| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "global_step": 12130, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.9175597691673537e-05, | |
| "loss": 0.2206, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.8351195383347076e-05, | |
| "loss": 0.1699, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9739486703772418, | |
| "eval_loss": 0.07855913788080215, | |
| "eval_runtime": 265.3419, | |
| "eval_samples_per_second": 292.513, | |
| "eval_steps_per_second": 4.571, | |
| "step": 1213 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.752679307502061e-05, | |
| "loss": 0.118, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 1.6702390766694146e-05, | |
| "loss": 0.0991, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9866393527107813, | |
| "eval_loss": 0.04234647750854492, | |
| "eval_runtime": 267.207, | |
| "eval_samples_per_second": 290.471, | |
| "eval_steps_per_second": 4.54, | |
| "step": 2426 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 1.5877988458367685e-05, | |
| "loss": 0.0979, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 1.5053586150041222e-05, | |
| "loss": 0.0573, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 1.4229183841714757e-05, | |
| "loss": 0.0653, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9903241599670171, | |
| "eval_loss": 0.029069583863019943, | |
| "eval_runtime": 266.0985, | |
| "eval_samples_per_second": 291.681, | |
| "eval_steps_per_second": 4.558, | |
| "step": 3639 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 1.3404781533388294e-05, | |
| "loss": 0.0492, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 1.2580379225061831e-05, | |
| "loss": 0.0442, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9942151102865389, | |
| "eval_loss": 0.01643509231507778, | |
| "eval_runtime": 266.3375, | |
| "eval_samples_per_second": 291.42, | |
| "eval_steps_per_second": 4.554, | |
| "step": 4852 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 1.1755976916735368e-05, | |
| "loss": 0.0361, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "learning_rate": 1.0931574608408903e-05, | |
| "loss": 0.0296, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 1.0107172300082442e-05, | |
| "loss": 0.0315, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9958256029684601, | |
| "eval_loss": 0.010281954891979694, | |
| "eval_runtime": 266.335, | |
| "eval_samples_per_second": 291.422, | |
| "eval_steps_per_second": 4.554, | |
| "step": 6065 | |
| }, | |
| { | |
| "epoch": 5.36, | |
| "learning_rate": 9.282769991755977e-06, | |
| "loss": 0.0189, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.77, | |
| "learning_rate": 8.458367683429514e-06, | |
| "loss": 0.0207, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9973587920016491, | |
| "eval_loss": 0.0071565331891179085, | |
| "eval_runtime": 266.5437, | |
| "eval_samples_per_second": 291.194, | |
| "eval_steps_per_second": 4.551, | |
| "step": 7278 | |
| }, | |
| { | |
| "epoch": 6.18, | |
| "learning_rate": 7.633965375103051e-06, | |
| "loss": 0.0171, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 6.809563066776587e-06, | |
| "loss": 0.0133, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9979514533085961, | |
| "eval_loss": 0.004438882227987051, | |
| "eval_runtime": 266.6565, | |
| "eval_samples_per_second": 291.071, | |
| "eval_steps_per_second": 4.549, | |
| "step": 8491 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 5.9851607584501235e-06, | |
| "loss": 0.0152, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 5.1607584501236605e-06, | |
| "loss": 0.0082, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 4.3363561417971975e-06, | |
| "loss": 0.0109, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9985183467326324, | |
| "eval_loss": 0.0034066857770085335, | |
| "eval_runtime": 266.0238, | |
| "eval_samples_per_second": 291.763, | |
| "eval_steps_per_second": 4.56, | |
| "step": 9704 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "learning_rate": 3.5119538334707345e-06, | |
| "loss": 0.0077, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 2.6875515251442706e-06, | |
| "loss": 0.0059, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9989177489177489, | |
| "eval_loss": 0.0024579998571425676, | |
| "eval_runtime": 264.7367, | |
| "eval_samples_per_second": 293.182, | |
| "eval_steps_per_second": 4.582, | |
| "step": 10917 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 1.8631492168178072e-06, | |
| "loss": 0.0077, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "learning_rate": 1.038746908491344e-06, | |
| "loss": 0.0033, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 2.1434460016488048e-07, | |
| "loss": 0.0043, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9992398474541332, | |
| "eval_loss": 0.0018210469279438257, | |
| "eval_runtime": 264.6034, | |
| "eval_samples_per_second": 293.33, | |
| "eval_steps_per_second": 4.584, | |
| "step": 12130 | |
| } | |
| ], | |
| "max_steps": 12130, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.8083196174016512e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |