pavan01729 commited on
Commit
1d310f2
·
verified ·
1 Parent(s): 57bc604

Upload sld_27.0/panel_counter_v27_grpo1/trainer_state.json with huggingface_hub

Browse files
sld_27.0/panel_counter_v27_grpo1/trainer_state.json ADDED
@@ -0,0 +1,264 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": null,
3
+ "best_metric": null,
4
+ "best_model_checkpoint": null,
5
+ "epoch": 0.02207505518763797,
6
+ "eval_steps": 5.0,
7
+ "global_step": 10,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "clip_ratio/high_max": 0.0,
14
+ "clip_ratio/high_mean": 0.0,
15
+ "clip_ratio/low_mean": 0.0,
16
+ "clip_ratio/low_min": 0.0,
17
+ "clip_ratio/region_mean": 0.0,
18
+ "completions/clipped_ratio": 1.0,
19
+ "completions/max_length": 2048.0,
20
+ "completions/mean_length": 2048.0,
21
+ "completions/min_length": 2048.0,
22
+ "epoch": 0.002207505518763797,
23
+ "frac_reward_zero_std": 1.0,
24
+ "grad_norm": 0.022355834022164345,
25
+ "kl": 0.034164416836574674,
26
+ "learning_rate": 4.999451708687114e-06,
27
+ "loss": 0.0013665767619386315,
28
+ "reward": 0.0,
29
+ "reward_std": 0.0,
30
+ "rewards/MultiComponentReward/mean": 0.0,
31
+ "rewards/MultiComponentReward/std": 0.0,
32
+ "step": 1,
33
+ "step_time": 641.0631351396441
34
+ },
35
+ {
36
+ "clip_ratio/high_max": 0.0,
37
+ "clip_ratio/high_mean": 0.0,
38
+ "clip_ratio/low_mean": 0.0,
39
+ "clip_ratio/low_min": 0.0,
40
+ "clip_ratio/region_mean": 0.0,
41
+ "completions/clipped_ratio": 1.0,
42
+ "completions/max_length": 2048.0,
43
+ "completions/mean_length": 2048.0,
44
+ "completions/min_length": 2048.0,
45
+ "epoch": 0.004415011037527594,
46
+ "frac_reward_zero_std": 1.0,
47
+ "grad_norm": 0.014668405055999756,
48
+ "kl": 0.03627943736501038,
49
+ "learning_rate": 4.997807075247147e-06,
50
+ "loss": 0.0014511775225400925,
51
+ "reward": 0.0,
52
+ "reward_std": 0.0,
53
+ "rewards/MultiComponentReward/mean": 0.0,
54
+ "rewards/MultiComponentReward/std": 0.0,
55
+ "step": 2,
56
+ "step_time": 643.0692910882644
57
+ },
58
+ {
59
+ "clip_ratio/high_max": 0.0,
60
+ "clip_ratio/high_mean": 0.0,
61
+ "clip_ratio/low_mean": 0.0,
62
+ "clip_ratio/low_min": 0.0,
63
+ "clip_ratio/region_mean": 0.0,
64
+ "completions/clipped_ratio": 1.0,
65
+ "completions/max_length": 2048.0,
66
+ "completions/mean_length": 2048.0,
67
+ "completions/min_length": 2048.0,
68
+ "epoch": 0.006622516556291391,
69
+ "frac_reward_zero_std": 1.0,
70
+ "grad_norm": 0.013016336597502232,
71
+ "kl": 0.04479851736687124,
72
+ "learning_rate": 4.9950668210706795e-06,
73
+ "loss": 0.0017919406527653337,
74
+ "reward": 0.0,
75
+ "reward_std": 0.0,
76
+ "rewards/MultiComponentReward/mean": 0.0,
77
+ "rewards/MultiComponentReward/std": 0.0,
78
+ "step": 3,
79
+ "step_time": 644.4912113547325
80
+ },
81
+ {
82
+ "clip_ratio/high_max": 0.0,
83
+ "clip_ratio/high_mean": 0.0,
84
+ "clip_ratio/low_mean": 0.0,
85
+ "clip_ratio/low_min": 0.0,
86
+ "clip_ratio/region_mean": 0.0,
87
+ "completions/clipped_ratio": 1.0,
88
+ "completions/max_length": 2048.0,
89
+ "completions/mean_length": 2048.0,
90
+ "completions/min_length": 2048.0,
91
+ "epoch": 0.008830022075055188,
92
+ "frac_reward_zero_std": 1.0,
93
+ "grad_norm": 0.013489930890500546,
94
+ "kl": 0.04416139982640743,
95
+ "learning_rate": 4.9912321481237616e-06,
96
+ "loss": 0.0017664558254182339,
97
+ "reward": 0.0,
98
+ "reward_std": 0.0,
99
+ "rewards/MultiComponentReward/mean": 0.0,
100
+ "rewards/MultiComponentReward/std": 0.0,
101
+ "step": 4,
102
+ "step_time": 642.5191293242387
103
+ },
104
+ {
105
+ "clip_ratio/high_max": 0.0,
106
+ "clip_ratio/high_mean": 0.0,
107
+ "clip_ratio/low_mean": 0.0,
108
+ "clip_ratio/low_min": 0.0,
109
+ "clip_ratio/region_mean": 0.0,
110
+ "completions/clipped_ratio": 1.0,
111
+ "completions/max_length": 2048.0,
112
+ "completions/mean_length": 2048.0,
113
+ "completions/min_length": 2048.0,
114
+ "epoch": 0.011037527593818985,
115
+ "frac_reward_zero_std": 1.0,
116
+ "grad_norm": 0.01270960085093975,
117
+ "kl": 0.04382203845307231,
118
+ "learning_rate": 4.986304738420684e-06,
119
+ "loss": 0.001752881333231926,
120
+ "reward": 0.0,
121
+ "reward_std": 0.0,
122
+ "rewards/MultiComponentReward/mean": 0.0,
123
+ "rewards/MultiComponentReward/std": 0.0,
124
+ "step": 5,
125
+ "step_time": 644.6477308589965
126
+ },
127
+ {
128
+ "clip_ratio/high_max": 0.0,
129
+ "clip_ratio/high_mean": 0.0,
130
+ "clip_ratio/low_mean": 0.0,
131
+ "clip_ratio/low_min": 0.0,
132
+ "clip_ratio/region_mean": 0.0,
133
+ "completions/clipped_ratio": 1.0,
134
+ "completions/max_length": 2048.0,
135
+ "completions/mean_length": 2048.0,
136
+ "completions/min_length": 2048.0,
137
+ "epoch": 0.013245033112582781,
138
+ "frac_reward_zero_std": 1.0,
139
+ "grad_norm": 0.016926726326346397,
140
+ "kl": 0.048070103861391544,
141
+ "learning_rate": 4.980286753286196e-06,
142
+ "loss": 0.0019228039309382439,
143
+ "reward": 0.0,
144
+ "reward_std": 0.0,
145
+ "rewards/MultiComponentReward/mean": 0.0,
146
+ "rewards/MultiComponentReward/std": 0.0,
147
+ "step": 6,
148
+ "step_time": 642.798067860771
149
+ },
150
+ {
151
+ "clip_ratio/high_max": 0.0,
152
+ "clip_ratio/high_mean": 0.0,
153
+ "clip_ratio/low_mean": 0.0,
154
+ "clip_ratio/low_min": 0.0,
155
+ "clip_ratio/region_mean": 0.0,
156
+ "completions/clipped_ratio": 1.0,
157
+ "completions/max_length": 2048.0,
158
+ "completions/mean_length": 2048.0,
159
+ "completions/min_length": 2048.0,
160
+ "epoch": 0.01545253863134658,
161
+ "frac_reward_zero_std": 1.0,
162
+ "grad_norm": 0.012467299588024616,
163
+ "kl": 0.042229132261127234,
164
+ "learning_rate": 4.973180832407471e-06,
165
+ "loss": 0.0016891652485355735,
166
+ "reward": 0.0,
167
+ "reward_std": 0.0,
168
+ "rewards/MultiComponentReward/mean": 0.0,
169
+ "rewards/MultiComponentReward/std": 0.0,
170
+ "step": 7,
171
+ "step_time": 641.6913179177791
172
+ },
173
+ {
174
+ "clip_ratio/high_max": 0.0,
175
+ "clip_ratio/high_mean": 0.0,
176
+ "clip_ratio/low_mean": 0.0,
177
+ "clip_ratio/low_min": 0.0,
178
+ "clip_ratio/region_mean": 0.0,
179
+ "completions/clipped_ratio": 1.0,
180
+ "completions/max_length": 2048.0,
181
+ "completions/mean_length": 2048.0,
182
+ "completions/min_length": 2048.0,
183
+ "epoch": 0.017660044150110375,
184
+ "frac_reward_zero_std": 1.0,
185
+ "grad_norm": 0.008632641285657883,
186
+ "kl": 0.027748343301936984,
187
+ "learning_rate": 4.964990092676263e-06,
188
+ "loss": 0.0011099337134510279,
189
+ "reward": 0.0,
190
+ "reward_std": 0.0,
191
+ "rewards/MultiComponentReward/mean": 0.0,
192
+ "rewards/MultiComponentReward/std": 0.0,
193
+ "step": 8,
194
+ "step_time": 638.3092312384397
195
+ },
196
+ {
197
+ "clip_ratio/high_max": 0.0,
198
+ "clip_ratio/high_mean": 0.0,
199
+ "clip_ratio/low_mean": 0.0,
200
+ "clip_ratio/low_min": 0.0,
201
+ "clip_ratio/region_mean": 0.0,
202
+ "completions/clipped_ratio": 1.0,
203
+ "completions/max_length": 2048.0,
204
+ "completions/mean_length": 2048.0,
205
+ "completions/min_length": 2048.0,
206
+ "epoch": 0.019867549668874173,
207
+ "frac_reward_zero_std": 1.0,
208
+ "grad_norm": 0.006320472341030836,
209
+ "kl": 0.020729208830744028,
210
+ "learning_rate": 4.9557181268217225e-06,
211
+ "loss": 0.0008291683625429869,
212
+ "reward": 0.0,
213
+ "reward_std": 0.0,
214
+ "rewards/MultiComponentReward/mean": 0.0,
215
+ "rewards/MultiComponentReward/std": 0.0,
216
+ "step": 9,
217
+ "step_time": 643.2193151013926
218
+ },
219
+ {
220
+ "clip_ratio/high_max": 0.0,
221
+ "clip_ratio/high_mean": 0.0,
222
+ "clip_ratio/low_mean": 0.0,
223
+ "clip_ratio/low_min": 0.0,
224
+ "clip_ratio/region_mean": 0.0,
225
+ "completions/clipped_ratio": 1.0,
226
+ "completions/max_length": 2048.0,
227
+ "completions/mean_length": 2048.0,
228
+ "completions/min_length": 2048.0,
229
+ "epoch": 0.02207505518763797,
230
+ "frac_reward_zero_std": 1.0,
231
+ "grad_norm": 0.021022867411375046,
232
+ "kl": 0.03890585224144161,
233
+ "learning_rate": 4.9453690018345144e-06,
234
+ "loss": 0.0015562340850010514,
235
+ "reward": 0.0,
236
+ "reward_std": 0.0,
237
+ "rewards/MultiComponentReward/mean": 0.0,
238
+ "rewards/MultiComponentReward/std": 0.0,
239
+ "step": 10,
240
+ "step_time": 644.6979443421587
241
+ }
242
+ ],
243
+ "logging_steps": 1,
244
+ "max_steps": 150,
245
+ "num_input_tokens_seen": 0,
246
+ "num_train_epochs": 1,
247
+ "save_steps": 5,
248
+ "stateful_callbacks": {
249
+ "TrainerControl": {
250
+ "args": {
251
+ "should_epoch_stop": false,
252
+ "should_evaluate": false,
253
+ "should_log": false,
254
+ "should_save": true,
255
+ "should_training_stop": false
256
+ },
257
+ "attributes": {}
258
+ }
259
+ },
260
+ "total_flos": 0.0,
261
+ "train_batch_size": 1,
262
+ "trial_name": null,
263
+ "trial_params": null
264
+ }