|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 1.0, |
|
"eval_steps": 500, |
|
"global_step": 325, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0, |
|
"grad_norm": 0.26868827253641747, |
|
"learning_rate": 6.060606060606061e-06, |
|
"loss": 0.7782, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"grad_norm": 0.21382632169384555, |
|
"learning_rate": 3.0303030303030306e-05, |
|
"loss": 0.7798, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"grad_norm": 0.22445606867737244, |
|
"learning_rate": 6.060606060606061e-05, |
|
"loss": 0.7851, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"grad_norm": 0.2072105558411773, |
|
"learning_rate": 9.090909090909092e-05, |
|
"loss": 0.7855, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 0.06, |
|
"grad_norm": 0.17995218448427217, |
|
"learning_rate": 0.00012121212121212122, |
|
"loss": 0.792, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.08, |
|
"grad_norm": 0.226508952573423, |
|
"learning_rate": 0.00015151515151515152, |
|
"loss": 0.7382, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"grad_norm": 0.18958014075029098, |
|
"learning_rate": 0.00018181818181818183, |
|
"loss": 0.7419, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"grad_norm": 0.19949889749656097, |
|
"learning_rate": 0.00019997685019798912, |
|
"loss": 0.7642, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 0.12, |
|
"grad_norm": 0.19567934772580736, |
|
"learning_rate": 0.0001997165380022878, |
|
"loss": 0.7607, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"grad_norm": 0.18969776374927835, |
|
"learning_rate": 0.000199167731989929, |
|
"loss": 0.795, |
|
"step": 45 |
|
}, |
|
{ |
|
"epoch": 0.15, |
|
"grad_norm": 0.1942706799722158, |
|
"learning_rate": 0.0001983320199330545, |
|
"loss": 0.764, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.17, |
|
"grad_norm": 0.19464988909031225, |
|
"learning_rate": 0.00019721181966290613, |
|
"loss": 0.7836, |
|
"step": 55 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"grad_norm": 0.1849352970071158, |
|
"learning_rate": 0.00019581037207470382, |
|
"loss": 0.757, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"grad_norm": 0.17976128059849958, |
|
"learning_rate": 0.00019413173175128473, |
|
"loss": 0.7616, |
|
"step": 65 |
|
}, |
|
{ |
|
"epoch": 0.22, |
|
"grad_norm": 0.1934718609207969, |
|
"learning_rate": 0.00019218075523263104, |
|
"loss": 0.7538, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"grad_norm": 0.16955946612078765, |
|
"learning_rate": 0.00018996308696522433, |
|
"loss": 0.7617, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"grad_norm": 0.19860710500579018, |
|
"learning_rate": 0.00018748514297187648, |
|
"loss": 0.7995, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"grad_norm": 0.17834039868116905, |
|
"learning_rate": 0.00018475409228928312, |
|
"loss": 0.7539, |
|
"step": 85 |
|
}, |
|
{ |
|
"epoch": 0.28, |
|
"grad_norm": 0.1889302979453787, |
|
"learning_rate": 0.00018177783622700327, |
|
"loss": 0.7754, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"grad_norm": 0.17928302611917235, |
|
"learning_rate": 0.00017856498550787144, |
|
"loss": 0.7924, |
|
"step": 95 |
|
}, |
|
{ |
|
"epoch": 0.31, |
|
"grad_norm": 0.18965987325021239, |
|
"learning_rate": 0.00017512483535597867, |
|
"loss": 0.7547, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"grad_norm": 0.16948470382160882, |
|
"learning_rate": 0.00017146733860429612, |
|
"loss": 0.7755, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"grad_norm": 0.20373009441973453, |
|
"learning_rate": 0.0001676030768997445, |
|
"loss": 0.7452, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.35, |
|
"grad_norm": 0.18030063668475196, |
|
"learning_rate": 0.00016354323008901776, |
|
"loss": 0.7499, |
|
"step": 115 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"grad_norm": 0.16132188005262432, |
|
"learning_rate": 0.00015929954387373103, |
|
"loss": 0.7344, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.38, |
|
"grad_norm": 0.2014655387761585, |
|
"learning_rate": 0.00015488429582847192, |
|
"loss": 0.7909, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 0.4, |
|
"grad_norm": 0.19701253442033323, |
|
"learning_rate": 0.00015031025988006936, |
|
"loss": 0.7846, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"grad_norm": 0.18599715688382074, |
|
"learning_rate": 0.00014559066935084588, |
|
"loss": 0.7884, |
|
"step": 135 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"grad_norm": 0.15916898081839567, |
|
"learning_rate": 0.00014073917867277557, |
|
"loss": 0.7478, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"grad_norm": 0.1693289847957747, |
|
"learning_rate": 0.0001357698238833126, |
|
"loss": 0.758, |
|
"step": 145 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"grad_norm": 0.1724296952005482, |
|
"learning_rate": 0.000130696982017182, |
|
"loss": 0.7349, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"grad_norm": 0.20365543265357713, |
|
"learning_rate": 0.0001255353295116187, |
|
"loss": 0.7581, |
|
"step": 155 |
|
}, |
|
{ |
|
"epoch": 0.49, |
|
"grad_norm": 0.17925645984310104, |
|
"learning_rate": 0.00012029979974539234, |
|
"loss": 0.7428, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.51, |
|
"grad_norm": 0.1849261223570645, |
|
"learning_rate": 0.00011500553983446527, |
|
"loss": 0.7915, |
|
"step": 165 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"grad_norm": 0.17147514694965366, |
|
"learning_rate": 0.00010966786680927874, |
|
"loss": 0.7282, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"grad_norm": 0.19718678657625408, |
|
"learning_rate": 0.00010430222330045304, |
|
"loss": 0.7807, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 0.55, |
|
"grad_norm": 0.1697756132509757, |
|
"learning_rate": 9.892413286110886e-05, |
|
"loss": 0.7522, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"grad_norm": 0.1745926097752596, |
|
"learning_rate": 9.354915505506839e-05, |
|
"loss": 0.7656, |
|
"step": 185 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"grad_norm": 0.17991554784663008, |
|
"learning_rate": 8.81928404408726e-05, |
|
"loss": 0.7425, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.6, |
|
"grad_norm": 0.1638975774298517, |
|
"learning_rate": 8.287068558185225e-05, |
|
"loss": 0.785, |
|
"step": 195 |
|
}, |
|
{ |
|
"epoch": 0.62, |
|
"grad_norm": 0.1721376274882642, |
|
"learning_rate": 7.759808821241406e-05, |
|
"loss": 0.747, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"grad_norm": 0.16645909181505114, |
|
"learning_rate": 7.239030269025311e-05, |
|
"loss": 0.7436, |
|
"step": 205 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"grad_norm": 0.17668278165163676, |
|
"learning_rate": 6.726239586337408e-05, |
|
"loss": 0.76, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"grad_norm": 0.1592821063866204, |
|
"learning_rate": 6.22292034796035e-05, |
|
"loss": 0.7532, |
|
"step": 215 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"grad_norm": 0.18211269157973645, |
|
"learning_rate": 5.730528726470792e-05, |
|
"loss": 0.7684, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"grad_norm": 0.1611844005218168, |
|
"learning_rate": 5.2504892793295e-05, |
|
"loss": 0.7432, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 0.71, |
|
"grad_norm": 0.1770721524356504, |
|
"learning_rate": 4.7841908274384616e-05, |
|
"loss": 0.7719, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 0.72, |
|
"grad_norm": 0.17458138059176753, |
|
"learning_rate": 4.332982437088825e-05, |
|
"loss": 0.7153, |
|
"step": 235 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"grad_norm": 0.1764916869298734, |
|
"learning_rate": 3.898169516924398e-05, |
|
"loss": 0.7509, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"grad_norm": 0.17474182112370204, |
|
"learning_rate": 3.4810100412128747e-05, |
|
"loss": 0.7562, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"grad_norm": 0.1741562771880736, |
|
"learning_rate": 3.0827109103512643e-05, |
|
"loss": 0.7669, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.78, |
|
"grad_norm": 0.17549490206623847, |
|
"learning_rate": 2.7044244591351232e-05, |
|
"loss": 0.7592, |
|
"step": 255 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"grad_norm": 0.16399003172635285, |
|
"learning_rate": 2.3472451228937253e-05, |
|
"loss": 0.7396, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"grad_norm": 0.1713214529866966, |
|
"learning_rate": 2.0122062711363532e-05, |
|
"loss": 0.7766, |
|
"step": 265 |
|
}, |
|
{ |
|
"epoch": 0.83, |
|
"grad_norm": 0.17574824064745212, |
|
"learning_rate": 1.7002772178705716e-05, |
|
"loss": 0.7329, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"grad_norm": 0.16897840325162736, |
|
"learning_rate": 1.4123604172419713e-05, |
|
"loss": 0.7723, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"grad_norm": 0.18239238197655358, |
|
"learning_rate": 1.149288852608743e-05, |
|
"loss": 0.7558, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"grad_norm": 0.16446806582557563, |
|
"learning_rate": 9.118236266049707e-06, |
|
"loss": 0.731, |
|
"step": 285 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"grad_norm": 0.17024389183471494, |
|
"learning_rate": 7.0065175916482095e-06, |
|
"loss": 0.7599, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"grad_norm": 0.15264356146830035, |
|
"learning_rate": 5.163841998782837e-06, |
|
"loss": 0.7644, |
|
"step": 295 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"grad_norm": 0.16324609453084554, |
|
"learning_rate": 3.595540604290437e-06, |
|
"loss": 0.7581, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.94, |
|
"grad_norm": 0.1710138375738759, |
|
"learning_rate": 2.30615072228183e-06, |
|
"loss": 0.7332, |
|
"step": 305 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"grad_norm": 0.1611044475082952, |
|
"learning_rate": 1.2994027370611173e-06, |
|
"loss": 0.7627, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"grad_norm": 0.1514646487097857, |
|
"learning_rate": 5.782093106048159e-07, |
|
"loss": 0.7366, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"grad_norm": 0.17261842858442877, |
|
"learning_rate": 1.446569558255395e-07, |
|
"loss": 0.7436, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 0.16882560969081586, |
|
"learning_rate": 0.0, |
|
"loss": 0.7491, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_loss": 0.7773829698562622, |
|
"eval_runtime": 46.5077, |
|
"eval_samples_per_second": 4.967, |
|
"eval_steps_per_second": 0.172, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"step": 325, |
|
"total_flos": 4962652747988992.0, |
|
"train_loss": 0.7601131853690514, |
|
"train_runtime": 11993.6222, |
|
"train_samples_per_second": 1.733, |
|
"train_steps_per_second": 0.027 |
|
} |
|
], |
|
"logging_steps": 5, |
|
"max_steps": 325, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 1, |
|
"save_steps": 100, |
|
"total_flos": 4962652747988992.0, |
|
"train_batch_size": 8, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|