| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.99936, | |
| "global_step": 1562, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5028, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.863201094391245e-05, | |
| "loss": 0.2106, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.72640218878249e-05, | |
| "loss": 0.1829, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 1.5896032831737347e-05, | |
| "loss": 0.165, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.4528043775649796e-05, | |
| "loss": 0.1482, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 1.3160054719562245e-05, | |
| "loss": 0.1394, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.1792065663474691e-05, | |
| "loss": 0.1451, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.96384, | |
| "eval_loss": 0.10218027979135513, | |
| "eval_runtime": 754.598, | |
| "eval_samples_per_second": 33.13, | |
| "eval_steps_per_second": 4.141, | |
| "step": 781 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.0424076607387142e-05, | |
| "loss": 0.1213, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 9.05608755129959e-06, | |
| "loss": 0.091, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 7.688098495212039e-06, | |
| "loss": 0.0988, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 6.3201094391244875e-06, | |
| "loss": 0.0822, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 4.952120383036936e-06, | |
| "loss": 0.092, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 3.584131326949385e-06, | |
| "loss": 0.0894, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.2161422708618332e-06, | |
| "loss": 0.1006, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 8.481532147742819e-07, | |
| "loss": 0.0767, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.96628, | |
| "eval_loss": 0.11105333268642426, | |
| "eval_runtime": 750.9747, | |
| "eval_samples_per_second": 33.29, | |
| "eval_steps_per_second": 4.161, | |
| "step": 1562 | |
| } | |
| ], | |
| "max_steps": 1562, | |
| "num_train_epochs": 2, | |
| "total_flos": 4.388029403227747e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |