Hugging Face's logo

ZhangShenao
/
math_gsm-Meta-Llama-3-8B-Instruct-msft-sample_7473_tp

Text Generation
Transformers
TensorBoard
Safetensors
llama
trl
sft
Generated from Trainer
conversational
text-generation-inference
Inference Endpoints
Model card Files Metrics Community
ZhangShenao commited on
Commit
d528f1c
·
verified ·
1 Parent(s): 3b9cede

Model save

Browse files
Files changed (15) hide show
  1. README.md +4 -4
  2. all_results.json +6 -6
  3. final_checkpoint/model-00001-of-00004.safetensors +1 -1
  4. final_checkpoint/model-00002-of-00004.safetensors +1 -1
  5. final_checkpoint/model-00003-of-00004.safetensors +1 -1
  6. final_checkpoint/model-00004-of-00004.safetensors +1 -1
  7. final_checkpoint/training_args.bin +1 -1
  8. model-00001-of-00004.safetensors +1 -1
  9. model-00002-of-00004.safetensors +1 -1
  10. model-00003-of-00004.safetensors +1 -1
  11. model-00004-of-00004.safetensors +1 -1
  12. runs/Jan07_07-25-38_n124-173-214/events.out.tfevents.1736234759.n124-173-214.63639.0 +3 -0
  13. train_results.json +6 -6
  14. trainer_state.json +207 -32
  15. training_args.bin +1 -1
README.md CHANGED
@@ -42,14 +42,14 @@ The following hyperparameters were used during training:
42
  - eval_batch_size: 2
43
  - seed: 42
44
  - distributed_type: multi-GPU
45
- - num_devices: 4
46
  - gradient_accumulation_steps: 4
47
- - total_train_batch_size: 128
48
- - total_eval_batch_size: 8
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: cosine
51
  - lr_scheduler_warmup_ratio: 0.1
52
- - num_epochs: 1
53
 
54
  ### Training results
55
 
 
42
  - eval_batch_size: 2
43
  - seed: 42
44
  - distributed_type: multi-GPU
45
+ - num_devices: 3
46
  - gradient_accumulation_steps: 4
47
+ - total_train_batch_size: 96
48
+ - total_eval_batch_size: 6
49
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
  - lr_scheduler_type: cosine
51
  - lr_scheduler_warmup_ratio: 0.1
52
+ - num_epochs: 5
53
 
54
  ### Training results
55
 
all_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 1.0,
3
- "total_flos": 1989106728960.0,
4
- "train_loss": 1.7512740612030029,
5
- "train_runtime": 60.9224,
6
  "train_samples": 7473,
7
- "train_samples_per_second": 10.144,
8
- "train_steps_per_second": 0.082
9
  }
 
