| { | |
| "best_metric": 0.4344363808631897, | |
| "best_model_checkpoint": "realFake-food/checkpoint-100", | |
| "epoch": 4.0, | |
| "eval_steps": 100, | |
| "global_step": 208, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19230769230769232, | |
| "grad_norm": 3.506465435028076, | |
| "learning_rate": 0.00019038461538461538, | |
| "loss": 0.8817, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 1.9431121349334717, | |
| "learning_rate": 0.00018076923076923077, | |
| "loss": 0.6353, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5769230769230769, | |
| "grad_norm": 1.186186671257019, | |
| "learning_rate": 0.00017115384615384616, | |
| "loss": 0.5035, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.654287099838257, | |
| "learning_rate": 0.00016153846153846155, | |
| "loss": 0.4855, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 3.7120602130889893, | |
| "learning_rate": 0.00015192307692307692, | |
| "loss": 0.5474, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 1.7464962005615234, | |
| "learning_rate": 0.0001423076923076923, | |
| "loss": 0.4245, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.3461538461538463, | |
| "grad_norm": 5.969783306121826, | |
| "learning_rate": 0.0001326923076923077, | |
| "loss": 0.3409, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.9254281520843506, | |
| "learning_rate": 0.0001230769230769231, | |
| "loss": 0.4396, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.7307692307692308, | |
| "grad_norm": 4.965914726257324, | |
| "learning_rate": 0.00011346153846153846, | |
| "loss": 0.4246, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 1.8691182136535645, | |
| "learning_rate": 0.00010384615384615386, | |
| "loss": 0.3941, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "eval_accuracy": 0.8013698630136986, | |
| "eval_loss": 0.4344363808631897, | |
| "eval_runtime": 6.6454, | |
| "eval_samples_per_second": 21.97, | |
| "eval_steps_per_second": 2.859, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.1153846153846154, | |
| "grad_norm": 1.4907684326171875, | |
| "learning_rate": 9.423076923076924e-05, | |
| "loss": 0.2651, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 3.2301297187805176, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.2996, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 3.9394240379333496, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.2653, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 1.300915002822876, | |
| "learning_rate": 6.538461538461539e-05, | |
| "loss": 0.2041, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.8846153846153846, | |
| "grad_norm": 5.265963077545166, | |
| "learning_rate": 5.576923076923077e-05, | |
| "loss": 0.1664, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 0.55958092212677, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.2053, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.269230769230769, | |
| "grad_norm": 0.2694534361362457, | |
| "learning_rate": 3.653846153846154e-05, | |
| "loss": 0.1057, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 3.485574960708618, | |
| "learning_rate": 2.6923076923076923e-05, | |
| "loss": 0.1739, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.6538461538461537, | |
| "grad_norm": 8.615453720092773, | |
| "learning_rate": 1.730769230769231e-05, | |
| "loss": 0.1928, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 0.40829136967658997, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.2366, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "eval_accuracy": 0.863013698630137, | |
| "eval_loss": 0.4852786064147949, | |
| "eval_runtime": 5.3876, | |
| "eval_samples_per_second": 27.099, | |
| "eval_steps_per_second": 3.527, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 208, | |
| "total_flos": 2.5603353368661197e+17, | |
| "train_loss": 0.3531340624277408, | |
| "train_runtime": 170.2734, | |
| "train_samples_per_second": 19.404, | |
| "train_steps_per_second": 1.222 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 208, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.5603353368661197e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |