| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.994482251241493, | |
| "eval_steps": 500, | |
| "global_step": 4074, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1.755522827687776e-05, | |
| "loss": 2.1963, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.6409082412719727, | |
| "eval_runtime": 92.0193, | |
| "eval_samples_per_second": 129.201, | |
| "eval_steps_per_second": 8.085, | |
| "step": 679 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 1.51006381934217e-05, | |
| "loss": 1.3139, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.5666508674621582, | |
| "eval_runtime": 91.885, | |
| "eval_samples_per_second": 129.39, | |
| "eval_steps_per_second": 8.097, | |
| "step": 1359 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 1.265095729013255e-05, | |
| "loss": 1.1653, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 1.0196367206676486e-05, | |
| "loss": 1.0453, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.5353167057037354, | |
| "eval_runtime": 91.6769, | |
| "eval_samples_per_second": 129.684, | |
| "eval_steps_per_second": 8.115, | |
| "step": 2038 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 7.741777123220424e-06, | |
| "loss": 0.9454, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.5275497436523438, | |
| "eval_runtime": 91.4939, | |
| "eval_samples_per_second": 129.943, | |
| "eval_steps_per_second": 8.132, | |
| "step": 2718 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 5.28718703976436e-06, | |
| "loss": 0.8888, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 1.5392436981201172, | |
| "eval_runtime": 91.9557, | |
| "eval_samples_per_second": 129.291, | |
| "eval_steps_per_second": 8.091, | |
| "step": 3398 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 2.832596956308297e-06, | |
| "loss": 0.8473, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 3.829160530191459e-07, | |
| "loss": 0.8072, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_loss": 1.5723119974136353, | |
| "eval_runtime": 91.9397, | |
| "eval_samples_per_second": 129.313, | |
| "eval_steps_per_second": 8.092, | |
| "step": 4074 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4074, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 1.021853879809367e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |