| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 19.336166381835938, | |
| "learning_rate": 4.999859193643945e-05, | |
| "loss": 2.4671, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 10.850088119506836, | |
| "learning_rate": 4.9949326364020314e-05, | |
| "loss": 2.4246, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 10.342315673828125, | |
| "learning_rate": 4.982981614472039e-05, | |
| "loss": 2.5563, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 11.594865798950195, | |
| "learning_rate": 4.9640397758692715e-05, | |
| "loss": 2.5333, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 9.311284065246582, | |
| "learning_rate": 4.938160451202521e-05, | |
| "loss": 2.5209, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 16.79537010192871, | |
| "learning_rate": 4.9054165035221236e-05, | |
| "loss": 2.8026, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 10.748820304870605, | |
| "learning_rate": 4.865900123174574e-05, | |
| "loss": 2.8341, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 11.450349807739258, | |
| "learning_rate": 4.819722568241274e-05, | |
| "loss": 2.7475, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 9.321686744689941, | |
| "learning_rate": 4.767013851292212e-05, | |
| "loss": 2.7298, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 9.671792984008789, | |
| "learning_rate": 4.7079223733365237e-05, | |
| "loss": 2.6166, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 8.407944679260254, | |
| "learning_rate": 4.642614506000523e-05, | |
| "loss": 2.5359, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 8.98971939086914, | |
| "learning_rate": 4.571274123109606e-05, | |
| "loss": 2.479, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 8.909531593322754, | |
| "learning_rate": 4.4941020829928245e-05, | |
| "loss": 2.4882, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 7.584418296813965, | |
| "learning_rate": 4.4113156629677316e-05, | |
| "loss": 2.3675, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 8.56999683380127, | |
| "learning_rate": 4.323147947597666e-05, | |
| "loss": 2.4285, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 9.06714153289795, | |
| "learning_rate": 4.229847172443866e-05, | |
| "loss": 2.5075, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 7.845032691955566, | |
| "learning_rate": 4.131676025160047e-05, | |
| "loss": 2.4794, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 7.793240070343018, | |
| "learning_rate": 4.028910905897229e-05, | |
| "loss": 2.5271, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 7.8635029792785645, | |
| "learning_rate": 3.9218411491011176e-05, | |
| "loss": 2.4032, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 8.155007362365723, | |
| "learning_rate": 3.8107682088930794e-05, | |
| "loss": 2.4891, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 8.073945999145508, | |
| "learning_rate": 3.6960048103282566e-05, | |
| "loss": 1.412, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 8.153766632080078, | |
| "learning_rate": 3.577874068920446e-05, | |
| "loss": 1.3875, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 6.7796630859375, | |
| "learning_rate": 3.456708580912725e-05, | |
| "loss": 1.3676, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 7.609086513519287, | |
| "learning_rate": 3.332849486855144e-05, | |
| "loss": 1.3438, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 7.6227126121521, | |
| "learning_rate": 3.2066455111259945e-05, | |
| "loss": 1.3895, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 7.66270637512207, | |
| "learning_rate": 3.0784519801008546e-05, | |
| "loss": 1.3399, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 7.801215171813965, | |
| "learning_rate": 2.9486298217337637e-05, | |
| "loss": 1.3101, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 7.925507545471191, | |
| "learning_rate": 2.8175445493671972e-05, | |
| "loss": 1.4303, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 7.386769771575928, | |
| "learning_rate": 2.6855652326319063e-05, | |
| "loss": 1.3833, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 7.175492763519287, | |
| "learning_rate": 2.5530634583340592e-05, | |
| "loss": 1.3684, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 7.5672607421875, | |
| "learning_rate": 2.4204122842552896e-05, | |
| "loss": 1.4123, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 7.482105731964111, | |
| "learning_rate": 2.287985188811228e-05, | |
| "loss": 1.3913, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 7.868821144104004, | |
| "learning_rate": 2.1561550195257378e-05, | |
| "loss": 1.4338, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 7.905065059661865, | |
| "learning_rate": 2.025292943281429e-05, | |
| "loss": 1.3506, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 8.045072555541992, | |
| "learning_rate": 1.895767401301994e-05, | |
| "loss": 1.4525, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 6.936304092407227, | |
| "learning_rate": 1.7679430718086243e-05, | |
| "loss": 1.3534, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 7.464107990264893, | |
| "learning_rate": 1.6421798432711345e-05, | |
| "loss": 1.322, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 7.77186393737793, | |
| "learning_rate": 1.5188318011445906e-05, | |
| "loss": 1.3326, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 7.153684139251709, | |
| "learning_rate": 1.3982462309442889e-05, | |
| "loss": 1.3804, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 6.911492824554443, | |
| "learning_rate": 1.2807626404659142e-05, | |
| "loss": 1.3938, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 5.748324394226074, | |
| "learning_rate": 1.1667118039038064e-05, | |
| "loss": 0.732, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 7.109584808349609, | |
| "learning_rate": 1.0564148305586296e-05, | |
| "loss": 0.6639, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 7.395934104919434, | |
| "learning_rate": 9.501822607564678e-06, | |
| "loss": 0.5978, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 8.013920783996582, | |
| "learning_rate": 8.483131915247968e-06, | |
| "loss": 0.6269, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 6.068602561950684, | |
| "learning_rate": 7.510944344869774e-06, | |
| "loss": 0.6189, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 5.850319862365723, | |
| "learning_rate": 6.587997083462197e-06, | |
| "loss": 0.6096, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 5.7608642578125, | |
| "learning_rate": 5.7168886823256355e-06, | |
| "loss": 0.5799, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 6.143069267272949, | |
| "learning_rate": 4.900071740826489e-06, | |
| "loss": 0.6156, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 5.740133285522461, | |
| "learning_rate": 4.139846001121514e-06, | |
| "loss": 0.6387, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 6.231141567230225, | |
| "learning_rate": 3.438351873250492e-06, | |
| "loss": 0.6032, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 6.172192096710205, | |
| "learning_rate": 2.797564408827319e-06, | |
| "loss": 0.5976, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 6.317423343658447, | |
| "learning_rate": 2.219287740296605e-06, | |
| "loss": 0.5687, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 9.063169479370117, | |
| "learning_rate": 1.7051500014118454e-06, | |
| "loss": 0.5843, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 6.360856056213379, | |
| "learning_rate": 1.2565987432367033e-06, | |
| "loss": 0.5979, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 6.228647232055664, | |
| "learning_rate": 8.748968585755274e-07, | |
| "loss": 0.5601, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 6.196378231048584, | |
| "learning_rate": 5.611190263078464e-07, | |
| "loss": 0.6004, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 6.535120964050293, | |
| "learning_rate": 3.1614868563778487e-07, | |
| "loss": 0.6026, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 6.039511203765869, | |
| "learning_rate": 1.406755487774386e-07, | |
| "loss": 0.6162, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 6.509077548980713, | |
| "learning_rate": 3.5193659067131455e-08, | |
| "loss": 0.6627, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 6.356975555419922, | |
| "learning_rate": 0.0, | |
| "loss": 0.5781, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2138970911670272.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |