| { | |
| "best_metric": 0.2657858431339264, | |
| "best_model_checkpoint": "deepfake_vs_real_image_detection/checkpoint-7998", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 7998, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18754688672168043, | |
| "grad_norm": 3.8869080543518066, | |
| "learning_rate": 9.43381982888777e-07, | |
| "loss": 0.9437, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.37509377344336087, | |
| "grad_norm": 5.849929332733154, | |
| "learning_rate": 8.804730749874182e-07, | |
| "loss": 0.5025, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5626406601650412, | |
| "grad_norm": 10.627140045166016, | |
| "learning_rate": 8.175641670860593e-07, | |
| "loss": 0.4137, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7501875468867217, | |
| "grad_norm": 6.05531644821167, | |
| "learning_rate": 7.546552591847005e-07, | |
| "loss": 0.3563, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9377344336084021, | |
| "grad_norm": 7.301521301269531, | |
| "learning_rate": 6.917463512833417e-07, | |
| "loss": 0.3179, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8625868284533544, | |
| "eval_loss": 0.3228318393230438, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 634.6946, | |
| "eval_samples_per_second": 89.594, | |
| "eval_steps_per_second": 11.201, | |
| "step": 2666 | |
| }, | |
| { | |
| "epoch": 1.1252813203300824, | |
| "grad_norm": 7.1418375968933105, | |
| "learning_rate": 6.288374433819828e-07, | |
| "loss": 0.2951, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.312828207051763, | |
| "grad_norm": 12.518142700195312, | |
| "learning_rate": 5.65928535480624e-07, | |
| "loss": 0.2749, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.5003750937734432, | |
| "grad_norm": 7.537821292877197, | |
| "learning_rate": 5.030196275792652e-07, | |
| "loss": 0.2542, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.6879219804951238, | |
| "grad_norm": 8.785564422607422, | |
| "learning_rate": 4.4011071967790636e-07, | |
| "loss": 0.2496, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.8754688672168043, | |
| "grad_norm": 8.230996131896973, | |
| "learning_rate": 3.7720181177654754e-07, | |
| "loss": 0.2347, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8783962015299394, | |
| "eval_loss": 0.28315049409866333, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 618.5578, | |
| "eval_samples_per_second": 91.932, | |
| "eval_steps_per_second": 11.493, | |
| "step": 5332 | |
| }, | |
| { | |
| "epoch": 2.0630157539384846, | |
| "grad_norm": 8.427594184875488, | |
| "learning_rate": 3.142929038751887e-07, | |
| "loss": 0.2317, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.250562640660165, | |
| "grad_norm": 7.748924732208252, | |
| "learning_rate": 2.5138399597382987e-07, | |
| "loss": 0.2347, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.4381095273818456, | |
| "grad_norm": 8.576868057250977, | |
| "learning_rate": 1.8847508807247105e-07, | |
| "loss": 0.2228, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.625656414103526, | |
| "grad_norm": 6.261688709259033, | |
| "learning_rate": 1.2556618017111224e-07, | |
| "loss": 0.2153, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.813203300825206, | |
| "grad_norm": 10.823939323425293, | |
| "learning_rate": 6.26572722697534e-08, | |
| "loss": 0.2143, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8868196605996659, | |
| "eval_loss": 0.2657858431339264, | |
| "eval_model_preparation_time": 0.0055, | |
| "eval_runtime": 609.2124, | |
| "eval_samples_per_second": 93.342, | |
| "eval_steps_per_second": 11.669, | |
| "step": 7998 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 7998, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "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": 1.982950271446756e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |