| { | |
| "best_metric": 2.202601909637451, | |
| "best_model_checkpoint": "./outputs/checkpoint-1900", | |
| "epoch": 1.384335154826958, | |
| "eval_steps": 100, | |
| "global_step": 1900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.7202, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.6173269748687744, | |
| "eval_runtime": 210.4944, | |
| "eval_samples_per_second": 29.806, | |
| "eval_steps_per_second": 3.729, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5807, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.566875696182251, | |
| "eval_runtime": 206.05, | |
| "eval_samples_per_second": 30.449, | |
| "eval_steps_per_second": 3.81, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5403, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.531456708908081, | |
| "eval_runtime": 205.7527, | |
| "eval_samples_per_second": 30.493, | |
| "eval_steps_per_second": 3.815, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5135, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 2.4987542629241943, | |
| "eval_runtime": 205.7584, | |
| "eval_samples_per_second": 30.492, | |
| "eval_steps_per_second": 3.815, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4703, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.4730148315429688, | |
| "eval_runtime": 205.8153, | |
| "eval_samples_per_second": 30.484, | |
| "eval_steps_per_second": 3.814, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4449, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 2.4445066452026367, | |
| "eval_runtime": 205.9496, | |
| "eval_samples_per_second": 30.464, | |
| "eval_steps_per_second": 3.812, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4265, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.421610116958618, | |
| "eval_runtime": 206.1548, | |
| "eval_samples_per_second": 30.433, | |
| "eval_steps_per_second": 3.808, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4059, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 2.401268720626831, | |
| "eval_runtime": 206.1413, | |
| "eval_samples_per_second": 30.435, | |
| "eval_steps_per_second": 3.808, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3694, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 2.3795337677001953, | |
| "eval_runtime": 206.144, | |
| "eval_samples_per_second": 30.435, | |
| "eval_steps_per_second": 3.808, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3664, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 2.3618950843811035, | |
| "eval_runtime": 205.9336, | |
| "eval_samples_per_second": 30.466, | |
| "eval_steps_per_second": 3.812, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3599, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 2.338806390762329, | |
| "eval_runtime": 205.8736, | |
| "eval_samples_per_second": 30.475, | |
| "eval_steps_per_second": 3.813, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3162, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 2.3209476470947266, | |
| "eval_runtime": 206.0155, | |
| "eval_samples_per_second": 30.454, | |
| "eval_steps_per_second": 3.81, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3272, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 2.3050100803375244, | |
| "eval_runtime": 205.8768, | |
| "eval_samples_per_second": 30.475, | |
| "eval_steps_per_second": 3.813, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2787, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 2.2802278995513916, | |
| "eval_runtime": 206.0479, | |
| "eval_samples_per_second": 30.449, | |
| "eval_steps_per_second": 3.81, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2387, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 2.263134479522705, | |
| "eval_runtime": 205.9604, | |
| "eval_samples_per_second": 30.462, | |
| "eval_steps_per_second": 3.811, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2303, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 2.246272325515747, | |
| "eval_runtime": 206.1382, | |
| "eval_samples_per_second": 30.436, | |
| "eval_steps_per_second": 3.808, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.0002, | |
| "loss": 2.2323, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_loss": 2.231445789337158, | |
| "eval_runtime": 205.9926, | |
| "eval_samples_per_second": 30.457, | |
| "eval_steps_per_second": 3.811, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0002, | |
| "loss": 2.193, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 2.2156155109405518, | |
| "eval_runtime": 217.8526, | |
| "eval_samples_per_second": 28.799, | |
| "eval_steps_per_second": 3.603, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0002, | |
| "loss": 2.1776, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 2.202601909637451, | |
| "eval_runtime": 529.3371, | |
| "eval_samples_per_second": 11.853, | |
| "eval_steps_per_second": 1.483, | |
| "step": 1900 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 5.588270155302912e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |