File size: 10,457 Bytes
7df5122 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 | {
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 5.0,
"eval_steps": 500,
"global_step": 50,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completion_length": 95.85,
"completions/clipped_ratio": 0.125,
"completions/max_length": 256.0,
"completions/max_terminated_length": 195.6,
"completions/mean_length": 95.85,
"completions/mean_terminated_length": 72.8880989074707,
"completions/min_length": 36.6,
"completions/min_terminated_length": 36.6,
"epoch": 0.5,
"frac_reward_zero_std": 0.2,
"grad_norm": 0.3041034936904907,
"kl": 0.00021898818758927518,
"learning_rate": 4e-05,
"loss": 0.0329527884721756,
"num_tokens": 21444.0,
"reward": 0.7767857074737549,
"reward_std": 0.7150911569595337,
"rewards/reward_fn/mean": 0.7767857074737549,
"rewards/reward_fn/std": 0.9959433197975158,
"step": 5
},
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completion_length": 95.1,
"completions/clipped_ratio": 0.0875,
"completions/max_length": 225.2,
"completions/max_terminated_length": 159.2,
"completions/mean_length": 95.1,
"completions/mean_terminated_length": 80.34466400146485,
"completions/min_length": 34.6,
"completions/min_terminated_length": 34.6,
"epoch": 1.0,
"frac_reward_zero_std": 0.4,
"grad_norm": 0.2708108425140381,
"kl": 0.006061523332027719,
"learning_rate": 9e-05,
"loss": -0.016053691506385803,
"num_tokens": 42220.0,
"reward": 0.6875000119209289,
"reward_std": 0.5175672590732574,
"rewards/reward_fn/mean": 0.6875,
"rewards/reward_fn/std": 0.8034753799438477,
"step": 10
},
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completion_length": 146.275,
"completions/clipped_ratio": 0.325,
"completions/max_length": 256.0,
"completions/max_terminated_length": 225.4,
"completions/mean_length": 146.275,
"completions/mean_terminated_length": 92.8663101196289,
"completions/min_length": 32.6,
"completions/min_terminated_length": 32.6,
"epoch": 1.5,
"frac_reward_zero_std": 0.2,
"grad_norm": 0.27529817819595337,
"kl": 0.01560373855754733,
"learning_rate": 9.989068136093873e-05,
"loss": 0.010054035484790802,
"num_tokens": 69198.0,
"reward": 1.1558928728103637,
"reward_std": 0.8465379416942597,
"rewards/reward_fn/mean": 1.1558928728103637,
"rewards/reward_fn/std": 1.1404805421829223,
"step": 15
},
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completion_length": 75.85,
"completions/clipped_ratio": 0.05,
"completions/max_length": 220.6,
"completions/max_terminated_length": 151.0,
"completions/mean_length": 75.85,
"completions/mean_terminated_length": 66.14857330322266,
"completions/min_length": 32.8,
"completions/min_terminated_length": 32.8,
"epoch": 2.0,
"frac_reward_zero_std": 0.35,
"grad_norm": 0.353487491607666,
"kl": 0.027986684162169696,
"learning_rate": 9.944739353007344e-05,
"loss": -0.030885905027389526,
"num_tokens": 87334.0,
"reward": 0.7932142853736878,
"reward_std": 0.4118983089923859,
"rewards/reward_fn/mean": 0.7932142853736878,
"rewards/reward_fn/std": 0.7167340397834778,
"step": 20
},
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completion_length": 90.85,
"completions/clipped_ratio": 0.0625,
"completions/max_length": 248.0,
"completions/max_terminated_length": 216.4,
"completions/mean_length": 90.85,
"completions/mean_terminated_length": 79.819287109375,
"completions/min_length": 28.0,
"completions/min_terminated_length": 28.0,
"epoch": 2.5,
"frac_reward_zero_std": 0.15,
"grad_norm": 0.3318491578102112,
"kl": 0.04692814685404301,
"learning_rate": 9.86663298624003e-05,
"loss": 0.05783940553665161,
"num_tokens": 108330.0,
"reward": 1.1987500190734863,
"reward_std": 0.5907092690467834,
"rewards/reward_fn/mean": 1.1987500190734863,
"rewards/reward_fn/std": 0.9482298731803894,
"step": 25
},
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completion_length": 91.775,
