| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9993367234136635, | |
| "eval_steps": 100, | |
| "global_step": 1130, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 4.904003870460323e-05, | |
| "loss": 0.2548, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 0.11478812247514725, | |
| "eval_runtime": 109.7793, | |
| "eval_samples_per_second": 9.164, | |
| "eval_steps_per_second": 4.582, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.6233876873505694e-05, | |
| "loss": 0.1149, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 0.0914301872253418, | |
| "eval_runtime": 109.8254, | |
| "eval_samples_per_second": 9.16, | |
| "eval_steps_per_second": 4.58, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.1797019046425264e-05, | |
| "loss": 0.0871, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_loss": 0.08308113366365433, | |
| "eval_runtime": 110.073, | |
| "eval_samples_per_second": 9.139, | |
| "eval_steps_per_second": 4.57, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 3.607020216633599e-05, | |
| "loss": 0.0865, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 0.07597807794809341, | |
| "eval_runtime": 109.949, | |
| "eval_samples_per_second": 9.15, | |
| "eval_steps_per_second": 4.575, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 2.9493228037294702e-05, | |
| "loss": 0.0802, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 0.07179657369852066, | |
| "eval_runtime": 110.0004, | |
| "eval_samples_per_second": 9.145, | |
| "eval_steps_per_second": 4.573, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 2.2571187907677853e-05, | |
| "loss": 0.0689, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 0.0701972097158432, | |
| "eval_runtime": 109.5978, | |
| "eval_samples_per_second": 9.179, | |
| "eval_steps_per_second": 4.59, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.583567302625469e-05, | |
| "loss": 0.0649, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_loss": 0.0648578405380249, | |
| "eval_runtime": 109.8826, | |
| "eval_samples_per_second": 9.155, | |
| "eval_steps_per_second": 4.578, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 9.803950080284005e-06, | |
| "loss": 0.0637, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 0.06319624185562134, | |
| "eval_runtime": 109.8291, | |
| "eval_samples_per_second": 9.16, | |
| "eval_steps_per_second": 4.58, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.939236715580884e-06, | |
| "loss": 0.0698, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.061882246285676956, | |
| "eval_runtime": 109.8313, | |
| "eval_samples_per_second": 9.16, | |
| "eval_steps_per_second": 4.58, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 1.615127855610496e-06, | |
| "loss": 0.0648, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_loss": 0.060829270631074905, | |
| "eval_runtime": 109.3936, | |
| "eval_samples_per_second": 9.196, | |
| "eval_steps_per_second": 4.598, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 8.690476815339244e-08, | |
| "loss": 0.0654, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 0.06053200736641884, | |
| "eval_runtime": 109.4208, | |
| "eval_samples_per_second": 9.194, | |
| "eval_steps_per_second": 4.597, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 1130, | |
| "total_flos": 1.4019616490441933e+17, | |
| "train_loss": 0.09174010732532603, | |
| "train_runtime": 4073.2029, | |
| "train_samples_per_second": 2.221, | |
| "train_steps_per_second": 0.277 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1130, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "total_flos": 1.4019616490441933e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |