| { | |
| "best_metric": 0.8496279716491699, | |
| "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": 16.404977798461914, | |
| "learning_rate": 6.751086527064401e-07, | |
| "loss": 0.7867, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39354584809130266, | |
| "grad_norm": 13.106764793395996, | |
| "learning_rate": 6.474516001580403e-07, | |
| "loss": 0.7802, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5903187721369539, | |
| "grad_norm": 9.88762092590332, | |
| "learning_rate": 6.197945476096404e-07, | |
| "loss": 0.7919, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7870916961826053, | |
| "grad_norm": 11.42548656463623, | |
| "learning_rate": 5.921374950612407e-07, | |
| "loss": 0.7778, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9838646202282566, | |
| "grad_norm": 10.138823509216309, | |
| "learning_rate": 5.644804425128407e-07, | |
| "loss": 0.7702, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7564585178624151, | |
| "eval_loss": 0.8620501756668091, | |
| "eval_model_preparation_time": 0.0068, | |
| "eval_runtime": 1259.1488, | |
| "eval_samples_per_second": 86.077, | |
| "eval_steps_per_second": 2.69, | |
| "step": 2541 | |
| }, | |
| { | |
| "epoch": 1.1806375442739079, | |
| "grad_norm": 9.205147743225098, | |
| "learning_rate": 5.368233899644408e-07, | |
| "loss": 0.7727, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.3774104683195592, | |
| "grad_norm": 12.592521667480469, | |
| "learning_rate": 5.091663374160411e-07, | |
| "loss": 0.7772, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.5741833923652107, | |
| "grad_norm": 13.292143821716309, | |
| "learning_rate": 4.815092848676412e-07, | |
| "loss": 0.7633, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.770956316410862, | |
| "grad_norm": 11.528284072875977, | |
| "learning_rate": 4.538522323192414e-07, | |
| "loss": 0.7718, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.9677292404565132, | |
| "grad_norm": 7.980528354644775, | |
| "learning_rate": 4.2619517977084157e-07, | |
| "loss": 0.7758, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7576856362562739, | |
| "eval_loss": 0.8577191829681396, | |
| "eval_model_preparation_time": 0.0068, | |
| "eval_runtime": 1266.4814, | |
| "eval_samples_per_second": 85.579, | |
| "eval_steps_per_second": 2.674, | |
| "step": 5082 | |
| }, | |
| { | |
| "epoch": 2.1645021645021645, | |
| "grad_norm": 12.733633995056152, | |
| "learning_rate": 3.9853812722244165e-07, | |
| "loss": 0.7577, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.3612750885478158, | |
| "grad_norm": 10.999346733093262, | |
| "learning_rate": 3.7088107467404185e-07, | |
| "loss": 0.7768, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.558048012593467, | |
| "grad_norm": 6.760105609893799, | |
| "learning_rate": 3.43224022125642e-07, | |
| "loss": 0.7585, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.7548209366391183, | |
| "grad_norm": 9.875105857849121, | |
| "learning_rate": 3.155669695772422e-07, | |
| "loss": 0.7586, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.9515938606847696, | |
| "grad_norm": 11.728323936462402, | |
| "learning_rate": 2.879099170288424e-07, | |
| "loss": 0.7717, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7591803218187186, | |
| "eval_loss": 0.8548997640609741, | |
| "eval_model_preparation_time": 0.0068, | |
| "eval_runtime": 1287.4937, | |
| "eval_samples_per_second": 84.182, | |
| "eval_steps_per_second": 2.631, | |
| "step": 7623 | |
| }, | |
| { | |
| "epoch": 3.1483667847304213, | |
| "grad_norm": 9.85474681854248, | |
| "learning_rate": 2.6025286448044247e-07, | |
| "loss": 0.7569, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.3451397087760726, | |
| "grad_norm": 11.589801788330078, | |
| "learning_rate": 2.3259581193204266e-07, | |
| "loss": 0.753, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.541912632821724, | |
| "grad_norm": 11.029313087463379, | |
| "learning_rate": 2.0493875938364283e-07, | |
| "loss": 0.7605, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.738685556867375, | |
| "grad_norm": 8.829140663146973, | |
| "learning_rate": 1.7728170683524297e-07, | |
| "loss": 0.754, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.9354584809130264, | |
| "grad_norm": 8.928058624267578, | |
| "learning_rate": 1.4962465428684314e-07, | |
| "loss": 0.755, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.759983023324476, | |
| "eval_loss": 0.8506004214286804, | |
| "eval_model_preparation_time": 0.0068, | |
| "eval_runtime": 1280.31, | |
| "eval_samples_per_second": 84.655, | |
| "eval_steps_per_second": 2.645, | |
| "step": 10164 | |
| }, | |
| { | |
| "epoch": 4.132231404958677, | |
| "grad_norm": 10.750606536865234, | |
| "learning_rate": 1.219676017384433e-07, | |
| "loss": 0.7593, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.329004329004329, | |
| "grad_norm": 9.377626419067383, | |
| "learning_rate": 9.431054919004346e-08, | |
| "loss": 0.7369, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.525777253049981, | |
| "grad_norm": 11.895365715026855, | |
| "learning_rate": 6.665349664164362e-08, | |
| "loss": 0.7445, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.7225501770956315, | |
| "grad_norm": 14.032135963439941, | |
| "learning_rate": 3.899644409324377e-08, | |
| "loss": 0.7589, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.919323101141283, | |
| "grad_norm": 12.187383651733398, | |
| "learning_rate": 1.1339391544843935e-08, | |
| "loss": 0.7735, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7599184381458518, | |
| "eval_loss": 0.8496279716491699, | |
| "eval_model_preparation_time": 0.0068, | |
| "eval_runtime": 1275.7742, | |
| "eval_samples_per_second": 84.955, | |
| "eval_steps_per_second": 2.655, | |
| "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 | |
| } | |