| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.861635220125786, | |
| "eval_steps": 500, | |
| "global_step": 2500, | |
| "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": 5.3992, | |
| "eval_samples_per_second": 574.162, | |
| "eval_steps_per_second": 12.039, | |
| "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": 5.4571, | |
| "eval_samples_per_second": 568.065, | |
| "eval_steps_per_second": 11.911, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8938709677419355, | |
| "eval_loss": 0.06310474872589111, | |
| "eval_runtime": 5.4013, | |
| "eval_samples_per_second": 573.935, | |
| "eval_steps_per_second": 12.034, | |
| "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": 5.4117, | |
| "eval_samples_per_second": 572.834, | |
| "eval_steps_per_second": 12.011, | |
| "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": 5.5557, | |
| "eval_samples_per_second": 557.989, | |
| "eval_steps_per_second": 11.7, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.917741935483871, | |
| "eval_loss": 0.03422646224498749, | |
| "eval_runtime": 5.3885, | |
| "eval_samples_per_second": 575.3, | |
| "eval_steps_per_second": 12.063, | |
| "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": 5.4997, | |
| "eval_samples_per_second": 563.666, | |
| "eval_steps_per_second": 11.819, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 4.276729559748428e-06, | |
| "loss": 0.0466, | |
| "step": 2500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3180, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 790936363356552.0, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.1451722512397559, | |
| "num_train_epochs": 10, | |
| "temperature": 14 | |
| } | |
| } | |