File size: 8,736 Bytes
80b1cee |
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 |
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9984,
"eval_steps": 83,
"global_step": 78,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"clip_ratio": 0.0,
"completion_length": 686.6350830078125,
"epoch": 0.064,
"grad_norm": 0.493407666683197,
"kl": 0.00012958049774169922,
"learning_rate": 1.2499999999999999e-06,
"loss": -0.0056,
"reward": 1.0069444552063942,
"reward_std": 0.580621787160635,
"rewards/accuracy_reward": 0.3986111156642437,
"rewards/format_reward": 0.18090277863666415,
"rewards/tag_count_reward": 0.42743055820465087,
"step": 5
},
{
"clip_ratio": 0.0,
"completion_length": 652.9118125915527,
"epoch": 0.128,
"grad_norm": 51.891143798828125,
"kl": 13.452931213378907,
"learning_rate": 1.995974293995239e-06,
"loss": 0.1344,
"reward": 1.3629340469837188,
"reward_std": 0.62072244733572,
"rewards/accuracy_reward": 0.3295138917863369,
"rewards/format_reward": 0.3850694507360458,
"rewards/tag_count_reward": 0.6483506999909878,
"step": 10
},
{
"clip_ratio": 0.0,
"completion_length": 676.9017471313476,
"epoch": 0.192,
"grad_norm": 17.215152740478516,
"kl": 1.034814453125,
"learning_rate": 1.9510565162951534e-06,
"loss": 0.0138,
"reward": 1.5778646022081375,
"reward_std": 0.6299503356218338,
"rewards/accuracy_reward": 0.3361111145466566,
"rewards/format_reward": 0.4927083343267441,
"rewards/tag_count_reward": 0.7490451440215111,
"step": 15
},
{
"clip_ratio": 0.0,
"completion_length": 641.2159820556641,
"epoch": 0.256,
"grad_norm": 0.4615473449230194,
"kl": 0.032623291015625,
"learning_rate": 1.858448793601866e-06,
"loss": 0.0025,
"reward": 1.623437511920929,
"reward_std": 0.6311033435165883,
"rewards/accuracy_reward": 0.3847222276031971,
"rewards/format_reward": 0.5031250067055225,
"rewards/tag_count_reward": 0.7355902805924416,
"step": 20
},
{
"clip_ratio": 0.0,
"completion_length": 653.2670181274414,
"epoch": 0.32,
"grad_norm": 0.2847181260585785,
"kl": 0.04327239990234375,
"learning_rate": 1.7227948638273915e-06,
"loss": 0.0167,
"reward": 1.8901041924953461,
"reward_std": 0.6056328244507313,
"rewards/accuracy_reward": 0.395138892903924,
"rewards/format_reward": 0.6468750037252903,
"rewards/tag_count_reward": 0.8480902835726738,
"step": 25
},
{
"clip_ratio": 0.0,
"completion_length": 616.6788215637207,
"epoch": 0.384,
"grad_norm": 0.27531540393829346,
"kl": 5.2246551513671875,
"learning_rate": 1.5508969814521024e-06,
"loss": 0.0613,
"reward": 2.1147569686174394,
"reward_std": 0.540030462294817,
"rewards/accuracy_reward": 0.37777778282761576,
"rewards/format_reward": 0.8107638955116272,
"rewards/tag_count_reward": 0.9262152895331383,
"step": 30
},
{
"clip_ratio": 0.0,
"completion_length": 638.4989616394043,
"epoch": 0.448,
"grad_norm": 0.2678464651107788,
"kl": 0.0264862060546875,
"learning_rate": 1.3513748240813427e-06,
"loss": 0.0124,
"reward": 2.2046007037162783,
"reward_std": 0.47457179576158526,
"rewards/accuracy_reward": 0.3527777839452028,
"rewards/format_reward": 0.8920138970017433,
"rewards/tag_count_reward": 0.9598090335726738,
"step": 35
},
{
"clip_ratio": 0.0,
"completion_length": 643.7434097290039,
"epoch": 0.512,
"grad_norm": 0.22826902568340302,
"kl": 0.0377349853515625,
"learning_rate": 1.1342332658176555e-06,
"loss": 0.0127,
"reward": 2.335156279802322,
"reward_std": 0.4647583156824112,
"rewards/accuracy_reward": 0.45034722685813905,
"rewards/format_reward": 0.9152777850627899,
"rewards/tag_count_reward": 0.9695312604308128,
"step": 40
},
{
"clip_ratio": 0.0,
"completion_length": 629.6038261413574,
"epoch": 0.576,
"grad_norm": 0.32088664174079895,
"kl": 0.04580535888671875,
"learning_rate": 9.103606910965665e-07,
"loss": 0.008,
"reward": 2.414062523841858,
"reward_std": 0.40865729935467243,
"rewards/accuracy_reward": 0.5135416679084301,
"rewards/format_reward": 0.9260416761040687,
"rewards/tag_count_reward": 0.974479179084301,
"step": 45
},
{
"clip_ratio": 0.0,
"completion_length": 656.8579933166504,
"epoch": 0.64,
"grad_norm": 2.3553261756896973,
"kl": 0.08855743408203125,
"learning_rate": 6.909830056250526e-07,
"loss": 0.0135,
"reward": 2.4554687619209288,
"reward_std": 0.38128191530704497,
"rewards/accuracy_reward": 0.5336805626749992,
"rewards/format_reward": 0.9430555611848831,
"rewards/tag_count_reward": 0.9787326440215111,
"step": 50
},
{
"clip_ratio": 0.0,
"completion_length": 652.9923721313477,
"epoch": 0.704,
"grad_norm": 2.0868406295776367,
"kl": 0.0515777587890625,
"learning_rate": 4.871007225940939e-07,
"loss": 0.015,
"reward": 2.487239581346512,
"reward_std": 0.3608484253287315,
"rewards/accuracy_reward": 0.5642361134290695,
"rewards/format_reward": 0.9427083373069763,
"rewards/tag_count_reward": 0.9802951425313949,
"step": 55
},
{
"clip_ratio": 0.0,
"completion_length": 650.6635467529297,
"epoch": 0.768,
"grad_norm": 0.19568555057048798,
"kl": 0.035595703125,
"learning_rate": 3.0893735101313535e-07,
"loss": 0.0199,
"reward": 2.501822918653488,
"reward_std": 0.35273106172680857,
"rewards/accuracy_reward": 0.5711805634200573,
"rewards/format_reward": 0.947569452226162,
"rewards/tag_count_reward": 0.9830729216337204,
"step": 60
},
{
"clip_ratio": 0.0,
"completion_length": 637.9875061035157,
"epoch": 0.832,
"grad_norm": 0.23244014382362366,
"kl": 0.03612060546875,
"learning_rate": 1.6542674627869734e-07,
"loss": 0.0137,
"reward": 2.5010416865348817,
"reward_std": 0.34778057783842087,
"rewards/accuracy_reward": 0.5725694492459297,
"rewards/format_reward": 0.9472222268581391,
"rewards/tag_count_reward": 0.9812500059604645,
"step": 65
},
{
"clip_ratio": 0.0,
"completion_length": 633.2107666015625,
"epoch": 0.896,
"grad_norm": 71.81304168701172,
"kl": 0.2681488037109375,
"learning_rate": 6.376512936026279e-08,
"loss": 0.0135,
"reward": 2.534635454416275,
"reward_std": 0.3476743776351213,
"rewards/accuracy_reward": 0.5899305626749992,
"rewards/format_reward": 0.9600694462656975,
"rewards/tag_count_reward": 0.9846354201436043,
"step": 70
},
{
"clip_ratio": 0.0,
"completion_length": 632.5708358764648,
"epoch": 0.96,
"grad_norm": 0.19250567257404327,
"kl": 0.038299560546875,
"learning_rate": 9.050238232065299e-09,
"loss": 0.0194,
"reward": 2.515277796983719,
"reward_std": 0.37559727430343626,
"rewards/accuracy_reward": 0.5836805649101734,
"rewards/format_reward": 0.9520833417773247,
"rewards/tag_count_reward": 0.9795139014720917,
"step": 75
},
{
"clip_ratio": 0.0,
"completion_length": 630.8084576924642,
"epoch": 0.9984,
"kl": 0.040715535481770836,
"reward": 2.5170717736085257,
"reward_std": 0.3583106442044179,
"rewards/accuracy_reward": 0.5885416728754839,
"rewards/format_reward": 0.9496527860562006,
"rewards/tag_count_reward": 0.978877325852712,
"step": 78,
"total_flos": 0.0,
"train_loss": 0.023221049398088302,
"train_runtime": 8217.2755,
"train_samples_per_second": 0.913,
"train_steps_per_second": 0.009
}
],
"logging_steps": 5,
"max_steps": 78,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 0.0,
"train_batch_size": 12,
"trial_name": null,
"trial_params": null
}
|