"completions/clipped_ratio": 0.125,
"completions/max_length": 248.2,
"completions/max_terminated_length": 212.6,
"completions/mean_length": 91.775,
"completions/mean_terminated_length": 68.68722763061524,
"completions/min_length": 25.2,
"completions/min_terminated_length": 25.2,
"epoch": 3.0,
"frac_reward_zero_std": 0.4,
"grad_norm": 0.24605490267276764,
"kl": 0.07254143096506596,
"learning_rate": 9.755282581475769e-05,
"loss": 0.03934891819953919,
"num_tokens": 129012.0,
"reward": 1.0242857217788697,
"reward_std": 0.5802445828914642,
"rewards/reward_fn/mean": 1.0242857217788697,
"rewards/reward_fn/std": 0.8706011414527893,
"step": 30
},
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completion_length": 59.7875,
"completions/clipped_ratio": 0.0375,
"completions/max_length": 175.6,
"completions/max_terminated_length": 118.4,
"completions/mean_length": 59.7875,
"completions/mean_terminated_length": 52.19559555053711,
"completions/min_length": 25.4,
"completions/min_terminated_length": 25.4,
"epoch": 3.5,
"frac_reward_zero_std": 0.6,
"grad_norm": 0.2970605790615082,
"kl": 0.053099883161485194,
"learning_rate": 9.611448774886924e-05,
"loss": 0.06977529525756836,
"num_tokens": 147103.0,
"reward": 0.9473214387893677,
"reward_std": 0.3923970699310303,
"rewards/reward_fn/mean": 0.9473214268684387,
"rewards/reward_fn/std": 0.649278998374939,
"step": 35
},
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completion_length": 47.85,
"completions/clipped_ratio": 0.0,
"completions/max_length": 94.8,
"completions/max_terminated_length": 94.8,
"completions/mean_length": 47.85,
"completions/mean_terminated_length": 47.85,
"completions/min_length": 31.2,
"completions/min_terminated_length": 31.2,
"epoch": 4.0,
"frac_reward_zero_std": 0.85,
"grad_norm": 0.16646049916744232,
"kl": 0.056573960930109024,
"learning_rate": 9.43611409721806e-05,
"loss": -0.0031644195318222047,
"num_tokens": 164643.0,
"reward": 0.7899999976158142,
"reward_std": 0.12681145668029786,
"rewards/reward_fn/mean": 0.7899999976158142,
"rewards/reward_fn/std": 0.5133758962154389,
"step": 40
},
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completion_length": 49.2625,
"completions/clipped_ratio": 0.0,
"completions/max_length": 93.6,
"completions/max_terminated_length": 93.6,
"completions/mean_length": 49.2625,
"completions/mean_terminated_length": 49.2625,
"completions/min_length": 33.8,
"completions/min_terminated_length": 33.8,
"epoch": 4.5,
"frac_reward_zero_std": 0.6,
"grad_norm": 0.3159082531929016,
"kl": 0.0353464649990201,
"learning_rate": 9.230476262104677e-05,
"loss": -0.03417131304740906,
"num_tokens": 182568.0,
"reward": 0.7924999833106995,
"reward_std": 0.3821385264396667,
"rewards/reward_fn/mean": 0.7924999833106995,
"rewards/reward_fn/std": 0.62625293135643,
"step": 45
},
{
"clip_ratio/high_max": 0.0,
"clip_ratio/high_mean": 0.0,
"clip_ratio/low_mean": 0.0,
"clip_ratio/low_min": 0.0,
"clip_ratio/region_mean": 0.0,
"completion_length": 58.25,
"completions/clipped_ratio": 0.0125,
"completions/max_length": 180.6,
"completions/max_terminated_length": 153.2,
"completions/mean_length": 58.25,
"completions/mean_terminated_length": 55.67250061035156,
"completions/min_length": 33.2,
"completions/min_terminated_length": 33.2,
"epoch": 5.0,
"frac_reward_zero_std": 0.55,
"grad_norm": 0.27360936999320984,
"kl": 0.04483541352674365,
"learning_rate": 8.995939984474624e-05,
"loss": 0.026852786540985107,
"num_tokens": 200660.0,
"reward": 0.774821424484253,
"reward_std": 0.32772802114486693,
"rewards/reward_fn/mean": 0.774821424484253,
"rewards/reward_fn/std": 0.5432307064533234,
"step": 50
}
],
"logging_steps": 5,
"max_steps": 200,
"num_input_tokens_seen": 200660,
"num_train_epochs": 20,
"save_steps": 50,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 0.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}
|