| { | |
| "best_metric": 0.9104332327842712, | |
| "best_model_checkpoint": "cat_breed_image_detection/checkpoint-12705", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 12705, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19677292404565133, | |
| "grad_norm": 10.679129600524902, | |
| "learning_rate": 7.715527459502173e-07, | |
| "loss": 0.8716, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39354584809130266, | |
| "grad_norm": 12.624375343322754, | |
| "learning_rate": 7.399446858949032e-07, | |
| "loss": 0.8608, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5903187721369539, | |
| "grad_norm": 12.887909889221191, | |
| "learning_rate": 7.08336625839589e-07, | |
| "loss": 0.8653, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7870916961826053, | |
| "grad_norm": 14.339140892028809, | |
| "learning_rate": 6.76728565784275e-07, | |
| "loss": 0.8493, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9838646202282566, | |
| "grad_norm": 10.709071159362793, | |
| "learning_rate": 6.451205057289608e-07, | |
| "loss": 0.8589, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7400815618541482, | |
| "eval_loss": 0.9309104084968567, | |
| "eval_model_preparation_time": 0.006, | |
| "eval_runtime": 1193.8367, | |
| "eval_samples_per_second": 90.786, | |
| "eval_steps_per_second": 2.837, | |
| "step": 2541 | |
| }, | |
| { | |
| "epoch": 1.1806375442739079, | |
| "grad_norm": 8.929093360900879, | |
| "learning_rate": 6.135124456736468e-07, | |
| "loss": 0.8467, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.3774104683195592, | |
| "grad_norm": 11.412522315979004, | |
| "learning_rate": 5.819043856183327e-07, | |
| "loss": 0.8531, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.5741833923652107, | |
| "grad_norm": 12.86628532409668, | |
| "learning_rate": 5.502963255630185e-07, | |
| "loss": 0.8443, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.770956316410862, | |
| "grad_norm": 10.795392990112305, | |
| "learning_rate": 5.186882655077045e-07, | |
| "loss": 0.8317, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.9677292404565132, | |
| "grad_norm": 12.041897773742676, | |
| "learning_rate": 4.870802054523904e-07, | |
| "loss": 0.8554, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7436983318571007, | |
| "eval_loss": 0.9218501448631287, | |
| "eval_model_preparation_time": 0.006, | |
| "eval_runtime": 1215.8937, | |
| "eval_samples_per_second": 89.139, | |
| "eval_steps_per_second": 2.786, | |
| "step": 5082 | |
| }, | |
| { | |
| "epoch": 2.1645021645021645, | |
| "grad_norm": 8.663507461547852, | |
| "learning_rate": 4.5547214539707617e-07, | |
| "loss": 0.8319, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.3612750885478158, | |
| "grad_norm": 13.493363380432129, | |
| "learning_rate": 4.238640853417621e-07, | |
| "loss": 0.8396, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.558048012593467, | |
| "grad_norm": 18.810260772705078, | |
| "learning_rate": 3.92256025286448e-07, | |
| "loss": 0.8415, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.7548209366391183, | |
| "grad_norm": 11.748995780944824, | |
| "learning_rate": 3.606479652311339e-07, | |
| "loss": 0.8427, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.9515938606847696, | |
| "grad_norm": 9.686206817626953, | |
| "learning_rate": 3.290399051758198e-07, | |
| "loss": 0.8367, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7437629170357248, | |
| "eval_loss": 0.9151368737220764, | |
| "eval_model_preparation_time": 0.006, | |
| "eval_runtime": 1207.1063, | |
| "eval_samples_per_second": 89.788, | |
| "eval_steps_per_second": 2.806, | |
| "step": 7623 | |
| }, | |
| { | |
| "epoch": 3.1483667847304213, | |
| "grad_norm": 10.070638656616211, | |
| "learning_rate": 2.974318451205057e-07, | |
| "loss": 0.8266, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.3451397087760726, | |
| "grad_norm": 8.807821273803711, | |
| "learning_rate": 2.658237850651916e-07, | |
| "loss": 0.8258, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.541912632821724, | |
| "grad_norm": 6.973104953765869, | |
| "learning_rate": 2.342157250098775e-07, | |
| "loss": 0.8382, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.738685556867375, | |
| "grad_norm": 9.397773742675781, | |
| "learning_rate": 2.026076649545634e-07, | |
| "loss": 0.8194, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.9354584809130264, | |
| "grad_norm": 9.227031707763672, | |
| "learning_rate": 1.709996048992493e-07, | |
| "loss": 0.8333, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7457742840271627, | |
| "eval_loss": 0.9115355610847473, | |
| "eval_model_preparation_time": 0.006, | |
| "eval_runtime": 1209.7892, | |
| "eval_samples_per_second": 89.589, | |
| "eval_steps_per_second": 2.8, | |
| "step": 10164 | |
| }, | |
| { | |
| "epoch": 4.132231404958677, | |
| "grad_norm": 10.666913032531738, | |
| "learning_rate": 1.393915448439352e-07, | |
| "loss": 0.8311, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.329004329004329, | |
| "grad_norm": 8.299125671386719, | |
| "learning_rate": 1.077834847886211e-07, | |
| "loss": 0.8312, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.525777253049981, | |
| "grad_norm": 10.474702835083008, | |
| "learning_rate": 7.617542473330699e-08, | |
| "loss": 0.8224, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.7225501770956315, | |
| "grad_norm": 9.30624008178711, | |
| "learning_rate": 4.4567364677992886e-08, | |
| "loss": 0.8169, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.919323101141283, | |
| "grad_norm": 11.784387588500977, | |
| "learning_rate": 1.2959304622678783e-08, | |
| "loss": 0.8285, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7457742840271627, | |
| "eval_loss": 0.9104332327842712, | |
| "eval_model_preparation_time": 0.006, | |
| "eval_runtime": 1215.0134, | |
| "eval_samples_per_second": 89.204, | |
| "eval_steps_per_second": 2.788, | |
| "step": 12705 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 12705, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 6.301765894857818e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |