| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.433962264150944, | |
| "eval_steps": 500, | |
| "global_step": 3000, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5929032258064516, | |
| "eval_loss": 0.19686882197856903, | |
| "eval_runtime": 5.2661, | |
| "eval_samples_per_second": 588.67, | |
| "eval_steps_per_second": 12.343, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.685534591194969e-05, | |
| "loss": 0.3149, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8416129032258064, | |
| "eval_loss": 0.09490782022476196, | |
| "eval_runtime": 5.3608, | |
| "eval_samples_per_second": 578.275, | |
| "eval_steps_per_second": 12.125, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8938709677419355, | |
| "eval_loss": 0.06310474872589111, | |
| "eval_runtime": 5.4541, | |
| "eval_samples_per_second": 568.382, | |
| "eval_steps_per_second": 11.918, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.1093, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9032258064516129, | |
| "eval_loss": 0.04744185879826546, | |
| "eval_runtime": 5.4951, | |
| "eval_samples_per_second": 564.135, | |
| "eval_steps_per_second": 11.829, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 1.0566037735849058e-05, | |
| "loss": 0.0689, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9151612903225806, | |
| "eval_loss": 0.03924418240785599, | |
| "eval_runtime": 5.5006, | |
| "eval_samples_per_second": 563.575, | |
| "eval_steps_per_second": 11.817, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.917741935483871, | |
| "eval_loss": 0.03422646224498749, | |
| "eval_runtime": 5.3805, | |
| "eval_samples_per_second": 576.158, | |
| "eval_steps_per_second": 12.081, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 7.421383647798742e-06, | |
| "loss": 0.0539, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9190322580645162, | |
| "eval_loss": 0.03100522980093956, | |
| "eval_runtime": 5.4281, | |
| "eval_samples_per_second": 571.102, | |
| "eval_steps_per_second": 11.975, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 4.276729559748428e-06, | |
| "loss": 0.0466, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9225806451612903, | |
| "eval_loss": 0.0292543713003397, | |
| "eval_runtime": 5.4919, | |
| "eval_samples_per_second": 564.465, | |
| "eval_steps_per_second": 11.836, | |
| "step": 2544 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9219354838709677, | |
| "eval_loss": 0.028007041662931442, | |
| "eval_runtime": 5.4757, | |
| "eval_samples_per_second": 566.138, | |
| "eval_steps_per_second": 11.871, | |
| "step": 2862 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "learning_rate": 1.1320754716981133e-06, | |
| "loss": 0.043, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3180, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 779832454697856.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.015447886949790768, | |
| "num_train_epochs": 10, | |
| "temperature": 14 | |
| } | |
| } | |