| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.4, | |
| "global_step": 20000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.1904e-05, | |
| "loss": 1.4759, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 2.3880000000000002e-05, | |
| "loss": 2.0784, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.9992187757207995e-05, | |
| "loss": 1.8615, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 2.9927417972317198e-05, | |
| "loss": 0.9561, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 2.97973743867291e-05, | |
| "loss": 0.8567, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 2.9602625647955514e-05, | |
| "loss": 0.7767, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 2.9344023342679696e-05, | |
| "loss": 0.7502, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 2.9022698272984482e-05, | |
| "loss": 0.8425, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 2.8640055511642487e-05, | |
| "loss": 0.7989, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 2.8197768258090375e-05, | |
| "loss": 0.8392, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.8492928069198428, | |
| "eval_loss": 0.6663709282875061, | |
| "eval_runtime": 121.7475, | |
| "eval_samples_per_second": 41.069, | |
| "eval_steps_per_second": 2.571, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.7697770521953444e-05, | |
| "loss": 0.8444, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.714224866611362e-05, | |
| "loss": 0.7575, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 2.6533631846300875e-05, | |
| "loss": 0.7919, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 2.587458138901327e-05, | |
| "loss": 0.833, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2.516797915421335e-05, | |
| "loss": 0.8335, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 2.4416914933687753e-05, | |
| "loss": 0.8382, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 2.3624672940173894e-05, | |
| "loss": 0.8253, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 2.27947174463333e-05, | |
| "loss": 0.8442, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 2.193067763636869e-05, | |
| "loss": 0.8794, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.1036331736524874e-05, | |
| "loss": 0.8631, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.855058580917128, | |
| "eval_loss": 0.6343714594841003, | |
| "eval_runtime": 151.1949, | |
| "eval_samples_per_second": 33.07, | |
| "eval_steps_per_second": 2.07, | |
| "step": 20000 | |
| } | |
| ], | |
| "max_steps": 50000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.34684268822528e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |