| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.0, | |
| "eval_steps": 500, | |
| "global_step": 2862, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5861290322580646, | |
| "eval_loss": 0.19721022248268127, | |
| "eval_runtime": 6.85, | |
| "eval_samples_per_second": 452.554, | |
| "eval_steps_per_second": 9.489, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.5261461734771729, | |
| "learning_rate": 1.650593990216632e-05, | |
| "loss": 0.3175, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8193548387096774, | |
| "eval_loss": 0.09563933312892914, | |
| "eval_runtime": 5.9266, | |
| "eval_samples_per_second": 523.068, | |
| "eval_steps_per_second": 10.968, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8774193548387097, | |
| "eval_loss": 0.06404151022434235, | |
| "eval_runtime": 5.6772, | |
| "eval_samples_per_second": 546.04, | |
| "eval_steps_per_second": 11.449, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.44202694296836853, | |
| "learning_rate": 1.3011879804332637e-05, | |
| "loss": 0.1109, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8996774193548387, | |
| "eval_loss": 0.04958628490567207, | |
| "eval_runtime": 5.7185, | |
| "eval_samples_per_second": 542.104, | |
| "eval_steps_per_second": 11.367, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.3023432493209839, | |
| "learning_rate": 9.517819706498952e-06, | |
| "loss": 0.071, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9077419354838709, | |
| "eval_loss": 0.04074610769748688, | |
| "eval_runtime": 5.8617, | |
| "eval_samples_per_second": 528.859, | |
| "eval_steps_per_second": 11.089, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9103225806451613, | |
| "eval_loss": 0.035789892077445984, | |
| "eval_runtime": 5.6338, | |
| "eval_samples_per_second": 550.247, | |
| "eval_steps_per_second": 11.537, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.2652527689933777, | |
| "learning_rate": 6.02375960866527e-06, | |
| "loss": 0.0559, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9190322580645162, | |
| "eval_loss": 0.03312822803854942, | |
| "eval_runtime": 5.9997, | |
| "eval_samples_per_second": 516.69, | |
| "eval_steps_per_second": 10.834, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.861635220125786, | |
| "grad_norm": 0.27172261476516724, | |
| "learning_rate": 2.5296995108315863e-06, | |
| "loss": 0.0491, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9238709677419354, | |
| "eval_loss": 0.03127439320087433, | |
| "eval_runtime": 5.8082, | |
| "eval_samples_per_second": 533.732, | |
| "eval_steps_per_second": 11.191, | |
| "step": 2544 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2862, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 651155886807636.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.7760026697362031, | |
| "num_train_epochs": 9, | |
| "temperature": 12 | |
| } | |
| } | |