| { | |
| "best_metric": 0.011737266555428505, | |
| "best_model_checkpoint": "./models/day-night/checkpoint-800", | |
| "epoch": 5.0, | |
| "global_step": 1680, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 9.404761904761905e-05, | |
| "loss": 0.2582, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.80952380952381e-05, | |
| "loss": 0.0899, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.9711316397228638, | |
| "eval_loss": 0.09337754547595978, | |
| "eval_runtime": 7.0345, | |
| "eval_samples_per_second": 123.107, | |
| "eval_steps_per_second": 15.495, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 8.220238095238095e-05, | |
| "loss": 0.1391, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 7.625e-05, | |
| "loss": 0.026, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_accuracy": 0.9942263279445728, | |
| "eval_loss": 0.02246282249689102, | |
| "eval_runtime": 7.0033, | |
| "eval_samples_per_second": 123.656, | |
| "eval_steps_per_second": 15.564, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 7.029761904761904e-05, | |
| "loss": 0.0434, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 6.43452380952381e-05, | |
| "loss": 0.0689, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "eval_accuracy": 0.7032332563510393, | |
| "eval_loss": 1.523564338684082, | |
| "eval_runtime": 6.9972, | |
| "eval_samples_per_second": 123.764, | |
| "eval_steps_per_second": 15.578, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 5.839285714285715e-05, | |
| "loss": 0.0315, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 5.24404761904762e-05, | |
| "loss": 0.0193, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "eval_accuracy": 0.9965357967667436, | |
| "eval_loss": 0.011737266555428505, | |
| "eval_runtime": 7.6431, | |
| "eval_samples_per_second": 113.304, | |
| "eval_steps_per_second": 14.261, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 4.648809523809524e-05, | |
| "loss": 0.0621, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 4.05952380952381e-05, | |
| "loss": 0.028, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.9919168591224018, | |
| "eval_loss": 0.01863272860646248, | |
| "eval_runtime": 7.0922, | |
| "eval_samples_per_second": 122.106, | |
| "eval_steps_per_second": 15.369, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 3.4642857142857144e-05, | |
| "loss": 0.0121, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "learning_rate": 2.869047619047619e-05, | |
| "loss": 0.0159, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "eval_accuracy": 0.9953810623556582, | |
| "eval_loss": 0.01500316709280014, | |
| "eval_runtime": 7.1901, | |
| "eval_samples_per_second": 120.444, | |
| "eval_steps_per_second": 15.16, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 2.273809523809524e-05, | |
| "loss": 0.0226, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 1.6785714285714285e-05, | |
| "loss": 0.0194, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "eval_accuracy": 0.9919168591224018, | |
| "eval_loss": 0.03691905736923218, | |
| "eval_runtime": 7.2131, | |
| "eval_samples_per_second": 120.059, | |
| "eval_steps_per_second": 15.111, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 1.0833333333333334e-05, | |
| "loss": 0.013, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 4.880952380952381e-06, | |
| "loss": 0.0081, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "eval_accuracy": 0.9849884526558892, | |
| "eval_loss": 0.04706377908587456, | |
| "eval_runtime": 7.1715, | |
| "eval_samples_per_second": 120.756, | |
| "eval_steps_per_second": 15.199, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 1680, | |
| "total_flos": 1.0500251940864e+16, | |
| "train_loss": 0.051665182056881136, | |
| "train_runtime": 245.9608, | |
| "train_samples_per_second": 68.222, | |
| "train_steps_per_second": 6.83 | |
| } | |
| ], | |
| "max_steps": 1680, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.0500251940864e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |