Hugging Face's logo

terry69
/
prometheus-7b-direct-p0.1_seed42level3

TensorBoard
Safetensors
mistral
alignment-handbook
trl
sft
Generated from Trainer
Model card Files Metrics Community
terry69 commited on
Commit
3f47377
•
1 Parent(s): 11512f6

Model save

Browse files
Files changed (6) hide show
  1. README.md +67 -0
  2. all_results.json +9 -0
  3. generation_config.json +6 -0
  4. runs/Aug12_03-18-36_COE-CS-sv003/events.out.tfevents.1723432798.COE-CS-sv003.3809886.0 +2 -2
  5. train_results.json +9 -0
  6. trainer_state.json +302 -0
README.md ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ base_model: mistralai/Mistral-7B-Instruct-v0.2
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ datasets:
9
+ - generator
10
+ model-index:
11
+ - name: prometheus-7b-direct-p0.1_seed42level3
12
+ results: []
13
+ ---
14
+
15
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
16
+ should probably proofread and complete it, then remove this comment. -->
17
+
18
+ # prometheus-7b-direct-p0.1_seed42level3
19
+
20
+ This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset.
21
+ It achieves the following results on the evaluation set:
22
+ - Loss: 0.6546
23
+
24
+ ## Model description
25
+
26
+ More information needed
27
+
28
+ ## Intended uses & limitations
29
+
30
+ More information needed
31
+
32
+ ## Training and evaluation data
33
+
34
+ More information needed
35
+
36
+ ## Training procedure
37
+
38
+ ### Training hyperparameters
39
+
40
+ The following hyperparameters were used during training:
41
+ - learning_rate: 1e-05
42
+ - train_batch_size: 2
43
+ - eval_batch_size: 1
44
+ - seed: 42
45
+ - distributed_type: multi-GPU
46
+ - num_devices: 4
47
+ - gradient_accumulation_steps: 4
48
+ - total_train_batch_size: 32
49
+ - total_eval_batch_size: 4
50
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
+ - lr_scheduler_type: cosine
52
+ - lr_scheduler_warmup_ratio: 0.1
53
+ - num_epochs: 1
54
+
55
+ ### Training results
56
+
57
+ | Training Loss | Epoch | Step | Validation Loss |
58
+ |:-------------:|:------:|:----:|:---------------:|
59
+ | 0.4721 | 0.9986 | 176 | 0.6546 |
60
+
61
+
62
+ ### Framework versions
63
+
64
+ - Transformers 4.43.4
65
+ - Pytorch 2.3.1+cu121
66
+ - Datasets 2.19.1
67
+ - Tokenizers 0.19.1
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9985815602836879,
3
+ "total_flos": 36798474485760.0,
4
+ "train_loss": 0.6350141577422619,
5
+ "train_runtime": 4225.9013,
6
+ "train_samples": 18801,
7
+ "train_samples_per_second": 1.333,
8
+ "train_steps_per_second": 0.042
9
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.43.4"
6
+ }
runs/Aug12_03-18-36_COE-CS-sv003/events.out.tfevents.1723432798.COE-CS-sv003.3809886.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:269e22db6dbe224c9e84f691c5122a24434641b26c1eb399dc59af54f074417c
3
- size 12736
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a604e2f2e44e38d29ec88aea9160f5141ffb16596985a4f062b7d8f718245ca7
3
+ size 13361
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9985815602836879,
3
+ "total_flos": 36798474485760.0,
4
+ "train_loss": 0.6350141577422619,
5
+ "train_runtime": 4225.9013,
6
+ "train_samples": 18801,
7
+ "train_samples_per_second": 1.333,
8
+ "train_steps_per_second": 0.042
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,302 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.9985815602836879,
5
+ "eval_steps": 500,
6
+ "global_step": 176,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.005673758865248227,
13
+ "grad_norm": 23.321381386134785,
14
+ "learning_rate": 5.555555555555555e-07,
15
+ "loss": 1.3529,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.028368794326241134,
20
+ "grad_norm": 8.891056445208546,
21
+ "learning_rate": 2.7777777777777783e-06,
22
+ "loss": 1.2787,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.05673758865248227,
27
+ "grad_norm": 4.590231197414839,
28
+ "learning_rate": 5.555555555555557e-06,
29
+ "loss": 1.0257,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.0851063829787234,
34
+ "grad_norm": 3.142582341930927,
35
+ "learning_rate": 8.333333333333334e-06,
36
+ "loss": 0.8807,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.11347517730496454,
41
+ "grad_norm": 2.843802286813948,
42
+ "learning_rate": 9.99604698613651e-06,
43
+ "loss": 0.843,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.14184397163120568,
48
+ "grad_norm": 2.4374270657643065,
49
+ "learning_rate": 9.951647332362511e-06,
50
+ "loss": 0.8103,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.1702127659574468,
55
+ "grad_norm": 2.3244501516240073,
56
+ "learning_rate": 9.85834670020205e-06,
57
+ "loss": 0.778,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.19858156028368795,
62
+ "grad_norm": 2.3307590830855576,
63
+ "learning_rate": 9.717066498610673e-06,
64
+ "loss": 0.7493,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.22695035460992907,
69
+ "grad_norm": 2.867283965545168,
70
+ "learning_rate": 9.529201968327618e-06,
71
+ "loss": 0.7304,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.2553191489361702,
76
+ "grad_norm": 2.503845781291395,
77
+ "learning_rate": 9.296608402898306e-06,
78
+ "loss": 0.711,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.28368794326241137,
83
+ "grad_norm": 2.4532179305680364,
84
+ "learning_rate": 9.021582826353825e-06,
85
+ "loss": 0.6915,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.3120567375886525,
90
+ "grad_norm": 2.133742309031153,
91
+ "learning_rate": 8.706841308493092e-06,
92
+ "loss": 0.6713,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.3404255319148936,
97
+ "grad_norm": 2.3597738337675187,
98
+ "learning_rate": 8.355492141795185e-06,
99
+ "loss": 0.6476,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.36879432624113473,
104
+ "grad_norm": 2.41256063962414,
105
+ "learning_rate": 7.971005144858554e-06,
106
+ "loss": 0.6409,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.3971631205673759,
111
+ "grad_norm": 2.5863453080504275,
112
+ "learning_rate": 7.5571773955171124e-06,
113
+ "loss": 0.6307,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.425531914893617,
118
+ "grad_norm": 2.1230662364450525,
119
+ "learning_rate": 7.118095732042643e-06,
120
+ "loss": 0.6241,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.45390070921985815,
125
+ "grad_norm": 2.339842319600617,
126
+ "learning_rate": 6.65809639276034e-06,
127
+ "loss": 0.61,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.48226950354609927,
132
+ "grad_norm": 2.203220986165918,
133
+ "learning_rate": 6.181722192664526e-06,
134
+ "loss": 0.607,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.5106382978723404,
139
+ "grad_norm": 2.057055175195757,
140
+ "learning_rate": 5.693677659945343e-06,
141
+ "loss": 0.568,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.5390070921985816,
146
+ "grad_norm": 2.0107641524633713,
147
+ "learning_rate": 5.19878257548463e-06,
148
+ "loss": 0.5581,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.5673758865248227,
153
+ "grad_norm": 2.342751630632786,
154
+ "learning_rate": 4.701924374150901e-06,
155
+ "loss": 0.5588,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.5957446808510638,
160
+ "grad_norm": 2.039824613266622,
161
+ "learning_rate": 4.2080098779639255e-06,
162
+ "loss": 0.5543,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.624113475177305,
167
+ "grad_norm": 2.1842313088053107,
168
+ "learning_rate": 3.721916837797627e-06,
169
+ "loss": 0.5382,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.6524822695035462,
174
+ "grad_norm": 2.3866340812713456,
175
+ "learning_rate": 3.2484457621808787e-06,
176
+ "loss": 0.5336,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.6808510638297872,
181
+ "grad_norm": 2.3598347444604637,
182
+ "learning_rate": 2.792272508920443e-06,
183
+ "loss": 0.5298,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.7092198581560284,
188
+ "grad_norm": 2.194353937213305,
189
+ "learning_rate": 2.3579021077369047e-06,
190
+ "loss": 0.5092,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.7375886524822695,
195
+ "grad_norm": 2.2206104013150805,
196
+ "learning_rate": 1.949624269947378e-06,
197
+ "loss": 0.5163,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.7659574468085106,
202
+ "grad_norm": 2.2710578602998623,
203
+ "learning_rate": 1.5714710245679348e-06,
204
+ "loss": 0.5106,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.7943262411347518,
209
+ "grad_norm": 2.0376221819656752,
210
+ "learning_rate": 1.227176899208849e-06,
211
+ "loss": 0.5101,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.8226950354609929,
216
+ "grad_norm": 2.0283659113726586,
217
+ "learning_rate": 9.201420390041965e-07,
218
+ "loss": 0.5071,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.851063829787234,
223
+ "grad_norm": 2.008652811277033,
224
+ "learning_rate": 6.533986278020876e-07,
225
+ "loss": 0.4827,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.8794326241134752,
230
+ "grad_norm": 2.0949057219648286,
231
+ "learning_rate": 4.2958094322982703e-07,
232
+ "loss": 0.5023,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.9078014184397163,
237
+ "grad_norm": 2.139405991925096,
238
+ "learning_rate": 2.5089934136108665e-07,
239
+ "loss": 0.4934,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.9361702127659575,
244
+ "grad_norm": 2.0539208771983923,
245
+ "learning_rate": 1.1911842790474637e-07,
246
+ "loss": 0.489,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.9645390070921985,
251
+ "grad_norm": 1.9792159065875536,
252
+ "learning_rate": 3.553963149013295e-08,
253
+ "loss": 0.4794,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.9929078014184397,
258
+ "grad_norm": 1.9554520695912883,
259
+ "learning_rate": 9.883511496722176e-10,
260
+ "loss": 0.4721,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.9985815602836879,
265
+ "eval_loss": 0.6546463370323181,
266
+ "eval_runtime": 96.7348,
267
+ "eval_samples_per_second": 3.122,
268
+ "eval_steps_per_second": 0.786,
269
+ "step": 176
270
+ },
271
+ {
272
+ "epoch": 0.9985815602836879,
273
+ "step": 176,
274
+ "total_flos": 36798474485760.0,
275
+ "train_loss": 0.6350141577422619,
276
+ "train_runtime": 4225.9013,
277
+ "train_samples_per_second": 1.333,
278
+ "train_steps_per_second": 0.042
279
+ }
280
+ ],
281
+ "logging_steps": 5,
282
+ "max_steps": 176,
283
+ "num_input_tokens_seen": 0,
284
+ "num_train_epochs": 1,
285
+ "save_steps": 100,
286
+ "stateful_callbacks": {
287
+ "TrainerControl": {
288
+ "args": {
289
+ "should_epoch_stop": false,
290
+ "should_evaluate": false,
291
+ "should_log": false,
292
+ "should_save": true,
293
+ "should_training_stop": true
294
+ },
295
+ "attributes": {}
296
+ }
297
+ },
298
+ "total_flos": 36798474485760.0,
299
+ "train_batch_size": 2,
300
+ "trial_name": null,
301
+ "trial_params": null
302
+ }