Hugging Face's logo

AmberYifan
/
Llama-3.1-8B-Mistral-7B-v0.3-mix

Text Generation
Transformers
Safetensors
llama
Generated from Trainer
trl
dpo
conversational
text-generation-inference
Inference Endpoints
Model card Files Community
AmberYifan commited on
Commit
0765ff4
·
verified ·
1 Parent(s): 425aea9

Model save

Browse files
Files changed (9) hide show
  1. README.md +68 -0
  2. all_results.json +9 -0
  3. generation_config.json +9 -0
  4. model-00001-of-00004.safetensors +1 -1
  5. model-00002-of-00004.safetensors +1 -1
  6. model-00003-of-00004.safetensors +1 -1
  7. model-00004-of-00004.safetensors +1 -1
  8. train_results.json +9 -0
  9. trainer_state.json +375 -0
README.md ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: AmberYifan/Llama-3.1-8B-sft-ultrachat-safeRLHF
3
+ library_name: transformers
4
+ model_name: Llama-3.1-8B-Mistral-7B-v0.3-mix
5
+ tags:
6
+ - generated_from_trainer
7
+ - trl
8
+ - dpo
9
+ licence: license
10
+ ---
11
+
12
+ # Model Card for Llama-3.1-8B-Mistral-7B-v0.3-mix
13
+
14
+ This model is a fine-tuned version of [AmberYifan/Llama-3.1-8B-sft-ultrachat-safeRLHF](https://huggingface.co/AmberYifan/Llama-3.1-8B-sft-ultrachat-safeRLHF).
15
+ It has been trained using [TRL](https://github.com/huggingface/trl).
16
+
17
+ ## Quick start
18
+
19
+ ```python
20
+ from transformers import pipeline
21
+
22
+ question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
23
+ generator = pipeline("text-generation", model="AmberYifan/Llama-3.1-8B-Mistral-7B-v0.3-mix", device="cuda")
24
+ output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
25
+ print(output["generated_text"])
26
+ ```
27
+
28
+ ## Training procedure
29
+
30
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/yifanwang/huggingface/runs/wo2kk71p)
31
+
32
+ This model was trained with DPO, a method introduced in [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://huggingface.co/papers/2305.18290).
33
+
34
+ ### Framework versions
35
+
36
+ - TRL: 0.12.2
37
+ - Transformers: 4.46.3
38
+ - Pytorch: 2.5.1+cu118
39
+ - Datasets: 3.2.0
40
+ - Tokenizers: 0.20.3
41
+
42
+ ## Citations
43
+
44
+ Cite DPO as:
45
+
46
+ ```bibtex
47
+ @inproceedings{rafailov2023direct,
48
+ title = {{Direct Preference Optimization: Your Language Model is Secretly a Reward Model}},
49
+ author = {Rafael Rafailov and Archit Sharma and Eric Mitchell and Christopher D. Manning and Stefano Ermon and Chelsea Finn},
50
+ year = 2023,
51
+ booktitle = {Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023},
52
+ url = {http://papers.nips.cc/paper_files/paper/2023/hash/a85b405ed65c6477a4fe8302b5e06ce7-Abstract-Conference.html},
53
+ editor = {Alice Oh and Tristan Naumann and Amir Globerson and Kate Saenko and Moritz Hardt and Sergey Levine},
54
+ }
55
+ ```
56
+
57
+ Cite TRL as:
58
+
59
+ ```bibtex
60
+ @misc{vonwerra2022trl,
61
+ title = {{TRL: Transformer Reinforcement Learning}},
62
+ author = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallouédec},
63
+ year = 2020,
64
+ journal = {GitHub repository},
65
+ publisher = {GitHub},
66
+ howpublished = {\url{https://github.com/huggingface/trl}}
67
+ }
68
+ ```
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 0.0,
4
+ "train_loss": 0.6665452422288359,
5
+ "train_runtime": 1664.0362,
6
+ "train_samples": 2000,
7
+ "train_samples_per_second": 3.606,
8
+ "train_steps_per_second": 0.114
9
+ }
generation_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 128000,
4
+ "do_sample": true,
5
+ "eos_token_id": 128001,
6
+ "temperature": 0.6,
7
+ "top_p": 0.9,
8
+ "transformers_version": "4.46.3"
9
+ }
model-00001-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ade34906b4cbdc902daa864eccbadf2a8cd52b90a0c82c00e5ea078e8ec76711
3
  size 4976698672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa158e478c1353a76cf54ca4154ab3238f1afe88f41549329c493476e3f3f912
3
  size 4976698672
model-00002-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:412ac5e7b41b49f159f1ab0e79fec711d071236d0c039b9213a6e0e970bac99c
3
  size 4999802720
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c355be4861bcc485847ddabf00ae7d5768edf3222398ac72bdd48e4c44da5375
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:cf0acadd41f37958f2d32c0e85d94b5f67f43915a39704444b1b9040fa49ec51
3
  size 4915916176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2d7b8aa906fec12c40d62cbddc56eb2bdbeccbebb3f7155ef847cca73f8bebf
3
  size 4915916176
model-00004-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:75cb0755a39def127f63de3fa5df3890ca0820f19b74340bb4adcb75c6f7b9f9
3
  size 1168138808
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:99c4cc43631de6e10ca54c0c8cb160a03f2bf9d2faf671b5e6ac68a9421d929c
3
  size 1168138808
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "total_flos": 0.0,
4
+ "train_loss": 0.6665452422288359,
5
+ "train_runtime": 1664.0362,
6
+ "train_samples": 2000,
7
+ "train_samples_per_second": 3.606,
8
+ "train_steps_per_second": 0.114
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,375 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.0,
5
+ "eval_steps": 500,
6
+ "global_step": 189,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.015873015873015872,
13
+ "grad_norm": 2.956210802841563,
14
+ "learning_rate": 2.6315789473684208e-08,
15
+ "logits/chosen": -0.84765625,
16
+ "logits/rejected": -0.84765625,
17
+ "logps/chosen": -14.875,
18
+ "logps/rejected": -14.875,
19
+ "loss": 0.6914,
20
+ "rewards/accuracies": 0.0,
21
+ "rewards/chosen": 0.0,
22
+ "rewards/margins": 0.0,
23
+ "rewards/rejected": 0.0,
24
+ "step": 1
25
+ },
26
+ {
27
+ "epoch": 0.15873015873015872,
28
+ "grad_norm": 0.0026515697903897284,
29
+ "learning_rate": 2.631578947368421e-07,
30
+ "logits/chosen": -0.8828125,
31
+ "logits/rejected": -0.88671875,
32
+ "logps/chosen": -15.9375,
33
+ "logps/rejected": -15.8125,
34
+ "loss": 0.6914,
35
+ "rewards/accuracies": 0.02777777798473835,
36
+ "rewards/chosen": 0.0008697509765625,
37
+ "rewards/margins": -0.00043487548828125,
38
+ "rewards/rejected": 0.00130462646484375,
39
+ "step": 10
40
+ },
41
+ {
42
+ "epoch": 0.31746031746031744,
43
+ "grad_norm": 2.201768075236467,
44
+ "learning_rate": 4.970588235294118e-07,
45
+ "logits/chosen": -0.890625,
46
+ "logits/rejected": -0.88671875,
47
+ "logps/chosen": -15.625,
48
+ "logps/rejected": -15.5625,
49
+ "loss": 0.6894,
50
+ "rewards/accuracies": 0.07500000298023224,
51
+ "rewards/chosen": 0.0228271484375,
52
+ "rewards/margins": 0.00860595703125,
53
+ "rewards/rejected": 0.01422119140625,
54
+ "step": 20
55
+ },
56
+ {
57
+ "epoch": 0.47619047619047616,
58
+ "grad_norm": 2.7819271307964892,
59
+ "learning_rate": 4.676470588235294e-07,
60
+ "logits/chosen": -0.890625,
61
+ "logits/rejected": -0.875,
62
+ "logps/chosen": -14.9375,
63
+ "logps/rejected": -14.75,
64
+ "loss": 0.689,
65
+ "rewards/accuracies": 0.0625,
66
+ "rewards/chosen": 0.0830078125,
67
+ "rewards/margins": 0.0111083984375,
68
+ "rewards/rejected": 0.07177734375,
69
+ "step": 30
70
+ },
71
+ {
72
+ "epoch": 0.6349206349206349,
73
+ "grad_norm": 2.468434193598462,
74
+ "learning_rate": 4.38235294117647e-07,
75
+ "logits/chosen": -0.8984375,
76
+ "logits/rejected": -0.91015625,
77
+ "logps/chosen": -14.375,
78
+ "logps/rejected": -14.8125,
79
+ "loss": 0.6849,
80
+ "rewards/accuracies": 0.11249999701976776,
81
+ "rewards/chosen": 0.1552734375,
82
+ "rewards/margins": 0.038330078125,
83
+ "rewards/rejected": 0.11669921875,
84
+ "step": 40
85
+ },
86
+ {
87
+ "epoch": 0.7936507936507936,
88
+ "grad_norm": 3.217396320056085,
89
+ "learning_rate": 4.0882352941176465e-07,
90
+ "logits/chosen": -0.87109375,
91
+ "logits/rejected": -0.8828125,
92
+ "logps/chosen": -13.125,
93
+ "logps/rejected": -12.625,
94
+ "loss": 0.6828,
95
+ "rewards/accuracies": 0.03750000149011612,
96
+ "rewards/chosen": 0.267578125,
97
+ "rewards/margins": 0.01129150390625,
98
+ "rewards/rejected": 0.255859375,
99
+ "step": 50
100
+ },
101
+ {
102
+ "epoch": 0.9523809523809523,
103
+ "grad_norm": 3.5576289077982315,
104
+ "learning_rate": 3.7941176470588235e-07,
105
+ "logits/chosen": -0.85546875,
106
+ "logits/rejected": -0.8671875,
107
+ "logps/chosen": -12.125,
108
+ "logps/rejected": -12.625,
109
+ "loss": 0.6772,
110
+ "rewards/accuracies": 0.02500000037252903,
111
+ "rewards/chosen": 0.3359375,
112
+ "rewards/margins": 0.0196533203125,
113
+ "rewards/rejected": 0.31640625,
114
+ "step": 60
115
+ },
116
+ {
117
+ "epoch": 1.0,
118
+ "eval_logits/chosen": -0.875,
119
+ "eval_logits/rejected": -0.87109375,
120
+ "eval_logps/chosen": -10.25,
121
+ "eval_logps/rejected": -10.625,
122
+ "eval_loss": 0.6818749904632568,
123
+ "eval_rewards/accuracies": 0.0357142873108387,
124
+ "eval_rewards/chosen": 0.458984375,
125
+ "eval_rewards/margins": -0.0004177093505859375,
126
+ "eval_rewards/rejected": 0.4609375,
127
+ "eval_runtime": 12.3547,
128
+ "eval_samples_per_second": 16.188,
129
+ "eval_steps_per_second": 0.567,
130
+ "step": 63
131
+ },
132
+ {
133
+ "epoch": 1.1111111111111112,
134
+ "grad_norm": 0.6480653257577673,
135
+ "learning_rate": 3.5e-07,
136
+ "logits/chosen": -0.83984375,
137
+ "logits/rejected": -0.8515625,
138
+ "logps/chosen": -9.8125,
139
+ "logps/rejected": -11.0,
140
+ "loss": 0.6654,
141
+ "rewards/accuracies": 0.0625,
142
+ "rewards/chosen": 0.52734375,
143
+ "rewards/margins": 0.05859375,
144
+ "rewards/rejected": 0.466796875,
145
+ "step": 70
146
+ },
147
+ {
148
+ "epoch": 1.2698412698412698,
149
+ "grad_norm": 1.8542768080035792,
150
+ "learning_rate": 3.205882352941177e-07,
151
+ "logits/chosen": -0.828125,
152
+ "logits/rejected": -0.83203125,
153
+ "logps/chosen": -10.1875,
154
+ "logps/rejected": -11.0,
155
+ "loss": 0.6656,
156
+ "rewards/accuracies": 0.03750000149011612,
157
+ "rewards/chosen": 0.49609375,
158
+ "rewards/margins": 0.05419921875,
159
+ "rewards/rejected": 0.44140625,
160
+ "step": 80
161
+ },
162
+ {
163
