| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 360, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 1.9213794469833374, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9704, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 2.6781961917877197, | |
| "learning_rate": 9.999789047591562e-05, | |
| "loss": 0.4178, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 1.419345498085022, | |
| "learning_rate": 9.974496289936769e-05, | |
| "loss": 0.354, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 0.947891116142273, | |
| "learning_rate": 9.90725746626209e-05, | |
| "loss": 0.3275, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.3888888888888888, | |
| "grad_norm": 1.2710075378417969, | |
| "learning_rate": 9.798639549376945e-05, | |
| "loss": 0.2647, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 1.4519425630569458, | |
| "learning_rate": 9.64955842986544e-05, | |
| "loss": 0.2325, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.9444444444444444, | |
| "grad_norm": 1.5266393423080444, | |
| "learning_rate": 9.46127119309197e-05, | |
| "loss": 0.2325, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 1.442086100578308, | |
| "learning_rate": 9.23536551917611e-05, | |
| "loss": 0.2141, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 1.3214430809020996, | |
| "learning_rate": 8.9737462953185e-05, | |
| "loss": 0.1961, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.7777777777777777, | |
| "grad_norm": 1.3879374265670776, | |
| "learning_rate": 8.678619553365659e-05, | |
| "loss": 0.1953, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.0555555555555554, | |
| "grad_norm": 1.3300167322158813, | |
| "learning_rate": 8.352473868055746e-05, | |
| "loss": 0.1732, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.8237183690071106, | |
| "learning_rate": 7.998059372799409e-05, | |
| "loss": 0.1702, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.611111111111111, | |
| "grad_norm": 1.216146469116211, | |
| "learning_rate": 7.618364569939391e-05, | |
| "loss": 0.1519, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.888888888888889, | |
| "grad_norm": 0.9486384987831116, | |
| "learning_rate": 7.21659113102993e-05, | |
| "loss": 0.1566, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "grad_norm": 0.7426595687866211, | |
| "learning_rate": 6.796126899625688e-05, | |
| "loss": 0.1495, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 0.8088931441307068, | |
| "learning_rate": 6.360517324226676e-05, | |
| "loss": 0.1478, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.722222222222222, | |
| "grad_norm": 1.0129106044769287, | |
| "learning_rate": 5.9134355622630356e-05, | |
| "loss": 0.1327, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.8650373220443726, | |
| "learning_rate": 5.458651507209518e-05, | |
| "loss": 0.1381, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.277777777777778, | |
| "grad_norm": 0.7770701050758362, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1403, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.555555555555555, | |
| "grad_norm": 0.7225720286369324, | |
| "learning_rate": 4.541348492790482e-05, | |
| "loss": 0.1359, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.833333333333333, | |
| "grad_norm": 1.0390045642852783, | |
| "learning_rate": 4.086564437736966e-05, | |
| "loss": 0.1274, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.111111111111111, | |
| "grad_norm": 1.121009349822998, | |
| "learning_rate": 3.639482675773324e-05, | |
| "loss": 0.1117, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 6.388888888888889, | |
| "grad_norm": 0.7137916088104248, | |
| "learning_rate": 3.203873100374314e-05, | |
| "loss": 0.1101, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.5132378935813904, | |
| "learning_rate": 2.783408868970071e-05, | |
| "loss": 0.1157, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.944444444444445, | |
| "grad_norm": 0.5895885229110718, | |
| "learning_rate": 2.381635430060611e-05, | |
| "loss": 0.1041, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 7.222222222222222, | |
| "grad_norm": 0.5823981165885925, | |
| "learning_rate": 2.0019406272005915e-05, | |
| "loss": 0.1066, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.6076810359954834, | |
| "learning_rate": 1.6475261319442553e-05, | |
| "loss": 0.1131, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.777777777777778, | |
| "grad_norm": 0.5810424089431763, | |
| "learning_rate": 1.3213804466343421e-05, | |
| "loss": 0.0964, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 8.055555555555555, | |
| "grad_norm": 0.5393444299697876, | |
| "learning_rate": 1.0262537046815018e-05, | |
| "loss": 0.0971, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "grad_norm": 0.5941327214241028, | |
| "learning_rate": 7.646344808238903e-06, | |
| "loss": 0.0933, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 8.61111111111111, | |
| "grad_norm": 0.4310052692890167, | |
| "learning_rate": 5.387288069080299e-06, | |
| "loss": 0.1067, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 0.4362979233264923, | |
| "learning_rate": 3.504415701345615e-06, | |
| "loss": 0.0891, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 9.166666666666666, | |
| "grad_norm": 0.46284258365631104, | |
| "learning_rate": 2.013604506230554e-06, | |
| "loss": 0.1027, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 9.444444444444445, | |
| "grad_norm": 0.5020444393157959, | |
| "learning_rate": 9.274253373791064e-07, | |
| "loss": 0.0852, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 9.722222222222221, | |
| "grad_norm": 0.5854896903038025, | |
| "learning_rate": 2.55037100632316e-07, | |
| "loss": 0.0919, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.7395591735839844, | |
| "learning_rate": 2.109524084381587e-09, | |
| "loss": 0.0918, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 360, | |
| "total_flos": 0.0, | |
| "train_loss": 0.1817794324623214, | |
| "train_runtime": 394.9899, | |
| "train_samples_per_second": 44.432, | |
| "train_steps_per_second": 0.911 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 360, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 3000, | |
| "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": 49, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |