| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.187604690117253, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.66499162479062e-05, | |
| "loss": 0.0795, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_accuracy": 0.9863571086644327, | |
| "eval_loss": 0.04342207312583923, | |
| "eval_runtime": 78.4916, | |
| "eval_samples_per_second": 106.457, | |
| "eval_steps_per_second": 2.535, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 4.32998324958124e-05, | |
| "loss": 0.0434, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_accuracy": 0.9869554810914313, | |
| "eval_loss": 0.035848040133714676, | |
| "eval_runtime": 78.5171, | |
| "eval_samples_per_second": 106.423, | |
| "eval_steps_per_second": 2.534, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 3.9949748743718597e-05, | |
| "loss": 0.0375, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_accuracy": 0.9876735280038296, | |
| "eval_loss": 0.03416126221418381, | |
| "eval_runtime": 78.4728, | |
| "eval_samples_per_second": 106.483, | |
| "eval_steps_per_second": 2.536, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 3.6599664991624795e-05, | |
| "loss": 0.0112, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "eval_accuracy": 0.9885112494016276, | |
| "eval_loss": 0.03930915519595146, | |
| "eval_runtime": 78.5108, | |
| "eval_samples_per_second": 106.431, | |
| "eval_steps_per_second": 2.535, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 3.324958123953099e-05, | |
| "loss": 0.0089, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "eval_accuracy": 0.9877932024892293, | |
| "eval_loss": 0.04387575387954712, | |
| "eval_runtime": 78.5288, | |
| "eval_samples_per_second": 106.407, | |
| "eval_steps_per_second": 2.534, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 2.989949748743719e-05, | |
| "loss": 0.0117, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_accuracy": 0.989588319770225, | |
| "eval_loss": 0.04102291911840439, | |
| "eval_runtime": 78.5424, | |
| "eval_samples_per_second": 106.388, | |
| "eval_steps_per_second": 2.534, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 2.6549413735343385e-05, | |
| "loss": 0.0028, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_accuracy": 0.989588319770225, | |
| "eval_loss": 0.055005114525556564, | |
| "eval_runtime": 78.492, | |
| "eval_samples_per_second": 106.457, | |
| "eval_steps_per_second": 2.535, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 2.3199329983249583e-05, | |
| "loss": 0.0014, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "eval_accuracy": 0.9889899473432264, | |
| "eval_loss": 0.054427579045295715, | |
| "eval_runtime": 78.5005, | |
| "eval_samples_per_second": 106.445, | |
| "eval_steps_per_second": 2.535, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 1.984924623115578e-05, | |
| "loss": 0.0011, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_accuracy": 0.989588319770225, | |
| "eval_loss": 0.05314817279577255, | |
| "eval_runtime": 78.5501, | |
| "eval_samples_per_second": 106.378, | |
| "eval_steps_per_second": 2.533, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 1.6499162479061976e-05, | |
| "loss": 0.0002, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "eval_accuracy": 0.9894686452848253, | |
| "eval_loss": 0.058279525488615036, | |
| "eval_runtime": 78.4971, | |
| "eval_samples_per_second": 106.45, | |
| "eval_steps_per_second": 2.535, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 1.3149078726968176e-05, | |
| "loss": 0.0002, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "eval_accuracy": 0.9897079942556247, | |
| "eval_loss": 0.05424511060118675, | |
| "eval_runtime": 78.4985, | |
| "eval_samples_per_second": 106.448, | |
| "eval_steps_per_second": 2.535, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 9.798994974874372e-06, | |
| "loss": 0.0002, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "eval_accuracy": 0.989588319770225, | |
| "eval_loss": 0.05800911784172058, | |
| "eval_runtime": 78.5022, | |
| "eval_samples_per_second": 106.443, | |
| "eval_steps_per_second": 2.535, | |
| "step": 2400 | |
| } | |
| ], | |
| "max_steps": 2985, | |
| "num_train_epochs": 5, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |