| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 1074, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.9068901303538176e-05, | |
| "loss": 0.8387, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.813780260707635e-05, | |
| "loss": 0.4021, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 1.7206703910614527e-05, | |
| "loss": 0.2884, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 1.62756052141527e-05, | |
| "loss": 0.2756, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.5344506517690876e-05, | |
| "loss": 0.2636, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 1.4413407821229052e-05, | |
| "loss": 0.2349, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 1.3482309124767226e-05, | |
| "loss": 0.2557, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 1.25512104283054e-05, | |
| "loss": 0.2014, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.1620111731843577e-05, | |
| "loss": 0.1959, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 1.0689013035381753e-05, | |
| "loss": 0.1964, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_accuracy": 0.9355653821857233, | |
| "eval_loss": 0.2315603792667389, | |
| "eval_runtime": 3.5804, | |
| "eval_samples_per_second": 442.134, | |
| "eval_steps_per_second": 27.651, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 9.757914338919926e-06, | |
| "loss": 0.2142, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.826815642458101e-06, | |
| "loss": 0.1913, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 7.895716945996277e-06, | |
| "loss": 0.1534, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 6.964618249534451e-06, | |
| "loss": 0.1755, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 6.033519553072626e-06, | |
| "loss": 0.1508, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 5.102420856610801e-06, | |
| "loss": 0.1244, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 4.171322160148976e-06, | |
| "loss": 0.1785, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 3.240223463687151e-06, | |
| "loss": 0.1347, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 2.3091247672253262e-06, | |
| "loss": 0.132, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 1.378026070763501e-06, | |
| "loss": 0.1334, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "eval_accuracy": 0.9519898926089703, | |
| "eval_loss": 0.21129217743873596, | |
| "eval_runtime": 3.5946, | |
| "eval_samples_per_second": 440.378, | |
| "eval_steps_per_second": 27.541, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 4.46927374301676e-07, | |
| "loss": 0.1359, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 1074, | |
| "total_flos": 3337406743067076.0, | |
| "train_loss": 0.23017916310876885, | |
| "train_runtime": 431.913, | |
| "train_samples_per_second": 79.363, | |
| "train_steps_per_second": 2.487 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 1074, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 3337406743067076.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |