| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 26.314754098360655, | |
| "global_step": 3000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 9.99957013773138e-06, | |
| "loss": 0.029, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 9.998254232827442e-06, | |
| "loss": 0.0249, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 9.996938327923503e-06, | |
| "loss": 0.0239, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 9.995622423019565e-06, | |
| "loss": 0.0223, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 9.994306518115625e-06, | |
| "loss": 0.0219, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 9.992990613211686e-06, | |
| "loss": 0.0215, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "eval_loss": 0.21139128506183624, | |
| "eval_runtime": 432.0782, | |
| "eval_samples_per_second": 9.443, | |
| "eval_steps_per_second": 1.18, | |
| "eval_wer": 0.1666585139167441, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "learning_rate": 9.991674708307746e-06, | |
| "loss": 0.0207, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 10.52, | |
| "learning_rate": 9.990358803403808e-06, | |
| "loss": 0.0209, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 11.84, | |
| "learning_rate": 9.989042898499869e-06, | |
| "loss": 0.0193, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 13.16, | |
| "learning_rate": 9.98772699359593e-06, | |
| "loss": 0.02, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.47, | |
| "learning_rate": 9.986411088691991e-06, | |
| "loss": 0.0191, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "learning_rate": 9.985095183788054e-06, | |
| "loss": 0.0189, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 17.1, | |
| "learning_rate": 9.983779278884114e-06, | |
| "loss": 0.0185, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 17.54, | |
| "eval_loss": 0.2301694005727768, | |
| "eval_runtime": 231.0348, | |
| "eval_samples_per_second": 17.66, | |
| "eval_steps_per_second": 2.207, | |
| "eval_wer": 0.16558235092696766, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 18.42, | |
| "learning_rate": 9.982463373980174e-06, | |
| "loss": 0.0184, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 19.73, | |
| "learning_rate": 9.981147469076235e-06, | |
| "loss": 0.0186, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 21.05, | |
| "learning_rate": 9.979831564172297e-06, | |
| "loss": 0.0174, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 22.37, | |
| "learning_rate": 9.978515659268357e-06, | |
| "loss": 0.0175, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 23.68, | |
| "learning_rate": 9.977199754364418e-06, | |
| "loss": 0.0169, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 9.97588384946048e-06, | |
| "loss": 0.0174, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 26.31, | |
| "learning_rate": 9.974567944556542e-06, | |
| "loss": 0.0175, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 26.31, | |
| "eval_loss": 0.23709918558597565, | |
| "eval_runtime": 224.4917, | |
| "eval_samples_per_second": 18.174, | |
| "eval_steps_per_second": 2.272, | |
| "eval_wer": 0.16504426943207945, | |
| "step": 3000 | |
| } | |
| ], | |
| "max_steps": 1140000, | |
| "num_train_epochs": 10000, | |
| "total_flos": 8.535940964167849e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |