "step": 480
178
+ },
179
+ {
180
+ "epoch": 7.79,
181
+ "grad_norm": 1.9850131273269653,
182
+ "learning_rate": 1.4822916666666667e-05,
183
+ "loss": 0.6199,
184
+ "step": 500
185
+ }
186
+ ],
187
+ "logging_steps": 20,
188
+ "max_steps": 1920,
189
+ "num_input_tokens_seen": 0,
190
+ "num_train_epochs": 30,
191
+ "save_steps": 500,
192
+ "total_flos": 1.2995638935552e+18,
193
+ "train_batch_size": 1,
194
+ "trial_name": null,
195
+ "trial_params": null
196
+ }
checkpoint-500/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf3bd8d0ae9745c03ca60e38c25c6801ef1b8e53b972f64cac4e257777c6a59b
3
+ size 4920