+ "epoch": 1.4285714285714286,
164
+ "grad_norm": 1.4880802722309492,
165
+ "learning_rate": 2.911764705882353e-07,
166
+ "logits/chosen": -0.83984375,
167
+ "logits/rejected": -0.84765625,
168
+ "logps/chosen": -10.6875,
169
+ "logps/rejected": -11.5,
170
+ "loss": 0.6672,
171
+ "rewards/accuracies": 0.10000000149011612,
172
+ "rewards/chosen": 0.474609375,
173
+ "rewards/margins": 0.10400390625,
174
+ "rewards/rejected": 0.37109375,
175
+ "step": 90
176
+ },
177
+ {
178
+ "epoch": 1.5873015873015874,
179
+ "grad_norm": 0.9541857190755878,
180
+ "learning_rate": 2.6176470588235295e-07,
181
+ "logits/chosen": -0.84765625,
182
+ "logits/rejected": -0.84765625,
183
+ "logps/chosen": -12.3125,
184
+ "logps/rejected": -13.875,
185
+ "loss": 0.667,
186
+ "rewards/accuracies": 0.08749999850988388,
187
+ "rewards/chosen": 0.34765625,
188
+ "rewards/margins": 0.1240234375,
189
+ "rewards/rejected": 0.224609375,
190
+ "step": 100
191
+ },
192
+ {
193
+ "epoch": 1.746031746031746,
194
+ "grad_norm": 1.3739568831557003,
195
+ "learning_rate": 2.323529411764706e-07,
196
+ "logits/chosen": -0.875,
197
+ "logits/rejected": -0.8671875,
198
+ "logps/chosen": -12.0,
199
+ "logps/rejected": -13.375,
200
+ "loss": 0.6519,
201
+ "rewards/accuracies": 0.10000000149011612,
202
+ "rewards/chosen": 0.419921875,
203
+ "rewards/margins": 0.1884765625,
204
+ "rewards/rejected": 0.23046875,
205
+ "step": 110
206
+ },
207
+ {
208
+ "epoch": 1.9047619047619047,
209
+ "grad_norm": 1.2013802625805676,
210
+ "learning_rate": 2.0294117647058823e-07,
211
+ "logits/chosen": -0.8359375,
212
+ "logits/rejected": -0.84375,
213
+ "logps/chosen": -12.25,
214
+ "logps/rejected": -13.4375,
215
+ "loss": 0.6564,
216
+ "rewards/accuracies": 0.07500000298023224,
217
+ "rewards/chosen": 0.36328125,
218
+ "rewards/margins": 0.10498046875,
219
+ "rewards/rejected": 0.2578125,
220
+ "step": 120
221
+ },
222
+ {
223
+ "epoch": 2.0,
224
+ "eval_logits/chosen": -0.85546875,
225
+ "eval_logits/rejected": -0.85546875,
226
+ "eval_logps/chosen": -11.9375,
227
+ "eval_logps/rejected": -12.5,
228
+ "eval_loss": 0.6790624856948853,
229
+ "eval_rewards/accuracies": 0.0357142873108387,
230
+ "eval_rewards/chosen": 0.2890625,
231
+ "eval_rewards/margins": 0.01324462890625,
232
+ "eval_rewards/rejected": 0.275390625,
233
+ "eval_runtime": 14.8138,
234
+ "eval_samples_per_second": 13.501,
235
+ "eval_steps_per_second": 0.473,
236
+ "step": 126
237
+ },
238
+ {
239
+ "epoch": 2.0634920634920633,
240
+ "grad_norm": 1.6810359268654005,
241
+ "learning_rate": 1.7352941176470587e-07,
242
+ "logits/chosen": -0.8359375,
243
+ "logits/rejected": -0.8359375,
244
+ "logps/chosen": -13.0,
245
+ "logps/rejected": -15.0,
246
+ "loss": 0.6466,
247
+ "rewards/accuracies": 0.08749999850988388,
248
+ "rewards/chosen": 0.255859375,
249
+ "rewards/margins": 0.1630859375,
250
+ "rewards/rejected": 0.09228515625,
251
+ "step": 130
252
+ },
253
+ {
254
+ "epoch": 2.2222222222222223,
255
+ "grad_norm": 1.6049829222017113,
256
+ "learning_rate": 1.441176470588235e-07,
257
+ "logits/chosen": -0.83984375,
258
+ "logits/rejected": -0.83984375,
259
+ "logps/chosen": -13.1875,
260
+ "logps/rejected": -13.0,
261
+ "loss": 0.6546,
262
+ "rewards/accuracies": 0.08749999850988388,
263
+ "rewards/chosen": 0.30859375,
264
+ "rewards/margins": 0.1064453125,
265
+ "rewards/rejected": 0.203125,
266
+ "step": 140
267
+ },
268
+ {
269
+ "epoch": 2.380952380952381,
270
+ "grad_norm": 0.8745967010917558,
271
+ "learning_rate": 1.1470588235294116e-07,
272
+ "logits/chosen": -0.85546875,
273
+ "logits/rejected": -0.84375,
274
+ "logps/chosen": -12.1875,
275
+ "logps/rejected": -13.1875,
276
+ "loss": 0.647,
277
+ "rewards/accuracies": 0.08749999850988388,
278
+ "rewards/chosen": 0.28125,
279
+ "rewards/margins": 0.099609375,
280
+ "rewards/rejected": 0.1806640625,
281
+ "step": 150
282
+ },
283
+ {
284
+ "epoch": 2.5396825396825395,
285
+ "grad_norm": 2.5161816747376817,
286
+ "learning_rate": 8.529411764705883e-08,
287
+ "logits/chosen": -0.87109375,
288
+ "logits/rejected": -0.87890625,
289
+ "logps/chosen": -13.625,
290
+ "logps/rejected": -15.3125,
291
+ "loss": 0.6455,
292
+ "rewards/accuracies": 0.08749999850988388,
293
+ "rewards/chosen": 0.2734375,
294
+ "rewards/margins": 0.1630859375,
295
+ "rewards/rejected": 0.11083984375,
296
+ "step": 160
297
+ },
298
+ {
299
+ "epoch": 2.6984126984126986,
300
+ "grad_norm": 1.779552803700363,
301
+ "learning_rate": 5.5882352941176474e-08,
302
+ "logits/chosen": -0.8203125,
303
+ "logits/rejected": -0.83203125,
304
+ "logps/chosen": -12.1875,
305
+ "logps/rejected": -13.75,
306
+ "loss": 0.6576,
307
+ "rewards/accuracies": 0.05000000074505806,
308
+ "rewards/chosen": 0.2890625,
309
+ "rewards/margins": 0.11328125,
310
+ "rewards/rejected": 0.17578125,
311
+ "step": 170
312
+ },
313
+ {
314
+ "epoch": 2.857142857142857,
315
+ "grad_norm": 1.0681708171714699,
316
+ "learning_rate": 2.6470588235294116e-08,
317
+ "logits/chosen": -0.80859375,
318
+ "logits/rejected": -0.796875,
319
+ "logps/chosen": -13.0625,
320
+ "logps/rejected": -14.375,
321
+ "loss": 0.6566,
322
+ "rewards/accuracies": 0.07500000298023224,
323
+ "rewards/chosen": 0.2294921875,
324
+ "rewards/margins": 0.09619140625,
325
+ "rewards/rejected": 0.1328125,
326
+ "step": 180
327
+ },
328
+ {
329
+ "epoch": 3.0,
330
+ "eval_logits/chosen": -0.85546875,
331
+ "eval_logits/rejected": -0.8515625,
332
+ "eval_logps/chosen": -12.375,
333
+ "eval_logps/rejected": -12.9375,
334
+ "eval_loss": 0.6785937547683716,
335
+ "eval_rewards/accuracies": 0.0357142873108387,
336
+ "eval_rewards/chosen": 0.2470703125,
337
+ "eval_rewards/margins": 0.016845703125,
338
+ "eval_rewards/rejected": 0.23046875,
339
+ "eval_runtime": 14.6047,
340
+ "eval_samples_per_second": 13.694,
341
+ "eval_steps_per_second": 0.479,
342
+ "step": 189
343
+ },
344
+ {
345
+ "epoch": 3.0,
346
+ "step": 189,
347
+ "total_flos": 0.0,
348
+ "train_loss": 0.6665452422288359,
349
+ "train_runtime": 1664.0362,
350
+ "train_samples_per_second": 3.606,
351
+ "train_steps_per_second": 0.114
352
+ }
353
+ ],
354
+ "logging_steps": 10,
355
+ "max_steps": 189,
356
+ "num_input_tokens_seen": 0,
357
+ "num_train_epochs": 3,
358
+ "save_steps": 500,
359
+ "stateful_callbacks": {
360
+ "TrainerControl": {
361
+ "args": {
362
+ "should_epoch_stop": false,
363
+ "should_evaluate": false,
364
+ "should_log": false,
365
+ "should_save": true,
366
+ "should_training_stop": true
367
+ },
368
+ "attributes": {}
369
+ }
370
+ },
371
+ "total_flos": 0.0,
372
+ "train_batch_size": 8,
373
+ "trial_name": null,
374
+ "trial_params": null
375
+ }