GaetanMichelet commited on
Commit
c5b8a3c
1 Parent(s): 9f4d733

Model save

Browse files
README.md ADDED
@@ -0,0 +1,78 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: mistralai/Mistral-7B-Instruct-v0.3
3
+ library_name: peft
4
+ license: apache-2.0
5
+ tags:
6
+ - trl
7
+ - sft
8
+ - generated_from_trainer
9
+ model-index:
10
+ - name: Mistral-7B_task-3_60-samples_config-2
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # Mistral-7B_task-3_60-samples_config-2
18
+
19
+ This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.3](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.3) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.9100
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 0.0001
41
+ - train_batch_size: 1
42
+ - eval_batch_size: 1
43
+ - seed: 42
44
+ - distributed_type: multi-GPU
45
+ - gradient_accumulation_steps: 16
46
+ - total_train_batch_size: 16
47
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
48
+ - lr_scheduler_type: cosine
49
+ - lr_scheduler_warmup_ratio: 0.1
50
+ - num_epochs: 50
51
+
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Validation Loss |
55
+ |:-------------:|:-------:|:----:|:---------------:|
56
+ | 3.6443 | 0.6957 | 2 | 3.4455 |
57
+ | 3.1998 | 1.7391 | 5 | 1.8330 |
58
+ | 1.1687 | 2.7826 | 8 | 0.7476 |
59
+ | 0.7861 | 3.8261 | 11 | 0.5044 |
60
+ | 0.3983 | 4.8696 | 14 | 0.4260 |
61
+ | 0.359 | 5.9130 | 17 | 0.3850 |
62
+ | 0.2531 | 6.9565 | 20 | 0.3752 |
63
+ | 0.1449 | 8.0 | 23 | 0.4619 |
64
+ | 0.1711 | 8.6957 | 25 | 0.5146 |
65
+ | 0.0704 | 9.7391 | 28 | 0.6076 |
66
+ | 0.0479 | 10.7826 | 31 | 0.7174 |
67
+ | 0.0096 | 11.8261 | 34 | 0.8062 |
68
+ | 0.0056 | 12.8696 | 37 | 0.8766 |
69
+ | 0.0024 | 13.9130 | 40 | 0.9100 |
70
+
71
+
72
+ ### Framework versions
73
+
74
+ - PEFT 0.12.0
75
+ - Transformers 4.44.0
76
+ - Pytorch 2.1.2+cu121
77
+ - Datasets 2.20.0
78
+ - Tokenizers 0.19.1
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:57fc433c67914aa2f50b346869802ee2fb5bdc8fc8f9c59eff7b948605ad4ed3
3
  size 167832240
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:09cbe18f8bcfb35713867a2e129dff7652f78ffc9c9c029967acc23a501e1d0d
3
  size 167832240
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 13.91304347826087,
3
+ "total_flos": 6.346894531887104e+16,
4
+ "train_loss": 0.6619909038650803,
5
+ "train_runtime": 2367.3473,
6
+ "train_samples": 46,
7
+ "train_samples_per_second": 0.972,
8
+ "train_steps_per_second": 0.042
9
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 13.91304347826087,
3
+ "total_flos": 6.346894531887104e+16,
4
+ "train_loss": 0.6619909038650803,
5
+ "train_runtime": 2367.3473,
6
+ "train_samples": 46,
7
+ "train_samples_per_second": 0.972,
8
+ "train_steps_per_second": 0.042
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,310 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.37516140937805176,
3
+ "best_model_checkpoint": "data/Mistral-7B_task-3_60-samples_config-2/checkpoint-20",
4
+ "epoch": 13.91304347826087,
5
+ "eval_steps": 500,
6
+ "global_step": 40,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.34782608695652173,
13
+ "grad_norm": 17.986906051635742,
14
+ "learning_rate": 1e-05,
15
+ "loss": 3.8263,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.6956521739130435,
20
+ "grad_norm": 16.079301834106445,
21
+ "learning_rate": 2e-05,
22
+ "loss": 3.6443,
23
+ "step": 2
24
+ },
25
+ {
26
+ "epoch": 0.6956521739130435,
27
+ "eval_loss": 3.4455251693725586,
28
+ "eval_runtime": 13.8084,
29
+ "eval_samples_per_second": 0.869,
30
+ "eval_steps_per_second": 0.869,
31
+ "step": 2
32
+ },
33
+ {
34
+ "epoch": 1.391304347826087,
35
+ "grad_norm": 14.32006549835205,
36
+ "learning_rate": 4e-05,
37
+ "loss": 3.1998,
38
+ "step": 4
39
+ },
40
+ {
41
+ "epoch": 1.7391304347826086,
42
+ "eval_loss": 1.8330297470092773,
43
+ "eval_runtime": 13.8064,
44
+ "eval_samples_per_second": 0.869,
45
+ "eval_steps_per_second": 0.869,
46
+ "step": 5
47
+ },
48
+ {
49
+ "epoch": 2.0869565217391304,
50
+ "grad_norm": 6.900447845458984,
51
+ "learning_rate": 6e-05,
52
+ "loss": 2.0772,
53
+ "step": 6
54
+ },
55
+ {
56
+ "epoch": 2.782608695652174,
57
+ "grad_norm": 4.268274784088135,
58
+ "learning_rate": 8e-05,
59
+ "loss": 1.1687,
60
+ "step": 8
61
+ },
62
+ {
63
+ "epoch": 2.782608695652174,
64
+ "eval_loss": 0.7475635409355164,
65
+ "eval_runtime": 13.8049,
66
+ "eval_samples_per_second": 0.869,
67
+ "eval_steps_per_second": 0.869,
68
+ "step": 8
69
+ },
70
+ {
71
+ "epoch": 3.4782608695652173,
72
+ "grad_norm": 1.7821996212005615,
73
+ "learning_rate": 0.0001,
74
+ "loss": 0.7861,
75
+ "step": 10
76
+ },
77
+ {
78
+ "epoch": 3.8260869565217392,
79
+ "eval_loss": 0.5044365525245667,
80
+ "eval_runtime": 13.806,
81
+ "eval_samples_per_second": 0.869,
82
+ "eval_steps_per_second": 0.869,
83
+ "step": 11
84
+ },
85
+ {
86
+ "epoch": 4.173913043478261,
87
+ "grad_norm": 0.9720113277435303,
88
+ "learning_rate": 9.987820251299122e-05,
89
+ "loss": 0.4289,
90
+ "step": 12
91
+ },
92
+ {
93
+ "epoch": 4.869565217391305,
94
+ "grad_norm": 1.0473393201828003,
95
+ "learning_rate": 9.951340343707852e-05,
96
+ "loss": 0.3983,
97
+ "step": 14
98
+ },
99
+ {
100
+ "epoch": 4.869565217391305,
101
+ "eval_loss": 0.4259566068649292,
102
+ "eval_runtime": 13.8111,
103
+ "eval_samples_per_second": 0.869,
104
+ "eval_steps_per_second": 0.869,
105
+ "step": 14
106
+ },
107
+ {
108
+ "epoch": 5.565217391304348,
109
+ "grad_norm": 0.6317740678787231,
110
+ "learning_rate": 9.890738003669029e-05,
111
+ "loss": 0.359,
112
+ "step": 16
113
+ },
114
+ {
115
+ "epoch": 5.913043478260869,
116
+ "eval_loss": 0.38502585887908936,
117
+ "eval_runtime": 13.8077,
118
+ "eval_samples_per_second": 0.869,
119
+ "eval_steps_per_second": 0.869,
120
+ "step": 17
121
+ },
122
+ {
123
+ "epoch": 6.260869565217392,
124
+ "grad_norm": 0.6458030343055725,
125
+ "learning_rate": 9.806308479691595e-05,
126
+ "loss": 0.2136,
127
+ "step": 18
128
+ },
129
+ {
130
+ "epoch": 6.956521739130435,
131
+ "grad_norm": 0.5151018500328064,
132
+ "learning_rate": 9.698463103929542e-05,
133
+ "loss": 0.2531,
134
+ "step": 20
135
+ },
136
+ {
137
+ "epoch": 6.956521739130435,
138
+ "eval_loss": 0.37516140937805176,
139
+ "eval_runtime": 13.8069,
140
+ "eval_samples_per_second": 0.869,
141
+ "eval_steps_per_second": 0.869,
142
+ "step": 20
143
+ },
144
+ {
145
+ "epoch": 7.6521739130434785,
146
+ "grad_norm": 0.46064192056655884,
147
+ "learning_rate": 9.567727288213005e-05,
148
+ "loss": 0.1449,
149
+ "step": 22
150
+ },
151
+ {
152
+ "epoch": 8.0,
153
+ "eval_loss": 0.4618639051914215,
154
+ "eval_runtime": 13.8049,
155
+ "eval_samples_per_second": 0.869,
156
+ "eval_steps_per_second": 0.869,
157
+ "step": 23
158
+ },
159
+ {
160
+ "epoch": 8.347826086956522,
161
+ "grad_norm": 0.44753408432006836,
162
+ "learning_rate": 9.414737964294636e-05,
163
+ "loss": 0.1711,
164
+ "step": 24
165
+ },
166
+ {
167
+ "epoch": 8.695652173913043,
168
+ "eval_loss": 0.5145867466926575,
169
+ "eval_runtime": 13.8075,
170
+ "eval_samples_per_second": 0.869,
171
+ "eval_steps_per_second": 0.869,
172
+ "step": 25
173
+ },
174
+ {
175
+ "epoch": 9.043478260869565,
176
+ "grad_norm": 0.706068754196167,
177
+ "learning_rate": 9.24024048078213e-05,
178
+ "loss": 0.1349,
179
+ "step": 26
180
+ },
181
+ {
182
+ "epoch": 9.73913043478261,
183
+ "grad_norm": 0.6422617435455322,
184
+ "learning_rate": 9.045084971874738e-05,
185
+ "loss": 0.0704,
186
+ "step": 28
187
+ },
188
+ {
189
+ "epoch": 9.73913043478261,
190
+ "eval_loss": 0.6075521111488342,
191
+ "eval_runtime": 13.8148,
192
+ "eval_samples_per_second": 0.869,
193
+ "eval_steps_per_second": 0.869,
194
+ "step": 28
195
+ },
196
+ {
197
+ "epoch": 10.434782608695652,
198
+ "grad_norm": 0.45618292689323425,
199
+ "learning_rate": 8.83022221559489e-05,
200
+ "loss": 0.0479,
201
+ "step": 30
202
+ },
203
+ {
204
+ "epoch": 10.782608695652174,
205
+ "eval_loss": 0.717354953289032,
206
+ "eval_runtime": 13.804,
207
+ "eval_samples_per_second": 0.869,
208
+ "eval_steps_per_second": 0.869,
209
+ "step": 31
210
+ },
211
+ {
212
+ "epoch": 11.130434782608695,
213
+ "grad_norm": 0.6182390451431274,
214
+ "learning_rate": 8.596699001693255e-05,
215
+ "loss": 0.0287,
216
+ "step": 32
217
+ },
218
+ {
219
+ "epoch": 11.826086956521738,
220
+ "grad_norm": 0.749740481376648,
221
+ "learning_rate": 8.345653031794292e-05,
222
+ "loss": 0.0096,
223
+ "step": 34
224
+ },
225
+ {
226
+ "epoch": 11.826086956521738,
227
+ "eval_loss": 0.8061732649803162,
228
+ "eval_runtime": 13.8091,
229
+ "eval_samples_per_second": 0.869,
230
+ "eval_steps_per_second": 0.869,
231
+ "step": 34
232
+ },
233
+ {
234
+ "epoch": 12.521739130434783,
235
+ "grad_norm": 0.5463645458221436,
236
+ "learning_rate": 8.07830737662829e-05,
237
+ "loss": 0.0056,
238
+ "step": 36
239
+ },
240
+ {
241
+ "epoch": 12.869565217391305,
242
+ "eval_loss": 0.876629650592804,
243
+ "eval_runtime": 13.8082,
244
+ "eval_samples_per_second": 0.869,
245
+ "eval_steps_per_second": 0.869,
246
+ "step": 37
247
+ },
248
+ {
249
+ "epoch": 13.217391304347826,
250
+ "grad_norm": 0.4603806734085083,
251
+ "learning_rate": 7.795964517353735e-05,
252
+ "loss": 0.0034,
253
+ "step": 38
254
+ },
255
+ {
256
+ "epoch": 13.91304347826087,
257
+ "grad_norm": 0.5449416041374207,
258
+ "learning_rate": 7.500000000000001e-05,
259
+ "loss": 0.0024,
260
+ "step": 40
261
+ },
262
+ {
263
+ "epoch": 13.91304347826087,
264
+ "eval_loss": 0.9099900722503662,
265
+ "eval_runtime": 13.8053,
266
+ "eval_samples_per_second": 0.869,
267
+ "eval_steps_per_second": 0.869,
268
+ "step": 40
269
+ },
270
+ {
271
+ "epoch": 13.91304347826087,
272
+ "step": 40,
273
+ "total_flos": 6.346894531887104e+16,
274
+ "train_loss": 0.6619909038650803,
275
+ "train_runtime": 2367.3473,
276
+ "train_samples_per_second": 0.972,
277
+ "train_steps_per_second": 0.042
278
+ }
279
+ ],
280
+ "logging_steps": 2,
281
+ "max_steps": 100,
282
+ "num_input_tokens_seen": 0,
283
+ "num_train_epochs": 50,
284
+ "save_steps": 25,
285
+ "stateful_callbacks": {
286
+ "EarlyStoppingCallback": {
287
+ "args": {
288
+ "early_stopping_patience": 7,
289
+ "early_stopping_threshold": 0.0
290
+ },
291
+ "attributes": {
292
+ "early_stopping_patience_counter": 0
293
+ }
294
+ },
295
+ "TrainerControl": {
296
+ "args": {
297
+ "should_epoch_stop": false,
298
+ "should_evaluate": false,
299
+ "should_log": false,
300
+ "should_save": true,
301
+ "should_training_stop": true
302
+ },
303
+ "attributes": {}
304
+ }
305
+ },
306
+ "total_flos": 6.346894531887104e+16,
307
+ "train_batch_size": 1,
308
+ "trial_name": null,
309
+ "trial_params": null
310
+ }