1
  {
2
+ "epoch": 4.615384615384615,
3
+ "total_flos": 9343567134720.0,
4
+ "train_loss": 1.0503198325634002,
5
+ "train_runtime": 473.9669,
6
  "train_samples": 7473,
7
+ "train_samples_per_second": 6.519,
8
+ "train_steps_per_second": 0.063
9
  }
final_checkpoint/model-00001-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6b2bc63bbf10dc034b8cecbbfd40d09b2f91128d5ae56972ea6ee4bbd867b8df
3
  size 4976698672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca64b968a06f69bcc06bd2d890b91670134b547700a758c879e4088eee568d1c
3
  size 4976698672
final_checkpoint/model-00002-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8fdc7cdc32157d0747579613f0840990d725486a32810803b22f51239189dd7f
3
  size 4999802720
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a8694067ac64052128f791d7ff03286ca98766b71ff57ce6ed9aa134a1b50d6
3
  size 4999802720
final_checkpoint/model-00003-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:091b911c6118350ed6ae9ec1fdc5a81eeceee0d503d13a85b58a2cd91c3b6754
3
  size 4915916176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4816d8cdb34b70df0c6b8eeab67571315b017bc2eac9e1ed00543d94bf3dda7
3
  size 4915916176
final_checkpoint/model-00004-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4fae73678956b4e086421f0509862dee6fcede4a04b02dec92d76ea30d0d8403
3
  size 1168138808
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85a008e424c01a32b5a5e132ddf3ba1106e787cace0b85304b8f2e9b0633df6c
3
  size 1168138808
final_checkpoint/training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a45fdecee2292ca7ccd22f2df90b781cd768df49eb485c5ffaf115365ed11511
3
  size 7608
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c33a19196bfe2bad2613a328bf8b5d72e314960f54e408ef1737430f48f21b92
3
  size 7608
model-00001-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6b2bc63bbf10dc034b8cecbbfd40d09b2f91128d5ae56972ea6ee4bbd867b8df
3
  size 4976698672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ca64b968a06f69bcc06bd2d890b91670134b547700a758c879e4088eee568d1c
3
  size 4976698672
model-00002-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8fdc7cdc32157d0747579613f0840990d725486a32810803b22f51239189dd7f
3
  size 4999802720
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9a8694067ac64052128f791d7ff03286ca98766b71ff57ce6ed9aa134a1b50d6
3
  size 4999802720
model-00003-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:091b911c6118350ed6ae9ec1fdc5a81eeceee0d503d13a85b58a2cd91c3b6754
3
  size 4915916176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4816d8cdb34b70df0c6b8eeab67571315b017bc2eac9e1ed00543d94bf3dda7
3
  size 4915916176
model-00004-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:4fae73678956b4e086421f0509862dee6fcede4a04b02dec92d76ea30d0d8403
3
  size 1168138808
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85a008e424c01a32b5a5e132ddf3ba1106e787cace0b85304b8f2e9b0633df6c
3
  size 1168138808
runs/Jan07_07-25-38_n124-173-214/events.out.tfevents.1736234759.n124-173-214.63639.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:84cb54523ae103dcd2bd4f6d9e528cb82183c7c9368b48db3e4f5582b9c366af
3
+ size 12216
train_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 1.0,
3
- "total_flos": 1989106728960.0,
4
- "train_loss": 1.7512740612030029,
5
- "train_runtime": 60.9224,
6
  "train_samples": 7473,
7
- "train_samples_per_second": 10.144,
8
- "train_steps_per_second": 0.082
9
  }
 
1
  {
2
+ "epoch": 4.615384615384615,
3
+ "total_flos": 9343567134720.0,
4
+ "train_loss": 1.0503198325634002,
5
+ "train_runtime": 473.9669,
6
  "train_samples": 7473,
7
+ "train_samples_per_second": 6.519,
8
+ "train_steps_per_second": 0.063
9
  }
trainer_state.json CHANGED
@@ -1,62 +1,237 @@
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
- "epoch": 1.0,
5
  "eval_steps": 500,
6
- "global_step": 5,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
- "epoch": 0.2,
13
- "grad_norm": 44.71906133277917,
14
- "learning_rate": 2e-05,
15
- "loss": 1.7586,
16
  "step": 1
17
  },
18
  {
19
- "epoch": 0.4,
20
- "grad_norm": 46.584378462644494,
21
- "learning_rate": 1.7071067811865477e-05,
22
- "loss": 1.7608,
23
  "step": 2
24
  },
25
  {
26
- "epoch": 0.6,
27
- "grad_norm": 46.80362412165293,
28
- "learning_rate": 1e-05,
29
- "loss": 2.0284,
30
  "step": 3
31
  },
32
  {
33
- "epoch": 0.8,
34
- "grad_norm": 46.9343362057302,
35
- "learning_rate": 2.9289321881345257e-06,
36
- "loss": 1.8336,
37
  "step": 4
38
  },
39
  {
40
- "epoch": 1.0,
41
- "grad_norm": 6.925820644312154,
42
- "learning_rate": 0.0,
43
- "loss": 1.375,
44
  "step": 5
45
  },
46
  {
47
- "epoch": 1.0,
48
- "step": 5,
49
- "total_flos": 1989106728960.0,
50
- "train_loss": 1.7512740612030029,
51
- "train_runtime": 60.9224,
52
- "train_samples_per_second": 10.144,
53
- "train_steps_per_second": 0.082
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
54
  }
55
  ],
56
  "logging_steps": 1,
57
- "max_steps": 5,
58
  "num_input_tokens_seen": 0,
59
- "num_train_epochs": 1,
60
  "save_steps": 999999,
61
  "stateful_callbacks": {
62
  "TrainerControl": {
@@ -70,7 +245,7 @@
70
  "attributes": {}
71
  }
72
  },
73
- "total_flos": 1989106728960.0,
74
  "train_batch_size": 8,
75
  "trial_name": null,
76
  "trial_params": null
 
1
  {
2
  "best_metric": null,
3
  "best_model_checkpoint": null,
4
+ "epoch": 4.615384615384615,
5
  "eval_steps": 500,
6
+ "global_step": 30,
7
  "is_hyper_param_search": false,
8
  "is_local_process_zero": true,
9
  "is_world_process_zero": true,
10
  "log_history": [
11
  {
12
+ "epoch": 0.15384615384615385,
13
+ "grad_norm": 45.207840048303645,
14
+ "learning_rate": 6.666666666666667e-06,
15
+ "loss": 1.7659,
16
  "step": 1
17
  },
18
  {
19
+ "epoch": 0.3076923076923077,
20
+ "grad_norm": 45.675403154272956,
21
+ "learning_rate": 1.3333333333333333e-05,
22
+ "loss": 1.7642,
23
  "step": 2
24
  },
25
  {
26
+ "epoch": 0.46153846153846156,
27
+ "grad_norm": 20.868641882690532,
28
+ "learning_rate": 2e-05,
29
+ "loss": 1.3575,
30
  "step": 3
31
  },
32
  {
33
+ "epoch": 0.6153846153846154,
34
+ "grad_norm": 34.06850705215301,
35
+ "learning_rate": 1.9932383577419432e-05,
36
+ "loss": 1.9345,
37
  "step": 4
38
  },
39
  {
40
+ "epoch": 0.7692307692307693,
41
+ "grad_norm": 37.41867508840468,
42
+ "learning_rate": 1.973044870579824e-05,
43
+ "loss": 1.6787,
44
  "step": 5
45
  },
46
  {
47
+ "epoch": 0.9230769230769231,
48
+ "grad_norm": 53.46088058831564,
49
+ "learning_rate": 1.9396926207859085e-05,
50
+ "loss": 1.488,
51
+ "step": 6
52
+ },
53
+ {
54
+ "epoch": 1.0769230769230769,
55
+ "grad_norm": 7.4627580332441275,
56
+ "learning_rate": 1.8936326403234125e-05,
57
+ "loss": 1.2373,
58
+ "step": 7
59
+ },
60
+ {
61
+ "epoch": 1.2307692307692308,
62
+ "grad_norm": 3.9518851165618223,
63
+ "learning_rate": 1.8354878114129368e-05,
64
+ "loss": 1.0725,
65
+ "step": 8
66
+ },
67
+ {
68
+ "epoch": 1.3846153846153846,
69
+ "grad_norm": 7.177914493789473,
70
+ "learning_rate": 1.766044443118978e-05,
71
+ "loss": 1.0964,
72
+ "step": 9
73
+ },
74
+ {
75
+ "epoch": 1.5384615384615383,
76
+ "grad_norm": 5.589669105212754,
77
+ "learning_rate": 1.686241637868734e-05,
78
+ "loss": 1.0627,
79
+ "step": 10
80
+ },
81
+ {
82
+ "epoch": 1.6923076923076923,
83
+ "grad_norm": 5.081123298359917,
84
+ "learning_rate": 1.5971585917027864e-05,
85
+ "loss": 1.0228,
86
+ "step": 11
87
+ },
88
+ {
89
+ "epoch": 1.8461538461538463,
90
+ "grad_norm": 5.35114385581476,
91
+ "learning_rate": 1.5000000000000002e-05,
92
+ "loss": 1.0027,
93
+ "step": 12
94
+ },
95
+ {
96
+ "epoch": 2.0,
97
+ "grad_norm": 1.8251196525450355,
98
+ "learning_rate": 1.396079766039157e-05,
99
+ "loss": 0.9503,
100
+ "step": 13
101
+ },
102
+ {
103
+ "epoch": 2.1538461538461537,
104
+ "grad_norm": 2.8409999360336515,
105
+ "learning_rate": 1.2868032327110904e-05,
106
+ "loss": 0.9229,
107
+ "step": 14
108
+ },
109
+ {
110
+ "epoch": 2.3076923076923075,
111
+ "grad_norm": 1.5687190110468443,
112
+ "learning_rate": 1.1736481776669307e-05,
113
+ "loss": 0.9072,
114
+ "step": 15
115
+ },
116
+ {
117
+ "epoch": 2.4615384615384617,
118
+ "grad_norm": 2.0183470286877503,
119
+ "learning_rate": 1.0581448289104759e-05,
120
+ "loss": 0.9059,
121
+ "step": 16
122
+ },
123
+ {
124
+ "epoch": 2.6153846153846154,
125
+ "grad_norm": 1.5424054889975665,
126
+ "learning_rate": 9.418551710895243e-06,
127
+ "loss": 0.8844,
128
+ "step": 17
129
+ },
