| { | |
| "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": 0.6738709677419354, | |
| "eval_loss": 0.40543094277381897, | |
| "eval_runtime": 1.5192, | |
| "eval_samples_per_second": 2040.5, | |
| "eval_steps_per_second": 42.785, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.685534591194969e-05, | |
| "loss": 0.6402, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8451612903225807, | |
| "eval_loss": 0.1374775767326355, | |
| "eval_runtime": 1.4813, | |
| "eval_samples_per_second": 2092.755, | |
| "eval_steps_per_second": 43.88, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8964516129032258, | |
| "eval_loss": 0.06959949433803558, | |
| "eval_runtime": 1.4706, | |
| "eval_samples_per_second": 2107.994, | |
| "eval_steps_per_second": 44.2, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.371069182389937e-05, | |
| "loss": 0.1606, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9161290322580645, | |
| "eval_loss": 0.05019702389836311, | |
| "eval_runtime": 1.469, | |
| "eval_samples_per_second": 2110.305, | |
| "eval_steps_per_second": 44.248, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 1.0566037735849058e-05, | |
| "loss": 0.0795, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9251612903225807, | |
| "eval_loss": 0.0416053868830204, | |
| "eval_runtime": 1.4758, | |
| "eval_samples_per_second": 2100.597, | |
| "eval_steps_per_second": 44.045, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9287096774193548, | |
| "eval_loss": 0.03693937882781029, | |
| "eval_runtime": 1.4766, | |
| "eval_samples_per_second": 2099.377, | |
| "eval_steps_per_second": 44.019, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 7.421383647798742e-06, | |
| "loss": 0.06, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9306451612903226, | |
| "eval_loss": 0.034136172384023666, | |
| "eval_runtime": 1.4764, | |
| "eval_samples_per_second": 2099.651, | |
| "eval_steps_per_second": 44.025, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 4.276729559748428e-06, | |
| "loss": 0.0515, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9335483870967742, | |
| "eval_loss": 0.03264344111084938, | |
| "eval_runtime": 1.4776, | |
| "eval_samples_per_second": 2098.02, | |
| "eval_steps_per_second": 43.991, | |
| "step": 2544 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9361290322580645, | |
| "eval_loss": 0.031439587473869324, | |
| "eval_runtime": 1.4731, | |
| "eval_samples_per_second": 2104.45, | |
| "eval_steps_per_second": 44.126, | |
| "step": 2862 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "learning_rate": 1.1320754716981133e-06, | |
| "loss": 0.0478, | |
| "step": 3000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 3180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 922119378435360.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.6113798620723033, | |
| "num_train_epochs": 10, | |
| "temperature": 2 | |
| } | |
| } | |