yiran-wang3 commited on
Commit
87abac9
1 Parent(s): 32d0ccf

End of training

Browse files
README.md ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ license: apache-2.0
4
+ base_model: yiran-wang3/qwen2_coder_adamw_iter4
5
+ tags:
6
+ - alignment-handbook
7
+ - generated_from_trainer
8
+ - trl
9
+ - dpo
10
+ datasets:
11
+ - self-generate/qwcoder2_sppo_hard_new_cn_mining_oj_iter4-binarized
12
+ model-index:
13
+ - name: qwen2_coder_adamw_iter5
14
+ results: []
15
+ ---
16
+
17
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
18
+ should probably proofread and complete it, then remove this comment. -->
19
+
20
+ # qwen2_coder_adamw_iter5
21
+
22
+ This model is a fine-tuned version of [yiran-wang3/qwen2_coder_adamw_iter4](https://huggingface.co/yiran-wang3/qwen2_coder_adamw_iter4) on the self-generate/qwcoder2_sppo_hard_new_cn_mining_oj_iter4-binarized dataset.
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-06
42
+ - train_batch_size: 8
43
+ - eval_batch_size: 4
44
+ - seed: 42
45
+ - distributed_type: multi-GPU
46
+ - num_devices: 8
47
+ - total_train_batch_size: 64
48
+ - total_eval_batch_size: 32
49
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
50
+ - lr_scheduler_type: constant
51
+ - lr_scheduler_warmup_ratio: 0.1
52
+ - lr_scheduler_warmup_steps: 100
53
+ - num_epochs: 1.0
54
+
55
+ ### Training results
56
+
57
+
58
+
59
+ ### Framework versions
60
+
61
+ - Transformers 4.45.0
62
+ - Pytorch 2.4.0+cu121
63
+ - Datasets 2.14.6
64
+ - Tokenizers 0.20.1
all_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 0.0,
4
+ "train_loss": 0.4979530750094233,
5
+ "train_runtime": 139.1789,
6
+ "train_samples": 2345,
7
+ "train_samples_per_second": 16.849,
8
+ "train_steps_per_second": 0.266
9
+ }
config.json CHANGED
@@ -23,7 +23,7 @@
23
  "tie_word_embeddings": true,
24
  "torch_dtype": "bfloat16",
25
  "transformers_version": "4.45.0",
26
- "use_cache": false,
27
  "use_sliding_window": false,
28
  "vocab_size": 151936
29
  }
 
23
  "tie_word_embeddings": true,
24
  "torch_dtype": "bfloat16",
25
  "transformers_version": "4.45.0",
26
+ "use_cache": true,
27
  "use_sliding_window": false,
28
  "vocab_size": 151936
29
  }
generation_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "pad_token_id": 151643,
9
+ "repetition_penalty": 1.1,
10
+ "temperature": 0.7,
11
+ "top_k": 20,
12
+ "top_p": 0.8,
13
+ "transformers_version": "4.45.0"
14
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 1.0,
3
+ "total_flos": 0.0,
4
+ "train_loss": 0.4979530750094233,
5
+ "train_runtime": 139.1789,
6
+ "train_samples": 2345,
7
+ "train_samples_per_second": 16.849,
8
+ "train_steps_per_second": 0.266
9
+ }
trainer_state.json ADDED
@@ -0,0 +1,819 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 1.0,
5
+ "eval_steps": 100,
6
+ "global_step": 37,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "debug/policy_chosen_logits": -3.41206693649292,
13
+ "debug/policy_chosen_logps": -143.21307373046875,
14
+ "debug/policy_rejected_logits": -3.2551214694976807,
15
+ "debug/policy_rejected_logps": -150.2328338623047,
16
+ "debug/reference_chosen_logps": -143.21307373046875,
17
+ "debug/reference_rejected_logps": -150.2328338623047,
18
+ "epoch": 0.02702702702702703,
19
+ "grad_norm": 3.5140838495338116,
20
+ "learning_rate": 1e-06,
21
+ "logits/chosen": -3.41206693649292,
22
+ "logits/rejected": -3.2551214694976807,
23
+ "logps/chosen": -143.21307373046875,
24
+ "logps/rejected": -150.2328338623047,
25
+ "loss": 0.5,
26
+ "rewards/accuracies": 0.0,
27
+ "rewards/chosen": 0.0,
28
+ "rewards/margins": 0.0,
29
+ "rewards/rejected": 0.0,
30
+ "step": 1
31
+ },
32
+ {
33
+ "debug/policy_chosen_logits": -3.4049649238586426,
34
+ "debug/policy_chosen_logps": -143.7670135498047,
35
+ "debug/policy_rejected_logits": -3.4008069038391113,
36
+ "debug/policy_rejected_logps": -138.0754852294922,
37
+ "debug/reference_chosen_logps": -143.82278442382812,
38
+ "debug/reference_rejected_logps": -138.1428985595703,
39
+ "epoch": 0.05405405405405406,
40
+ "grad_norm": 3.526231895119079,
41
+ "learning_rate": 1e-06,
42
+ "logits/chosen": -3.4049649238586426,
43
+ "logits/rejected": -3.4008069038391113,
44
+ "logps/chosen": -143.7670135498047,
45
+ "logps/rejected": -138.0754852294922,
46
+ "loss": 0.4994,
47
+ "rewards/accuracies": 0.375,
48
+ "rewards/chosen": 0.0005577945848926902,
49
+ "rewards/margins": -0.00011638639261946082,
50
+ "rewards/rejected": 0.000674180977512151,
51
+ "step": 2
52
+ },
53
+ {
54
+ "debug/policy_chosen_logits": -3.2883780002593994,
55
+ "debug/policy_chosen_logps": -143.35385131835938,
56
+ "debug/policy_rejected_logits": -3.2103564739227295,
57
+ "debug/policy_rejected_logps": -139.85372924804688,
58
+ "debug/reference_chosen_logps": -143.38311767578125,
59
+ "debug/reference_rejected_logps": -139.98326110839844,
60
+ "epoch": 0.08108108108108109,
61
+ "grad_norm": 3.618148106626628,
62
+ "learning_rate": 1e-06,
63
+ "logits/chosen": -3.2883780002593994,
64
+ "logits/rejected": -3.2103564739227295,
65
+ "logps/chosen": -143.35385131835938,
66
+ "logps/rejected": -139.85372924804688,
67
+ "loss": 0.5006,
68
+ "rewards/accuracies": 0.25,
69
+ "rewards/chosen": 0.00029258732683956623,
70
+ "rewards/margins": -0.001002683537080884,
71
+ "rewards/rejected": 0.0012952708639204502,
72
+ "step": 3
73
+ },
74
+ {
75
+ "debug/policy_chosen_logits": -3.294776678085327,
76
+ "debug/policy_chosen_logps": -132.79519653320312,
77
+ "debug/policy_rejected_logits": -3.4150912761688232,
78
+ "debug/policy_rejected_logps": -138.08180236816406,
79
+ "debug/reference_chosen_logps": -132.62167358398438,
80
+ "debug/reference_rejected_logps": -137.927978515625,
81
+ "epoch": 0.10810810810810811,
82
+ "grad_norm": 3.5093502617713077,
83
+ "learning_rate": 1e-06,
84
+ "logits/chosen": -3.294776678085327,
85
+ "logits/rejected": -3.4150912761688232,
86
+ "logps/chosen": -132.79519653320312,
87
+ "logps/rejected": -138.08180236816406,
88
+ "loss": 0.4994,
89
+ "rewards/accuracies": 0.5,
90
+ "rewards/chosen": -0.0017352199647575617,
91
+ "rewards/margins": -0.0001968955621123314,
92
+ "rewards/rejected": -0.0015383241698145866,
93
+ "step": 4
94
+ },
95
+ {
96
+ "debug/policy_chosen_logits": -3.343012809753418,
97
+ "debug/policy_chosen_logps": -126.95188903808594,
98
+ "debug/policy_rejected_logits": -3.3124771118164062,
99
+ "debug/policy_rejected_logps": -129.9374237060547,
100
+ "debug/reference_chosen_logps": -126.82910919189453,
101
+ "debug/reference_rejected_logps": -129.91493225097656,
102
+ "epoch": 0.13513513513513514,
103
+ "grad_norm": 3.250245901043351,
104
+ "learning_rate": 1e-06,
105
+ "logits/chosen": -3.343012809753418,
106
+ "logits/rejected": -3.3124771118164062,
107
+ "logps/chosen": -126.95188903808594,
108
+ "logps/rejected": -129.9374237060547,
109
+ "loss": 0.5009,
110
+ "rewards/accuracies": 0.625,
111
+ "rewards/chosen": -0.001227750675752759,
112
+ "rewards/margins": -0.001002778997644782,
113
+ "rewards/rejected": -0.00022497176541946828,
114
+ "step": 5
115
+ },
116
+ {
117
+ "debug/policy_chosen_logits": -3.2374277114868164,
118
+ "debug/policy_chosen_logps": -154.26406860351562,
119
+ "debug/policy_rejected_logits": -3.2978382110595703,
120
+ "debug/policy_rejected_logps": -143.40101623535156,
121
+ "debug/reference_chosen_logps": -153.9961395263672,
122
+ "debug/reference_rejected_logps": -143.3165283203125,
123
+ "epoch": 0.16216216216216217,
124
+ "grad_norm": 3.439711062354016,
125
+ "learning_rate": 1e-06,
126
+ "logits/chosen": -3.2374277114868164,
127
+ "logits/rejected": -3.2978382110595703,
128
+ "logps/chosen": -154.26406860351562,
129
+ "logps/rejected": -143.40101623535156,
130
+ "loss": 0.5009,
131
+ "rewards/accuracies": 0.375,
132
+ "rewards/chosen": -0.002679176162928343,
133
+ "rewards/margins": -0.0018340873066335917,
134
+ "rewards/rejected": -0.000845088972710073,
135
+ "step": 6
136
+ },
137
+ {
138
+ "debug/policy_chosen_logits": -3.2252252101898193,
139
+ "debug/policy_chosen_logps": -162.3474578857422,
140
+ "debug/policy_rejected_logits": -3.230537176132202,
141
+ "debug/policy_rejected_logps": -147.3297576904297,
142
+ "debug/reference_chosen_logps": -162.26889038085938,
143
+ "debug/reference_rejected_logps": -147.57199096679688,
144
+ "epoch": 0.1891891891891892,
145
+ "grad_norm": 3.5128248367150468,
146
+ "learning_rate": 1e-06,
147
+ "logits/chosen": -3.2252252101898193,
148
+ "logits/rejected": -3.230537176132202,
149
+ "logps/chosen": -162.3474578857422,
150
+ "logps/rejected": -147.3297576904297,
151
+ "loss": 0.5005,
152
+ "rewards/accuracies": 0.25,
153
+ "rewards/chosen": -0.0007855510921217501,
154
+ "rewards/margins": -0.003207969479262829,
155
+ "rewards/rejected": 0.0024224184453487396,
156
+ "step": 7
157
+ },
158
+ {
159
+ "debug/policy_chosen_logits": -3.3006181716918945,
160
+ "debug/policy_chosen_logps": -149.18630981445312,
161
+ "debug/policy_rejected_logits": -3.3146488666534424,
162
+ "debug/policy_rejected_logps": -147.72348022460938,
163
+ "debug/reference_chosen_logps": -149.20960998535156,
164
+ "debug/reference_rejected_logps": -147.54893493652344,
165
+ "epoch": 0.21621621621621623,
166
+ "grad_norm": 3.499360817762659,
167
+ "learning_rate": 1e-06,
168
+ "logits/chosen": -3.3006181716918945,
169
+ "logits/rejected": -3.3146488666534424,
170
+ "logps/chosen": -149.18630981445312,
171
+ "logps/rejected": -147.72348022460938,
172
+ "loss": 0.4992,
173
+ "rewards/accuracies": 0.75,
174
+ "rewards/chosen": 0.000233030179515481,
175
+ "rewards/margins": 0.001978444866836071,
176
+ "rewards/rejected": -0.00174541468732059,
177
+ "step": 8
178
+ },
179
+ {
180
+ "debug/policy_chosen_logits": -3.3503048419952393,
181
+ "debug/policy_chosen_logps": -150.99197387695312,
182
+ "debug/policy_rejected_logits": -3.2822837829589844,
183
+ "debug/policy_rejected_logps": -145.26345825195312,
184
+ "debug/reference_chosen_logps": -151.06939697265625,
185
+ "debug/reference_rejected_logps": -145.52816772460938,
186
+ "epoch": 0.24324324324324326,
187
+ "grad_norm": 3.4264003876223215,
188
+ "learning_rate": 1e-06,
189
+ "logits/chosen": -3.3503048419952393,
190
+ "logits/rejected": -3.2822837829589844,
191
+ "logps/chosen": -150.99197387695312,
192
+ "logps/rejected": -145.26345825195312,
193
+ "loss": 0.4987,
194
+ "rewards/accuracies": 0.375,
195
+ "rewards/chosen": 0.0007742022862657905,
196
+ "rewards/margins": -0.0018729494186118245,
197
+ "rewards/rejected": 0.002647151704877615,
198
+ "step": 9
199
+ },
200
+ {
201
+ "debug/policy_chosen_logits": -3.304420232772827,
202
+ "debug/policy_chosen_logps": -133.21044921875,
203
+ "debug/policy_rejected_logits": -3.412870168685913,
204
+ "debug/policy_rejected_logps": -125.90888977050781,
205
+ "debug/reference_chosen_logps": -133.26504516601562,
206
+ "debug/reference_rejected_logps": -125.95523834228516,
207
+ "epoch": 0.2702702702702703,
208
+ "grad_norm": 3.3820605783151074,
209
+ "learning_rate": 1e-06,
210
+ "logits/chosen": -3.304420232772827,
211
+ "logits/rejected": -3.412870168685913,
212
+ "logps/chosen": -133.21044921875,
213
+ "logps/rejected": -125.90888977050781,
214
+ "loss": 0.5005,
215
+ "rewards/accuracies": 0.5,
216
+ "rewards/chosen": 0.0005457591032609344,
217
+ "rewards/margins": 8.22543297545053e-05,
218
+ "rewards/rejected": 0.0004635047516785562,
219
+ "step": 10
220
+ },
221
+ {
222
+ "debug/policy_chosen_logits": -3.3585329055786133,
223
+ "debug/policy_chosen_logps": -139.380859375,
224
+ "debug/policy_rejected_logits": -3.3599538803100586,
225
+ "debug/policy_rejected_logps": -132.3596649169922,
226
+ "debug/reference_chosen_logps": -139.71542358398438,
227
+ "debug/reference_rejected_logps": -132.61981201171875,
228
+ "epoch": 0.2972972972972973,
229
+ "grad_norm": 3.288218575224273,
230
+ "learning_rate": 1e-06,
231
+ "logits/chosen": -3.3585329055786133,
232
+ "logits/rejected": -3.3599538803100586,
233
+ "logps/chosen": -139.380859375,
234
+ "logps/rejected": -132.3596649169922,
235
+ "loss": 0.4999,
236
+ "rewards/accuracies": 0.5,
237
+ "rewards/chosen": 0.0033456420060247183,
238
+ "rewards/margins": 0.0007441901252605021,
239
+ "rewards/rejected": 0.0026014517061412334,
240
+ "step": 11
241
+ },
242
+ {
243
+ "debug/policy_chosen_logits": -3.2882096767425537,
244
+ "debug/policy_chosen_logps": -156.53334045410156,
245
+ "debug/policy_rejected_logits": -3.2960500717163086,
246
+ "debug/policy_rejected_logps": -143.13995361328125,
247
+ "debug/reference_chosen_logps": -156.6325225830078,
248
+ "debug/reference_rejected_logps": -142.83905029296875,
249
+ "epoch": 0.32432432432432434,
250
+ "grad_norm": 3.813316903837083,
251
+ "learning_rate": 1e-06,
252
+ "logits/chosen": -3.2882096767425537,
253
+ "logits/rejected": -3.2960500717163086,
254
+ "logps/chosen": -156.53334045410156,
255
+ "logps/rejected": -143.13995361328125,
256
+ "loss": 0.4989,
257
+ "rewards/accuracies": 0.625,
258
+ "rewards/chosen": 0.0009917640127241611,
259
+ "rewards/margins": 0.004000749439001083,
260
+ "rewards/rejected": -0.0030089854262769222,
261
+ "step": 12
262
+ },
263
+ {
264
+ "debug/policy_chosen_logits": -3.2771499156951904,
265
+ "debug/policy_chosen_logps": -133.85403442382812,
266
+ "debug/policy_rejected_logits": -3.3577096462249756,
267
+ "debug/policy_rejected_logps": -145.81051635742188,
268
+ "debug/reference_chosen_logps": -134.21636962890625,
269
+ "debug/reference_rejected_logps": -145.71588134765625,
270
+ "epoch": 0.35135135135135137,
271
+ "grad_norm": 3.5238547026932254,
272
+ "learning_rate": 1e-06,
273
+ "logits/chosen": -3.2771499156951904,
274
+ "logits/rejected": -3.3577096462249756,
275
+ "logps/chosen": -133.85403442382812,
276
+ "logps/rejected": -145.81051635742188,
277
+ "loss": 0.5004,
278
+ "rewards/accuracies": 0.75,
279
+ "rewards/chosen": 0.003623266238719225,
280
+ "rewards/margins": 0.004569797310978174,
281
+ "rewards/rejected": -0.0009465311886742711,
282
+ "step": 13
283
+ },
284
+ {
285
+ "debug/policy_chosen_logits": -3.4804084300994873,
286
+ "debug/policy_chosen_logps": -165.59515380859375,
287
+ "debug/policy_rejected_logits": -3.5207834243774414,
288
+ "debug/policy_rejected_logps": -139.4119873046875,
289
+ "debug/reference_chosen_logps": -165.61618041992188,
290
+ "debug/reference_rejected_logps": -139.546142578125,
291
+ "epoch": 0.3783783783783784,
292
+ "grad_norm": 3.271576173715924,
293
+ "learning_rate": 1e-06,
294
+ "logits/chosen": -3.4804084300994873,
295
+ "logits/rejected": -3.5207834243774414,
296
+ "logps/chosen": -165.59515380859375,
297
+ "logps/rejected": -139.4119873046875,
298
+ "loss": 0.4997,
299
+ "rewards/accuracies": 0.625,
300
+ "rewards/chosen": 0.00021032325457781553,
301
+ "rewards/margins": -0.0011312393471598625,
302
+ "rewards/rejected": 0.0013415624853223562,
303
+ "step": 14
304
+ },
305
+ {
306
+ "debug/policy_chosen_logits": -3.371809244155884,
307
+ "debug/policy_chosen_logps": -138.2847900390625,
308
+ "debug/policy_rejected_logits": -3.2725796699523926,
309
+ "debug/policy_rejected_logps": -146.83644104003906,
310
+ "debug/reference_chosen_logps": -138.04966735839844,
311
+ "debug/reference_rejected_logps": -146.4534454345703,
312
+ "epoch": 0.40540540540540543,
313
+ "grad_norm": 3.5549880331732084,
314
+ "learning_rate": 1e-06,
315
+ "logits/chosen": -3.371809244155884,
316
+ "logits/rejected": -3.2725796699523926,
317
+ "logps/chosen": -138.2847900390625,
318
+ "logps/rejected": -146.83644104003906,
319
+ "loss": 0.4978,
320
+ "rewards/accuracies": 0.5,
321
+ "rewards/chosen": -0.002351150382310152,
322
+ "rewards/margins": 0.0014788818079978228,
323
+ "rewards/rejected": -0.0038300324231386185,
324
+ "step": 15
325
+ },
326
+ {
327
+ "debug/policy_chosen_logits": -3.2811360359191895,
328
+ "debug/policy_chosen_logps": -145.2406005859375,
329
+ "debug/policy_rejected_logits": -3.2666900157928467,
330
+ "debug/policy_rejected_logps": -136.08499145507812,
331
+ "debug/reference_chosen_logps": -145.38519287109375,
332
+ "debug/reference_rejected_logps": -136.1790313720703,
333
+ "epoch": 0.43243243243243246,
334
+ "grad_norm": 3.390241199656903,
335
+ "learning_rate": 1e-06,
336
+ "logits/chosen": -3.2811360359191895,
337
+ "logits/rejected": -3.2666900157928467,
338
+ "logps/chosen": -145.2406005859375,
339
+ "logps/rejected": -136.08499145507812,
340
+ "loss": 0.4992,
341
+ "rewards/accuracies": 0.375,
342
+ "rewards/chosen": 0.0014458559453487396,
343
+ "rewards/margins": 0.0005054186331108212,
344
+ "rewards/rejected": 0.0009404371376149356,
345
+ "step": 16
346
+ },
347
+ {
348
+ "debug/policy_chosen_logits": -3.4209792613983154,
349
+ "debug/policy_chosen_logps": -154.63943481445312,
350
+ "debug/policy_rejected_logits": -3.3486948013305664,
351
+ "debug/policy_rejected_logps": -133.35781860351562,
352
+ "debug/reference_chosen_logps": -154.70895385742188,
353
+ "debug/reference_rejected_logps": -133.4116668701172,
354
+ "epoch": 0.4594594594594595,
355
+ "grad_norm": 3.6742049248327437,
356
+ "learning_rate": 1e-06,
357
+ "logits/chosen": -3.4209792613983154,
358
+ "logits/rejected": -3.3486948013305664,
359
+ "logps/chosen": -154.63943481445312,
360
+ "logps/rejected": -133.35781860351562,
361
+ "loss": 0.4989,
362
+ "rewards/accuracies": 0.5,
363
+ "rewards/chosen": 0.0006950664683245122,
364
+ "rewards/margins": 0.00015656478353776038,
365
+ "rewards/rejected": 0.0005385016556829214,
366
+ "step": 17
367
+ },
368
+ {
369
+ "debug/policy_chosen_logits": -3.298342704772949,
370
+ "debug/policy_chosen_logps": -146.52044677734375,
371
+ "debug/policy_rejected_logits": -3.3201894760131836,
372
+ "debug/policy_rejected_logps": -156.97808837890625,
373
+ "debug/reference_chosen_logps": -146.34840393066406,
374
+ "debug/reference_rejected_logps": -156.84378051757812,
375
+ "epoch": 0.4864864864864865,
376
+ "grad_norm": 3.5194962148324103,
377
+ "learning_rate": 1e-06,
378
+ "logits/chosen": -3.298342704772949,
379
+ "logits/rejected": -3.3201894760131836,
380
+ "logps/chosen": -146.52044677734375,
381
+ "logps/rejected": -156.97808837890625,
382
+ "loss": 0.4973,
383
+ "rewards/accuracies": 0.5,
384
+ "rewards/chosen": -0.0017202282324433327,
385
+ "rewards/margins": -0.0003771304036490619,
386
+ "rewards/rejected": -0.001343097654171288,
387
+ "step": 18
388
+ },
389
+ {
390
+ "debug/policy_chosen_logits": -3.3400371074676514,
391
+ "debug/policy_chosen_logps": -132.01437377929688,
392
+ "debug/policy_rejected_logits": -3.36023211479187,
393
+ "debug/policy_rejected_logps": -148.31887817382812,
394
+ "debug/reference_chosen_logps": -131.88931274414062,
395
+ "debug/reference_rejected_logps": -148.13345336914062,
396
+ "epoch": 0.5135135135135135,
397
+ "grad_norm": 3.61313418435521,
398
+ "learning_rate": 1e-06,
399
+ "logits/chosen": -3.3400371074676514,
400
+ "logits/rejected": -3.36023211479187,
401
+ "logps/chosen": -132.01437377929688,
402
+ "logps/rejected": -148.31887817382812,
403
+ "loss": 0.498,
404
+ "rewards/accuracies": 0.5,
405
+ "rewards/chosen": -0.00125067715998739,
406
+ "rewards/margins": 0.0006034375401213765,
407
+ "rewards/rejected": -0.001854114467278123,
408
+ "step": 19
409
+ },
410
+ {
411
+ "debug/policy_chosen_logits": -3.2641985416412354,
412
+ "debug/policy_chosen_logps": -134.37350463867188,
413
+ "debug/policy_rejected_logits": -3.255669355392456,
414
+ "debug/policy_rejected_logps": -124.11454772949219,
415
+ "debug/reference_chosen_logps": -134.53549194335938,
416
+ "debug/reference_rejected_logps": -123.25413513183594,
417
+ "epoch": 0.5405405405405406,
418
+ "grad_norm": 3.4240940627433663,
419
+ "learning_rate": 1e-06,
420
+ "logits/chosen": -3.2641985416412354,
421
+ "logits/rejected": -3.255669355392456,
422
+ "logps/chosen": -134.37350463867188,
423
+ "logps/rejected": -124.11454772949219,
424
+ "loss": 0.4969,
425
+ "rewards/accuracies": 0.5,
426
+ "rewards/chosen": 0.0016198731027543545,
427
+ "rewards/margins": 0.010224045254290104,
428
+ "rewards/rejected": -0.008604173548519611,
429
+ "step": 20
430
+ },
431
+ {
432
+ "debug/policy_chosen_logits": -3.389690637588501,
433
+ "debug/policy_chosen_logps": -144.95230102539062,
434
+ "debug/policy_rejected_logits": -3.3237528800964355,
435
+ "debug/policy_rejected_logps": -140.4608917236328,
436
+ "debug/reference_chosen_logps": -145.19155883789062,
437
+ "debug/reference_rejected_logps": -139.9624481201172,
438
+ "epoch": 0.5675675675675675,
439
+ "grad_norm": 3.5636895297483804,
440
+ "learning_rate": 1e-06,
441
+ "logits/chosen": -3.389690637588501,
442
+ "logits/rejected": -3.3237528800964355,
443
+ "logps/chosen": -144.95230102539062,
444
+ "logps/rejected": -140.4608917236328,
445
+ "loss": 0.4972,
446
+ "rewards/accuracies": 0.375,
447
+ "rewards/chosen": 0.0023925015702843666,
448
+ "rewards/margins": 0.007377022877335548,
449
+ "rewards/rejected": -0.004984522238373756,
450
+ "step": 21
451
+ },
452
+ {
453
+ "debug/policy_chosen_logits": -3.2673821449279785,
454
+ "debug/policy_chosen_logps": -159.7883758544922,
455
+ "debug/policy_rejected_logits": -3.2593047618865967,
456
+ "debug/policy_rejected_logps": -141.72947692871094,
457
+ "debug/reference_chosen_logps": -159.80300903320312,
458
+ "debug/reference_rejected_logps": -142.28695678710938,
459
+ "epoch": 0.5945945945945946,
460
+ "grad_norm": 3.8014985351330988,
461
+ "learning_rate": 1e-06,
462
+ "logits/chosen": -3.2673821449279785,
463
+ "logits/rejected": -3.2593047618865967,
464
+ "logps/chosen": -159.7883758544922,
465
+ "logps/rejected": -141.72947692871094,
466
+ "loss": 0.499,
467
+ "rewards/accuracies": 0.125,
468
+ "rewards/chosen": 0.00014627468772232533,
469
+ "rewards/margins": -0.005428638309240341,
470
+ "rewards/rejected": 0.00557491322979331,
471
+ "step": 22
472
+ },
473
+ {
474
+ "debug/policy_chosen_logits": -3.3076963424682617,
475
+ "debug/policy_chosen_logps": -147.4810791015625,
476
+ "debug/policy_rejected_logits": -3.3063511848449707,
477
+ "debug/policy_rejected_logps": -155.9129638671875,
478
+ "debug/reference_chosen_logps": -147.6890869140625,
479
+ "debug/reference_rejected_logps": -155.93359375,
480
+ "epoch": 0.6216216216216216,
481
+ "grad_norm": 3.5943677168048143,
482
+ "learning_rate": 1e-06,
483
+ "logits/chosen": -3.3076963424682617,
484
+ "logits/rejected": -3.3063511848449707,
485
+ "logps/chosen": -147.4810791015625,
486
+ "logps/rejected": -155.9129638671875,
487
+ "loss": 0.499,
488
+ "rewards/accuracies": 0.625,
489
+ "rewards/chosen": 0.0020801923237740993,
490
+ "rewards/margins": 0.001873721950687468,
491
+ "rewards/rejected": 0.00020647048950195312,
492
+ "step": 23
493
+ },
494
+ {
495
+ "debug/policy_chosen_logits": -3.3162076473236084,
496
+ "debug/policy_chosen_logps": -146.12327575683594,
497
+ "debug/policy_rejected_logits": -3.2808167934417725,
498
+ "debug/policy_rejected_logps": -150.74459838867188,
499
+ "debug/reference_chosen_logps": -146.20852661132812,
500
+ "debug/reference_rejected_logps": -151.0212860107422,
501
+ "epoch": 0.6486486486486487,
502
+ "grad_norm": 3.490255067904084,
503
+ "learning_rate": 1e-06,
504
+ "logits/chosen": -3.3162076473236084,
505
+ "logits/rejected": -3.2808167934417725,
506
+ "logps/chosen": -146.12327575683594,
507
+ "logps/rejected": -150.74459838867188,
508
+ "loss": 0.4979,
509
+ "rewards/accuracies": 0.75,
510
+ "rewards/chosen": 0.0008525657467544079,
511
+ "rewards/margins": -0.0019143482204526663,
512
+ "rewards/rejected": 0.002766914200037718,
513
+ "step": 24
514
+ },
515
+ {
516
+ "debug/policy_chosen_logits": -3.341583490371704,
517
+ "debug/policy_chosen_logps": -145.74295043945312,
518
+ "debug/policy_rejected_logits": -3.2622642517089844,
519
+ "debug/policy_rejected_logps": -148.95095825195312,
520
+ "debug/reference_chosen_logps": -145.28700256347656,
521
+ "debug/reference_rejected_logps": -148.38320922851562,
522
+ "epoch": 0.6756756756756757,
523
+ "grad_norm": 3.4930826080691433,
524
+ "learning_rate": 1e-06,
525
+ "logits/chosen": -3.341583490371704,
526
+ "logits/rejected": -3.2622642517089844,
527
+ "logps/chosen": -145.74295043945312,
528
+ "logps/rejected": -148.95095825195312,
529
+ "loss": 0.4992,
530
+ "rewards/accuracies": 0.375,
531
+ "rewards/chosen": -0.0045594689436256886,
532
+ "rewards/margins": 0.001117925625294447,
533
+ "rewards/rejected": -0.0056773945689201355,
534
+ "step": 25
535
+ },
536
+ {
537
+ "debug/policy_chosen_logits": -3.3710758686065674,
538
+ "debug/policy_chosen_logps": -134.76580810546875,
539
+ "debug/policy_rejected_logits": -3.268131971359253,
540
+ "debug/policy_rejected_logps": -164.87716674804688,
541
+ "debug/reference_chosen_logps": -134.9357452392578,
542
+ "debug/reference_rejected_logps": -164.06057739257812,
543
+ "epoch": 0.7027027027027027,
544
+ "grad_norm": 3.5324806343379795,
545
+ "learning_rate": 1e-06,
546
+ "logits/chosen": -3.3710758686065674,
547
+ "logits/rejected": -3.268131971359253,
548
+ "logps/chosen": -134.76580810546875,
549
+ "logps/rejected": -164.87716674804688,
550
+ "loss": 0.4961,
551
+ "rewards/accuracies": 0.875,
552
+ "rewards/chosen": 0.0016993426252156496,
553
+ "rewards/margins": 0.009865359403192997,
554
+ "rewards/rejected": -0.008166017010807991,
555
+ "step": 26
556
+ },
557
+ {
558
+ "debug/policy_chosen_logits": -3.276531934738159,
559
+ "debug/policy_chosen_logps": -128.13336181640625,
560
+ "debug/policy_rejected_logits": -3.2201032638549805,
561
+ "debug/policy_rejected_logps": -148.01954650878906,
562
+ "debug/reference_chosen_logps": -127.79987335205078,
563
+ "debug/reference_rejected_logps": -148.602783203125,
564
+ "epoch": 0.7297297297297297,
565
+ "grad_norm": 3.5504923092772165,
566
+ "learning_rate": 1e-06,
567
+ "logits/chosen": -3.276531934738159,
568
+ "logits/rejected": -3.2201032638549805,
569
+ "logps/chosen": -128.13336181640625,
570
+ "logps/rejected": -148.01954650878906,
571
+ "loss": 0.4978,
572
+ "rewards/accuracies": 0.25,
573
+ "rewards/chosen": -0.0033348274882882833,
574
+ "rewards/margins": -0.009167090058326721,
575
+ "rewards/rejected": 0.005832261871546507,
576
+ "step": 27
577
+ },
578
+ {
579
+ "debug/policy_chosen_logits": -3.3670849800109863,
580
+ "debug/policy_chosen_logps": -142.864013671875,
581
+ "debug/policy_rejected_logits": -3.314009428024292,
582
+ "debug/policy_rejected_logps": -159.29888916015625,
583
+ "debug/reference_chosen_logps": -143.54417419433594,
584
+ "debug/reference_rejected_logps": -159.19894409179688,
585
+ "epoch": 0.7567567567567568,
586
+ "grad_norm": 3.6333481085660533,
587
+ "learning_rate": 1e-06,
588
+ "logits/chosen": -3.3670849800109863,
589
+ "logits/rejected": -3.314009428024292,
590
+ "logps/chosen": -142.864013671875,
591
+ "logps/rejected": -159.29888916015625,
592
+ "loss": 0.4946,
593
+ "rewards/accuracies": 0.875,
594
+ "rewards/chosen": 0.006801585666835308,
595
+ "rewards/margins": 0.007801036350429058,
596
+ "rewards/rejected": -0.0009994508000090718,
597
+ "step": 28
598
+ },
599
+ {
600
+ "debug/policy_chosen_logits": -3.4510252475738525,
601
+ "debug/policy_chosen_logps": -144.7528839111328,
602
+ "debug/policy_rejected_logits": -3.4992246627807617,
603
+ "debug/policy_rejected_logps": -129.56163024902344,
604
+ "debug/reference_chosen_logps": -144.27435302734375,
605
+ "debug/reference_rejected_logps": -128.7982177734375,
606
+ "epoch": 0.7837837837837838,
607
+ "grad_norm": 3.519707420288352,
608
+ "learning_rate": 1e-06,
609
+ "logits/chosen": -3.4510252475738525,
610
+ "logits/rejected": -3.4992246627807617,
611
+ "logps/chosen": -144.7528839111328,
612
+ "logps/rejected": -129.56163024902344,
613
+ "loss": 0.4917,
614
+ "rewards/accuracies": 0.625,
615
+ "rewards/chosen": -0.0047853561118245125,
616
+ "rewards/margins": 0.0028487201780080795,
617
+ "rewards/rejected": -0.007634077221155167,
618
+ "step": 29
619
+ },
620
+ {
621
+ "debug/policy_chosen_logits": -3.442349910736084,
622
+ "debug/policy_chosen_logps": -129.01779174804688,
623
+ "debug/policy_rejected_logits": -3.355781078338623,
624
+ "debug/policy_rejected_logps": -125.97911071777344,
625
+ "debug/reference_chosen_logps": -129.1351318359375,
626
+ "debug/reference_rejected_logps": -125.61131286621094,
627
+ "epoch": 0.8108108108108109,
628
+ "grad_norm": 3.436282122029211,
629
+ "learning_rate": 1e-06,
630
+ "logits/chosen": -3.442349910736084,
631
+ "logits/rejected": -3.355781078338623,
632
+ "logps/chosen": -129.01779174804688,
633
+ "logps/rejected": -125.97911071777344,
634
+ "loss": 0.4957,
635
+ "rewards/accuracies": 0.375,
636
+ "rewards/chosen": 0.0011734389699995518,
637
+ "rewards/margins": 0.004851359874010086,
638
+ "rewards/rejected": -0.003677921136841178,
639
+ "step": 30
640
+ },
641
+ {
642
+ "debug/policy_chosen_logits": -3.46421480178833,
643
+ "debug/policy_chosen_logps": -148.1017608642578,
644
+ "debug/policy_rejected_logits": -3.4284045696258545,
645
+ "debug/policy_rejected_logps": -155.14401245117188,
646
+ "debug/reference_chosen_logps": -147.857666015625,
647
+ "debug/reference_rejected_logps": -153.7999267578125,
648
+ "epoch": 0.8378378378378378,
649
+ "grad_norm": 3.573380219675295,
650
+ "learning_rate": 1e-06,
651
+ "logits/chosen": -3.46421480178833,
652
+ "logits/rejected": -3.4284045696258545,
653
+ "logps/chosen": -148.1017608642578,
654
+ "logps/rejected": -155.14401245117188,
655
+ "loss": 0.4994,
656
+ "rewards/accuracies": 0.5,
657
+ "rewards/chosen": -0.002440805546939373,
658
+ "rewards/margins": 0.010999973863363266,
659
+ "rewards/rejected": -0.013440780341625214,
660
+ "step": 31
661
+ },
662
+ {
663
+ "debug/policy_chosen_logits": -3.296452522277832,
664
+ "debug/policy_chosen_logps": -146.0049285888672,
665
+ "debug/policy_rejected_logits": -3.3844947814941406,
666
+ "debug/policy_rejected_logps": -145.01739501953125,
667
+ "debug/reference_chosen_logps": -146.03688049316406,
668
+ "debug/reference_rejected_logps": -144.96478271484375,
669
+ "epoch": 0.8648648648648649,
670
+ "grad_norm": 3.7844677501283552,
671
+ "learning_rate": 1e-06,
672
+ "logits/chosen": -3.296452522277832,
673
+ "logits/rejected": -3.3844947814941406,
674
+ "logps/chosen": -146.0049285888672,
675
+ "logps/rejected": -145.01739501953125,
676
+ "loss": 0.4983,
677
+ "rewards/accuracies": 0.625,
678
+ "rewards/chosen": 0.0003195476019755006,
679
+ "rewards/margins": 0.0008457758231088519,
680
+ "rewards/rejected": -0.000526227755472064,
681
+ "step": 32
682
+ },
683
+ {
684
+ "debug/policy_chosen_logits": -3.393280506134033,
685
+ "debug/policy_chosen_logps": -135.49090576171875,
686
+ "debug/policy_rejected_logits": -3.4107859134674072,
687
+ "debug/policy_rejected_logps": -126.17267608642578,
688
+ "debug/reference_chosen_logps": -135.9659423828125,
689
+ "debug/reference_rejected_logps": -123.26311492919922,
690
+ "epoch": 0.8918918918918919,
691
+ "grad_norm": 3.6987210522943323,
692
+ "learning_rate": 1e-06,
693
+ "logits/chosen": -3.393280506134033,
694
+ "logits/rejected": -3.4107859134674072,
695
+ "logps/chosen": -135.49090576171875,
696
+ "logps/rejected": -126.17267608642578,
697
+ "loss": 0.4982,
698
+ "rewards/accuracies": 0.5,
699
+ "rewards/chosen": 0.00475044222548604,
700
+ "rewards/margins": 0.03384600579738617,
701
+ "rewards/rejected": -0.029095562174916267,
702
+ "step": 33
703
+ },
704
+ {
705
+ "debug/policy_chosen_logits": -3.3232836723327637,
706
+ "debug/policy_chosen_logps": -159.23611450195312,
707
+ "debug/policy_rejected_logits": -3.3686249256134033,
708
+ "debug/policy_rejected_logps": -165.2159423828125,
709
+ "debug/reference_chosen_logps": -158.15670776367188,
710
+ "debug/reference_rejected_logps": -164.94976806640625,
711
+ "epoch": 0.918918918918919,
712
+ "grad_norm": 3.553957133112115,
713
+ "learning_rate": 1e-06,
714
+ "logits/chosen": -3.3232836723327637,
715
+ "logits/rejected": -3.3686249256134033,
716
+ "logps/chosen": -159.23611450195312,
717
+ "logps/rejected": -165.2159423828125,
718
+ "loss": 0.4998,
719
+ "rewards/accuracies": 0.25,
720
+ "rewards/chosen": -0.010794105939567089,
721
+ "rewards/margins": -0.008132400922477245,
722
+ "rewards/rejected": -0.0026617045514285564,
723
+ "step": 34
724
+ },
725
+ {
726
+ "debug/policy_chosen_logits": -3.3675332069396973,
727
+ "debug/policy_chosen_logps": -129.2323455810547,
728
+ "debug/policy_rejected_logits": -3.39815092086792,
729
+ "debug/policy_rejected_logps": -157.40768432617188,
730
+ "debug/reference_chosen_logps": -129.46514892578125,
731
+ "debug/reference_rejected_logps": -157.58755493164062,
732
+ "epoch": 0.9459459459459459,
733
+ "grad_norm": 3.6687608874400897,
734
+ "learning_rate": 1e-06,
735
+ "logits/chosen": -3.3675332069396973,
736
+ "logits/rejected": -3.39815092086792,
737
+ "logps/chosen": -129.2323455810547,
738
+ "logps/rejected": -157.40768432617188,
739
+ "loss": 0.4972,
740
+ "rewards/accuracies": 0.625,
741
+ "rewards/chosen": 0.0023280431050807238,
742
+ "rewards/margins": 0.0005296231247484684,
743
+ "rewards/rejected": 0.0017984198639169335,
744
+ "step": 35
745
+ },
746
+ {
747
+ "debug/policy_chosen_logits": -3.2530903816223145,
748
+ "debug/policy_chosen_logps": -139.84768676757812,
749
+ "debug/policy_rejected_logits": -3.380523204803467,
750
+ "debug/policy_rejected_logps": -127.18656921386719,
751
+ "debug/reference_chosen_logps": -140.35733032226562,
752
+ "debug/reference_rejected_logps": -127.09129333496094,
753
+ "epoch": 0.972972972972973,
754
+ "grad_norm": 3.747493235351115,
755
+ "learning_rate": 1e-06,
756
+ "logits/chosen": -3.2530903816223145,
757
+ "logits/rejected": -3.380523204803467,
758
+ "logps/chosen": -139.84768676757812,
759
+ "logps/rejected": -127.18656921386719,
760
+ "loss": 0.4949,
761
+ "rewards/accuracies": 0.75,
762
+ "rewards/chosen": 0.005096264183521271,
763
+ "rewards/margins": 0.006049060728400946,
764
+ "rewards/rejected": -0.0009527970105409622,
765
+ "step": 36
766
+ },
767
+ {
768
+ "debug/policy_chosen_logits": -3.1672980785369873,
769
+ "debug/policy_chosen_logps": -165.81317138671875,
770
+ "debug/policy_rejected_logits": -3.1891133785247803,
771
+ "debug/policy_rejected_logps": -135.89328002929688,
772
+ "debug/reference_chosen_logps": -164.97256469726562,
773
+ "debug/reference_rejected_logps": -135.76358032226562,
774
+ "epoch": 1.0,
775
+ "grad_norm": 3.5111470282705084,
776
+ "learning_rate": 1e-06,
777
+ "logits/chosen": -3.1672980785369873,
778
+ "logits/rejected": -3.1891133785247803,
779
+ "logps/chosen": -165.81317138671875,
780
+ "logps/rejected": -135.89328002929688,
781
+ "loss": 0.4809,
782
+ "rewards/accuracies": 0.375,
783
+ "rewards/chosen": -0.00840616226196289,
784
+ "rewards/margins": -0.007109222002327442,
785
+ "rewards/rejected": -0.0012969397939741611,
786
+ "step": 37
787
+ },
788
+ {
789
+ "epoch": 1.0,
790
+ "step": 37,
791
+ "total_flos": 0.0,
792
+ "train_loss": 0.4979530750094233,
793
+ "train_runtime": 139.1789,
794
+ "train_samples_per_second": 16.849,
795
+ "train_steps_per_second": 0.266
796
+ }
797
+ ],
798
+ "logging_steps": 1,
799
+ "max_steps": 37,
800
+ "num_input_tokens_seen": 0,
801
+ "num_train_epochs": 1,
802
+ "save_steps": 500,
803
+ "stateful_callbacks": {
804
+ "TrainerControl": {
805
+ "args": {
806
+ "should_epoch_stop": false,
807
+ "should_evaluate": false,
808
+ "should_log": false,
809
+ "should_save": true,
810
+ "should_training_stop": true
811
+ },
812
+ "attributes": {}
813
+ }
814
+ },
815
+ "total_flos": 0.0,
816
+ "train_batch_size": 8,
817
+ "trial_name": null,
818
+ "trial_params": null
819
+ }