| { | |
| "best_metric": 0.4889937106918239, | |
| "best_model_checkpoint": "resnet-50-finetuned-student_kaggle/checkpoint-11", | |
| "epoch": 4.680851063829787, | |
| "eval_steps": 500, | |
| "global_step": 55, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.851063829787234, | |
| "grad_norm": 492.8765869140625, | |
| "learning_rate": 4.591836734693878e-05, | |
| "loss": 3.371893679888266e+22, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.9361702127659575, | |
| "eval_accuracy": 0.4889937106918239, | |
| "eval_loss": 3.489738920977707e+22, | |
| "eval_runtime": 7.7509, | |
| "eval_samples_per_second": 82.055, | |
| "eval_steps_per_second": 2.58, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.702127659574468, | |
| "grad_norm": 954.1488647460938, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 3.243846974994898e+22, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.9574468085106385, | |
| "eval_accuracy": 0.4889937106918239, | |
| "eval_loss": 3.489738920977707e+22, | |
| "eval_runtime": 8.4777, | |
| "eval_samples_per_second": 75.021, | |
| "eval_steps_per_second": 2.359, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 2.5531914893617023, | |
| "grad_norm": 704.236083984375, | |
| "learning_rate": 2.5510204081632654e-05, | |
| "loss": 3.336324610319264e+22, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.978723404255319, | |
| "eval_accuracy": 0.4889937106918239, | |
| "eval_loss": 3.489738920977707e+22, | |
| "eval_runtime": 8.3616, | |
| "eval_samples_per_second": 76.062, | |
| "eval_steps_per_second": 2.392, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 3.404255319148936, | |
| "grad_norm": 400.68768310546875, | |
| "learning_rate": 1.5306122448979594e-05, | |
| "loss": 3.295420756775664e+22, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.4889937106918239, | |
| "eval_loss": 3.489738920977707e+22, | |
| "eval_runtime": 8.378, | |
| "eval_samples_per_second": 75.913, | |
| "eval_steps_per_second": 2.387, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 4.25531914893617, | |
| "grad_norm": 1062.6368408203125, | |
| "learning_rate": 5.102040816326531e-06, | |
| "loss": 3.2794156842759295e+22, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.680851063829787, | |
| "eval_accuracy": 0.4889937106918239, | |
| "eval_loss": 3.489738920977707e+22, | |
| "eval_runtime": 8.4114, | |
| "eval_samples_per_second": 75.612, | |
| "eval_steps_per_second": 2.378, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 4.680851063829787, | |
| "step": 55, | |
| "total_flos": 1.477984078577664e+17, | |
| "train_loss": 3.320157440986396e+22, | |
| "train_runtime": 166.0473, | |
| "train_samples_per_second": 44.686, | |
| "train_steps_per_second": 0.331 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 55, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 1.477984078577664e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |