| { | |
| "best_metric": 1.6452902555465698, | |
| "best_model_checkpoint": "fruit_100_types_image_detection/checkpoint-4220", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 4220, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 2.6183602809906006, | |
| "learning_rate": 1.7841726618705034e-06, | |
| "loss": 1.8505, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8403333333333334, | |
| "eval_loss": 1.8119271993637085, | |
| "eval_runtime": 167.1143, | |
| "eval_samples_per_second": 107.711, | |
| "eval_steps_per_second": 6.732, | |
| "step": 844 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 3.343961000442505, | |
| "learning_rate": 1.5443645083932852e-06, | |
| "loss": 1.7915, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 3.8417131900787354, | |
| "learning_rate": 1.3045563549160672e-06, | |
| "loss": 1.7335, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8433333333333334, | |
| "eval_loss": 1.7391046285629272, | |
| "eval_runtime": 169.3646, | |
| "eval_samples_per_second": 106.28, | |
| "eval_steps_per_second": 6.642, | |
| "step": 1688 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 2.7001466751098633, | |
| "learning_rate": 1.0647482014388488e-06, | |
| "loss": 1.7029, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 3.4065492153167725, | |
| "learning_rate": 8.249400479616306e-07, | |
| "loss": 1.6636, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8451666666666666, | |
| "eval_loss": 1.6863653659820557, | |
| "eval_runtime": 183.0164, | |
| "eval_samples_per_second": 98.352, | |
| "eval_steps_per_second": 6.147, | |
| "step": 2532 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "grad_norm": 2.9417073726654053, | |
| "learning_rate": 5.851318944844125e-07, | |
| "loss": 1.6379, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8458333333333333, | |
| "eval_loss": 1.655003309249878, | |
| "eval_runtime": 182.964, | |
| "eval_samples_per_second": 98.38, | |
| "eval_steps_per_second": 6.149, | |
| "step": 3376 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "grad_norm": 3.41286563873291, | |
| "learning_rate": 3.4532374100719426e-07, | |
| "loss": 1.617, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "grad_norm": 3.1814427375793457, | |
| "learning_rate": 1.0551558752997601e-07, | |
| "loss": 1.6055, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8466666666666667, | |
| "eval_loss": 1.6452902555465698, | |
| "eval_runtime": 165.9895, | |
| "eval_samples_per_second": 108.441, | |
| "eval_steps_per_second": 6.778, | |
| "step": 4220 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4220, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 1.047060731179008e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |