| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.1027095148078134, | |
| "global_step": 140000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.9921235034656587e-06, | |
| "loss": 0.2621, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.984247006931317e-06, | |
| "loss": 0.2548, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.9763705103969753e-06, | |
| "loss": 0.2451, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.9684940138626337e-06, | |
| "loss": 0.2479, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.9606175173282924e-06, | |
| "loss": 0.2477, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.9227449297904968, | |
| "eval_loss": 0.42459121346473694, | |
| "eval_runtime": 532.8547, | |
| "eval_samples_per_second": 53.054, | |
| "eval_steps_per_second": 13.264, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.9527410207939508e-06, | |
| "loss": 0.253, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.9448645242596095e-06, | |
| "loss": 0.2466, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.936988027725268e-06, | |
| "loss": 0.25, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 1.929111531190926e-06, | |
| "loss": 0.2402, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.9212350346565845e-06, | |
| "loss": 0.2515, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_accuracy": 0.926034688949585, | |
| "eval_loss": 0.3926495909690857, | |
| "eval_runtime": 506.3742, | |
| "eval_samples_per_second": 55.828, | |
| "eval_steps_per_second": 13.958, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1.9133585381222433e-06, | |
| "loss": 0.2383, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.9054820415879016e-06, | |
| "loss": 0.2523, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 1.8976055450535602e-06, | |
| "loss": 0.2372, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.8897290485192185e-06, | |
| "loss": 0.2395, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 1.881852551984877e-06, | |
| "loss": 0.2376, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_accuracy": 0.9263883829116821, | |
| "eval_loss": 0.3989144265651703, | |
| "eval_runtime": 505.8843, | |
| "eval_samples_per_second": 55.882, | |
| "eval_steps_per_second": 13.972, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.8739760554505356e-06, | |
| "loss": 0.2331, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.8660995589161941e-06, | |
| "loss": 0.2426, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.8582230623818525e-06, | |
| "loss": 0.2493, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.850346565847511e-06, | |
| "loss": 0.2379, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 1.8424700693131694e-06, | |
| "loss": 0.2428, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_accuracy": 0.9267421364784241, | |
| "eval_loss": 0.3985295295715332, | |
| "eval_runtime": 549.265, | |
| "eval_samples_per_second": 51.469, | |
| "eval_steps_per_second": 12.868, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1.834593572778828e-06, | |
| "loss": 0.2429, | |
| "step": 105000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1.8267170762444864e-06, | |
| "loss": 0.238, | |
| "step": 110000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 1.818840579710145e-06, | |
| "loss": 0.2322, | |
| "step": 115000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.8109640831758033e-06, | |
| "loss": 0.2371, | |
| "step": 120000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.8030875866414619e-06, | |
| "loss": 0.2303, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_accuracy": 0.9282631874084473, | |
| "eval_loss": 0.40024659037590027, | |
| "eval_runtime": 548.3799, | |
| "eval_samples_per_second": 51.552, | |
| "eval_steps_per_second": 12.889, | |
| "step": 125000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.7952110901071202e-06, | |
| "loss": 0.221, | |
| "step": 130000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 1.7873345935727788e-06, | |
| "loss": 0.2199, | |
| "step": 135000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.779458097038437e-06, | |
| "loss": 0.2097, | |
| "step": 140000 | |
| } | |
| ], | |
| "max_steps": 1269600, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.473443106221998e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |