Hugging Face's logo

terry69
/
sanity_syntax

PEFT
TensorBoard
Safetensors
llama
alignment-handbook
trl
sft
Generated from Trainer
Model card Files
xet
 Metrics Community
terry69 commited on
Commit
ee16748
·
verified ·
1 Parent(s): 6addd63

Model save

Browse files
Files changed (5) hide show
  1. README.md +69 -0
  2. all_results.json +9 -0
  3. runs/Aug21_17-09-14_COE-CS-sv003/events.out.tfevents.1724260172.COE-CS-sv003.2438382.0 +2 -2
  4. train_results.json +9 -0
  5. trainer_state.json +302 -0
README.md ADDED
@@ -0,0 +1,69 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: llama3
3
+ library_name: peft
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ base_model: meta-llama/Meta-Llama-3-8B
9
+ datasets:
10
+ - generator
11
+ model-index:
12
+ - name: sanity_syntax
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # sanity_syntax
20
+
21
+ This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 1.2360
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 0.0002
43
+ - train_batch_size: 4
44
+ - eval_batch_size: 4
45
+ - seed: 42
46
+ - distributed_type: multi-GPU
47
+ - num_devices: 4
48
+ - gradient_accumulation_steps: 4
49
+ - total_train_batch_size: 64
50
+ - total_eval_batch_size: 16
51
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
52
+ - lr_scheduler_type: cosine
53
+ - lr_scheduler_warmup_ratio: 0.1
54
+ - num_epochs: 1
55
+
56
+ ### Training results
57
+
58
+ | Training Loss | Epoch | Step | Validation Loss |
59
+ |:-------------:|:------:|:----:|:---------------:|
60
+ | 1.4333 | 0.9958 | 176 | 1.2360 |
61
+
62
+
63
+ ### Framework versions
64
+
65
+ - PEFT 0.11.1
66
+ - Transformers 4.43.4
67
+ - Pytorch 2.3.1+cu121
68
+ - Datasets 2.19.1
69
+ - Tokenizers 0.19.1
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9957567185289957,
3
+ "total_flos": 2207623390691328.0,
4
+ "train_loss": 1.0380965369668873,
5
+ "train_runtime": 4181.7688,
6
+ "train_samples": 20700,
7
+ "train_samples_per_second": 2.703,
8
+ "train_steps_per_second": 0.042
9
+ }
runs/Aug21_17-09-14_COE-CS-sv003/events.out.tfevents.1724260172.COE-CS-sv003.2438382.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8d4f9c258c24d51fd2abbf407305b7c72f6fa9f909aecf2b83093bc0416ed78f
3
- size 10406
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3100aa8a3c1e2d7d3fdd83f602dd1d7ed424798d8b2448112c063738f13a2ca6
3
+ size 11031
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.9957567185289957,
3
+ "total_flos": 2207623390691328.0,
4
+ "train_loss": 1.0380965369668873,
5
+ "train_runtime": 4181.7688,
6
+ "train_samples": 20700,
7
+ "train_samples_per_second": 2.703,
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": 0.9957567185289957,
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": 0.005657708628005658,
13
+ "grad_norm": 0.43130557078322906,
14
+ "learning_rate": 1.1111111111111112e-05,
15
+ "loss": 1.9358,
16
+ "step": 1
17
+ },
18
+ {
19
+ "epoch": 0.028288543140028287,
20
+ "grad_norm": 0.47181074868090295,
21
+ "learning_rate": 5.555555555555556e-05,
22
+ "loss": 1.8755,
23
+ "step": 5
24
+ },
25
+ {
26
+ "epoch": 0.056577086280056574,
27
+ "grad_norm": 0.24499816695787435,
28
+ "learning_rate": 0.00011111111111111112,
29
+ "loss": 1.8709,
30
+ "step": 10
31
+ },
32
+ {
33
+ "epoch": 0.08486562942008487,
34
+ "grad_norm": 0.1946906656901304,
35
+ "learning_rate": 0.0001666666666666667,
36
+ "loss": 1.783,
37
+ "step": 15
38
+ },
39
+ {
40
+ "epoch": 0.11315417256011315,
41
+ "grad_norm": 0.16173800943880282,
42
+ "learning_rate": 0.00019992093972273018,
43
+ "loss": 1.6992,
44
+ "step": 20
45
+ },
46
+ {
47
+ "epoch": 0.14144271570014144,
48
+ "grad_norm": 0.13810234079187747,
49
+ "learning_rate": 0.0001990329466472502,
50
+ "loss": 1.6291,
51
+ "step": 25
52
+ },
53
+ {
54
+ "epoch": 0.16973125884016974,
55
+ "grad_norm": 0.11800201667039435,
56
+ "learning_rate": 0.000197166934004041,
57
+ "loss": 1.6379,
58
+ "step": 30
59
+ },
60
+ {
61
+ "epoch": 0.19801980198019803,
62
+ "grad_norm": 0.11256043626490136,
63
+ "learning_rate": 0.00019434132997221345,
64
+ "loss": 1.5736,
65
+ "step": 35
66
+ },
67
+ {
68
+ "epoch": 0.2263083451202263,
69
+ "grad_norm": 0.10718182461540707,
70
+ "learning_rate": 0.00019058403936655233,
71
+ "loss": 1.5491,
72
+ "step": 40
73
+ },
74
+ {
75
+ "epoch": 0.2545968882602546,
76
+ "grad_norm": 0.112326110136943,
77
+ "learning_rate": 0.00018593216805796612,
78
+ "loss": 1.51,
79
+ "step": 45
80
+ },
81
+ {
82
+ "epoch": 0.2828854314002829,
83
+ "grad_norm": 0.08515607635399414,
84
+ "learning_rate": 0.00018043165652707649,
85
+ "loss": 1.5018,
86
+ "step": 50
87
+ },
88
+ {
89
+ "epoch": 0.31117397454031115,
90
+ "grad_norm": 0.08598762497133733,
91
+ "learning_rate": 0.00017413682616986185,
92
+ "loss": 1.4985,
93
+ "step": 55
94
+ },
95
+ {
96
+ "epoch": 0.33946251768033947,
97
+ "grad_norm": 0.08605574729719911,
98
+ "learning_rate": 0.0001671098428359037,
99
+ "loss": 1.4957,
100
+ "step": 60
101
+ },
102
+ {
103
+ "epoch": 0.36775106082036774,
104
+ "grad_norm": 0.08270099201021953,
105
+ "learning_rate": 0.00015942010289717105,
106
+ "loss": 1.4897,
107
+ "step": 65
108
+ },
109
+ {
110
+ "epoch": 0.39603960396039606,
111
+ "grad_norm": 0.09139373323353543,
112
+ "learning_rate": 0.00015114354791034225,
113
+ "loss": 1.4661,
114
+ "step": 70
115
+ },
116
+ {
117
+ "epoch": 0.4243281471004243,
118
+ "grad_norm": 0.09213917858670279,
119
+ "learning_rate": 0.00014236191464085286,
120
+ "loss": 1.4825,
121
+ "step": 75
122
+ },
123
+ {
124
+ "epoch": 0.4526166902404526,
125
+ "grad_norm": 0.09741017460001462,
126
+ "learning_rate": 0.0001331619278552068,
127
+ "loss": 1.4796,
128
+ "step": 80
129
+ },
130
+ {
131
+ "epoch": 0.4809052333804809,
132
+ "grad_norm": 0.09522302084100907,
133
+ "learning_rate": 0.0001236344438532905,
134
+ "loss": 1.4522,
135
+ "step": 85
136
+ },
137
+ {
138
+ "epoch": 0.5091937765205092,
139
+ "grad_norm": 0.10005523175818275,
140
+ "learning_rate": 0.00011387355319890685,
141
+ "loss": 1.4512,
142
+ "step": 90
143
+ },
144
+ {
145
+ "epoch": 0.5374823196605375,
146
+ "grad_norm": 0.09774894920042008,
147
+ "learning_rate": 0.0001039756515096926,
148
+ "loss": 1.4531,
149
+ "step": 95
150
+ },
151
+ {
152
+ "epoch": 0.5657708628005658,
153
+ "grad_norm": 0.09337090709938024,
154
+ "learning_rate": 9.403848748301802e-05,
155
+ "loss": 1.431,
156
+ "step": 100
157
+ },
158
+ {
159
+ "epoch": 0.594059405940594,
160
+ "grad_norm": 0.10862536785181011,
161
+ "learning_rate": 8.416019755927851e-05,
162
+ "loss": 1.4497,
163
+ "step": 105
164
+ },
165
+ {
166
+ "epoch": 0.6223479490806223,
167
+ "grad_norm": 0.09931981935895794,
168
+ "learning_rate": 7.443833675595255e-05,
169
+ "loss": 1.4277,
170
+ "step": 110
171
+ },
172
+ {
173
+ "epoch": 0.6506364922206507,
174
+ "grad_norm": 0.09784818782800696,
175
+ "learning_rate": 6.496891524361757e-05,
176
+ "loss": 1.4467,
177
+ "step": 115
178
+ },
179
+ {
180
+ "epoch": 0.6789250353606789,
181
+ "grad_norm": 0.09291305679558234,
182
+ "learning_rate": 5.584545017840885e-05,
183
+ "loss": 1.4384,
184
+ "step": 120
185
+ },
186
+ {
187
+ "epoch": 0.7072135785007072,
188
+ "grad_norm": 0.10423529826360275,
189
+ "learning_rate": 4.715804215473809e-05,
190
+ "loss": 1.4476,
191
+ "step": 125
192
+ },
193
+ {
194
+ "epoch": 0.7355021216407355,
195
+ "grad_norm": 0.10820562792911577,
196
+ "learning_rate": 3.899248539894757e-05,
197
+ "loss": 1.4327,
198
+ "step": 130
199
+ },
200
+ {
201
+ "epoch": 0.7637906647807637,
202
+ "grad_norm": 0.09010400983561788,
203
+ "learning_rate": 3.14294204913587e-05,
204
+ "loss": 1.4426,
205
+ "step": 135
206
+ },
207
+ {
208
+ "epoch": 0.7920792079207921,
209
+ "grad_norm": 0.093194757033809,
210
+ "learning_rate": 2.4543537984176978e-05,
211
+ "loss": 1.4434,
212
+ "step": 140
213
+ },
214
+ {
215
+ "epoch": 0.8203677510608204,
216
+ "grad_norm": 0.09797367971916428,
217
+ "learning_rate": 1.840284078008393e-05,
218
+ "loss": 1.4363,
219
+ "step": 145
220
+ },
221
+ {
222
+ "epoch": 0.8486562942008486,
223
+ "grad_norm": 0.09241502492056575,
224
+ "learning_rate": 1.3067972556041752e-05,
225
+ "loss": 1.4347,
226
+ "step": 150
227
+ },
228
+ {
229
+ "epoch": 0.8769448373408769,
230
+ "grad_norm": 0.0929511573879462,
231
+ "learning_rate": 8.59161886459654e-06,
232
+ "loss": 1.4129,
233
+ "step": 155
234
+ },
235
+ {
236
+ "epoch": 0.9052333804809052,
237
+ "grad_norm": 0.08984575236333624,
238
+ "learning_rate": 5.017986827221733e-06,
239
+ "loss": 1.4555,
240
+ "step": 160
241
+ },
242
+ {
243
+ "epoch": 0.9335219236209336,
244
+ "grad_norm": 0.0905814734943296,
245
+ "learning_rate": 2.3823685580949273e-06,
246
+ "loss": 1.4418,
247
+ "step": 165
248
+ },
249
+ {
250
+ "epoch": 0.9618104667609618,
251
+ "grad_norm": 0.08926337770672606,
252
+ "learning_rate": 7.10792629802659e-07,
253
+ "loss": 1.4399,
254
+ "step": 170
255
+ },
256
+ {
257
+ "epoch": 0.9900990099009901,
258
+ "grad_norm": 0.09160289967908435,
259
+ "learning_rate": 1.976702299344435e-08,
260
+ "loss": 1.4333,
261
+ "step": 175
262
+ },
263
+ {
264
+ "epoch": 0.9957567185289957,
265
+ "eval_loss": 1.235967755317688,
266
+ "eval_runtime": 1555.7732,
267
+ "eval_samples_per_second": 8.599,
268
+ "eval_steps_per_second": 0.538,
269
+ "step": 176
270
+ },
271
+ {
272
+ "epoch": 0.9957567185289957,
273
+ "step": 176,
274
+ "total_flos": 2207623390691328.0,
275
+ "train_loss": 1.0380965369668873,
276
+ "train_runtime": 4181.7688,
277
+ "train_samples_per_second": 2.703,
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": 25,
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": 2207623390691328.0,
299
+ "train_batch_size": 4,
300
+ "trial_name": null,
301
+ "trial_params": null
302
+ }