130
+ {
131
+ "epoch": 2.769230769230769,
132
+ "grad_norm": 1.4748838136546052,
133
+ "learning_rate": 8.263518223330698e-06,
134
+ "loss": 0.8603,
135
+ "step": 18
136
+ },
137
+ {
138
+ "epoch": 2.9230769230769234,
139
+ "grad_norm": 1.2236878389423478,
140
+ "learning_rate": 7.131967672889101e-06,
141
+ "loss": 0.8656,
142
+ "step": 19
143
+ },
144
+ {
145
+ "epoch": 3.076923076923077,
146
+ "grad_norm": 1.1955478413282419,
147
+ "learning_rate": 6.039202339608432e-06,
148
+ "loss": 0.8482,
149
+ "step": 20
150
+ },
151
+ {
152
+ "epoch": 3.230769230769231,
153
+ "grad_norm": 0.7738899384102409,
154
+ "learning_rate": 5.000000000000003e-06,
155
+ "loss": 0.8156,
156
+ "step": 21
157
+ },
158
+ {
159
+ "epoch": 3.3846153846153846,
160
+ "grad_norm": 1.0901149643020054,
161
+ "learning_rate": 4.028414082972141e-06,
162
+ "loss": 0.8103,
163
+ "step": 22
164
+ },
165
+ {
166
+ "epoch": 3.5384615384615383,
167
+ "grad_norm": 0.8933210291425059,
168
+ "learning_rate": 3.1375836213126653e-06,
169
+ "loss": 0.7943,
170
+ "step": 23
171
+ },
172
+ {
173
+ "epoch": 3.6923076923076925,
174
+ "grad_norm": 0.6465350340912998,
175
+ "learning_rate": 2.339555568810221e-06,
176
+ "loss": 0.801,
177
+ "step": 24
178
+ },
179
+ {
180
+ "epoch": 3.8461538461538463,
181
+ "grad_norm": 0.7231924138082919,
182
+ "learning_rate": 1.6451218858706374e-06,
183
+ "loss": 0.7893,
184
+ "step": 25
185
+ },
186
+ {
187
+ "epoch": 4.0,
188
+ "grad_norm": 0.7957272625370657,
189
+ "learning_rate": 1.0636735967658785e-06,
190
+ "loss": 0.7955,
191
+ "step": 26
192
+ },
193
+ {
194
+ "epoch": 4.153846153846154,
195
+ "grad_norm": 0.7102065034979811,
196
+ "learning_rate": 6.030737921409169e-07,
197
+ "loss": 0.7631,
198
+ "step": 27
199
+ },
200
+ {
201
+ "epoch": 4.3076923076923075,
202
+ "grad_norm": 0.7243607288304533,
203
+ "learning_rate": 2.6955129420176193e-07,
204
+ "loss": 0.7785,
205
+ "step": 28
206
+ },
207
+ {
208
+ "epoch": 4.461538461538462,
209
+ "grad_norm": 0.687719390104318,
210
+ "learning_rate": 6.761642258056977e-08,
211
+ "loss": 0.7678,
212
+ "step": 29
213
+ },
214
+ {
215
+ "epoch": 4.615384615384615,
216
+ "grad_norm": 0.6115431503277491,
217
+ "learning_rate": 0.0,
218
+ "loss": 0.7664,
219
+ "step": 30
220
+ },
221
+ {
222
+ "epoch": 4.615384615384615,
223
+ "step": 30,
224
+ "total_flos": 9343567134720.0,
225
+ "train_loss": 1.0503198325634002,
226
+ "train_runtime": 473.9669,
227
+ "train_samples_per_second": 6.519,
228
+ "train_steps_per_second": 0.063
229
  }
230
  ],
231
  "logging_steps": 1,
232
+ "max_steps": 30,
233
  "num_input_tokens_seen": 0,
234
+ "num_train_epochs": 5,
235
  "save_steps": 999999,
236
  "stateful_callbacks": {
237
  "TrainerControl": {
 
245
  "attributes": {}
246
  }
247
  },
248
+ "total_flos": 9343567134720.0,
249
  "train_batch_size": 8,
250
  "trial_name": null,
251
  "trial_params": null
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a45fdecee2292ca7ccd22f2df90b781cd768df49eb485c5ffaf115365ed11511
3
  size 7608
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c33a19196bfe2bad2613a328bf8b5d72e314960f54e408ef1737430f48f21b92
3
  size 7608