| { | |
| "best_metric": 1.0, | |
| "best_model_checkpoint": "/kaggle/working/new_downloads22/checkpoint-317", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 951, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07886435331230283, | |
| "grad_norm": 8.923051834106445, | |
| "learning_rate": 1.75e-05, | |
| "loss": 9.5858, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.15772870662460567, | |
| "grad_norm": 7.801741600036621, | |
| "learning_rate": 3.8333333333333334e-05, | |
| "loss": 7.4178, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.23659305993690852, | |
| "grad_norm": 6.98698091506958, | |
| "learning_rate": 5.916666666666667e-05, | |
| "loss": 6.5061, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.31545741324921134, | |
| "grad_norm": 15.704840660095215, | |
| "learning_rate": 8e-05, | |
| "loss": 6.3719, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3943217665615142, | |
| "grad_norm": 18.519981384277344, | |
| "learning_rate": 9.9983922829582e-05, | |
| "loss": 6.2496, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.47318611987381703, | |
| "grad_norm": 5.073540210723877, | |
| "learning_rate": 9.958199356913184e-05, | |
| "loss": 6.1319, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5520504731861199, | |
| "grad_norm": 5.464381217956543, | |
| "learning_rate": 9.918006430868168e-05, | |
| "loss": 6.0139, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.6309148264984227, | |
| "grad_norm": 7.70945405960083, | |
| "learning_rate": 9.877813504823152e-05, | |
| "loss": 5.8082, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7097791798107256, | |
| "grad_norm": 10.696139335632324, | |
| "learning_rate": 9.837620578778135e-05, | |
| "loss": 5.4794, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.7886435331230284, | |
| "grad_norm": 11.641499519348145, | |
| "learning_rate": 9.79742765273312e-05, | |
| "loss": 5.0549, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8675078864353313, | |
| "grad_norm": 9.7868013381958, | |
| "learning_rate": 9.757234726688104e-05, | |
| "loss": 4.5482, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.9463722397476341, | |
| "grad_norm": 8.010425567626953, | |
| "learning_rate": 9.717041800643088e-05, | |
| "loss": 4.0112, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cer": 1.0, | |
| "eval_loss": 3.591109037399292, | |
| "eval_runtime": 822.2984, | |
| "eval_samples_per_second": 2.674, | |
| "eval_steps_per_second": 0.022, | |
| "eval_wer": 1.0, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.025236593059937, | |
| "grad_norm": 6.642086982727051, | |
| "learning_rate": 9.676848874598072e-05, | |
| "loss": 3.5042, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.1041009463722398, | |
| "grad_norm": 8.283796310424805, | |
| "learning_rate": 9.636655948553056e-05, | |
| "loss": 2.9869, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.1829652996845426, | |
| "grad_norm": 6.643803596496582, | |
| "learning_rate": 9.596463022508039e-05, | |
| "loss": 2.759, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.2618296529968454, | |
| "grad_norm": 7.243183135986328, | |
| "learning_rate": 9.556270096463023e-05, | |
| "loss": 2.5283, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.3406940063091484, | |
| "grad_norm": 6.6391425132751465, | |
| "learning_rate": 9.516077170418007e-05, | |
| "loss": 2.3984, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.4195583596214512, | |
| "grad_norm": 6.673182010650635, | |
| "learning_rate": 9.47588424437299e-05, | |
| "loss": 2.2356, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.498422712933754, | |
| "grad_norm": 6.318761825561523, | |
| "learning_rate": 9.435691318327976e-05, | |
| "loss": 2.1016, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.5772870662460567, | |
| "grad_norm": 5.9026360511779785, | |
| "learning_rate": 9.39549839228296e-05, | |
| "loss": 2.019, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.6561514195583595, | |
| "grad_norm": 5.887834072113037, | |
| "learning_rate": 9.355305466237943e-05, | |
| "loss": 1.9171, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.7350157728706623, | |
| "grad_norm": 5.294543743133545, | |
| "learning_rate": 9.315112540192927e-05, | |
| "loss": 1.8466, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.8138801261829653, | |
| "grad_norm": 5.4380364418029785, | |
| "learning_rate": 9.27491961414791e-05, | |
| "loss": 1.7883, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.8927444794952681, | |
| "grad_norm": 5.8729777336120605, | |
| "learning_rate": 9.234726688102894e-05, | |
| "loss": 1.6805, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.971608832807571, | |
| "grad_norm": 5.349132537841797, | |
| "learning_rate": 9.194533762057878e-05, | |
| "loss": 1.6442, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 1.0, | |
| "eval_loss": 2.0905776023864746, | |
| "eval_runtime": 809.9594, | |
| "eval_samples_per_second": 2.715, | |
| "eval_steps_per_second": 0.022, | |
| "eval_wer": 1.0, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.050473186119874, | |
| "grad_norm": 4.628177642822266, | |
| "learning_rate": 9.154340836012862e-05, | |
| "loss": 1.3757, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.1293375394321767, | |
| "grad_norm": 4.613383769989014, | |
| "learning_rate": 9.114147909967846e-05, | |
| "loss": 1.2338, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.2082018927444795, | |
| "grad_norm": 4.620671272277832, | |
| "learning_rate": 9.07395498392283e-05, | |
| "loss": 1.1955, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.2870662460567823, | |
| "grad_norm": 5.2364325523376465, | |
| "learning_rate": 9.033762057877814e-05, | |
| "loss": 1.1995, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.365930599369085, | |
| "grad_norm": 4.981749057769775, | |
| "learning_rate": 8.993569131832798e-05, | |
| "loss": 1.1839, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.444794952681388, | |
| "grad_norm": 4.923290252685547, | |
| "learning_rate": 8.953376205787782e-05, | |
| "loss": 1.1529, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.5236593059936907, | |
| "grad_norm": 4.748891830444336, | |
| "learning_rate": 8.913183279742766e-05, | |
| "loss": 1.156, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.6025236593059935, | |
| "grad_norm": 4.635627269744873, | |
| "learning_rate": 8.872990353697749e-05, | |
| "loss": 1.152, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.6813880126182967, | |
| "grad_norm": 4.900181770324707, | |
| "learning_rate": 8.832797427652733e-05, | |
| "loss": 1.1358, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.7602523659305995, | |
| "grad_norm": 4.697984218597412, | |
| "learning_rate": 8.792604501607717e-05, | |
| "loss": 1.1274, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.8391167192429023, | |
| "grad_norm": 4.6810078620910645, | |
| "learning_rate": 8.752411575562702e-05, | |
| "loss": 1.1098, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.917981072555205, | |
| "grad_norm": 4.750789165496826, | |
| "learning_rate": 8.712218649517686e-05, | |
| "loss": 1.0976, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.996845425867508, | |
| "grad_norm": 4.741580486297607, | |
| "learning_rate": 8.67202572347267e-05, | |
| "loss": 1.0836, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_cer": 1.0, | |
| "eval_loss": 1.82232666015625, | |
| "eval_runtime": 816.0044, | |
| "eval_samples_per_second": 2.695, | |
| "eval_steps_per_second": 0.022, | |
| "eval_wer": 1.0, | |
| "step": 951 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 6340, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.7456402765824e+18, | |
| "train_batch_size": 480, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |