| { | |
| "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": 0.5880645161290322, | |
| "eval_loss": 0.20545053482055664, | |
| "eval_runtime": 5.7316, | |
| "eval_samples_per_second": 540.858, | |
| "eval_steps_per_second": 11.341, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.5723270440251573, | |
| "grad_norm": 0.5401266813278198, | |
| "learning_rate": 1.606918238993711e-05, | |
| "loss": 0.329, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8193548387096774, | |
| "eval_loss": 0.09952054172754288, | |
| "eval_runtime": 6.0215, | |
| "eval_samples_per_second": 514.821, | |
| "eval_steps_per_second": 10.795, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8770967741935484, | |
| "eval_loss": 0.06660476326942444, | |
| "eval_runtime": 5.7607, | |
| "eval_samples_per_second": 538.131, | |
| "eval_steps_per_second": 11.283, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.1446540880503147, | |
| "grad_norm": 0.46109867095947266, | |
| "learning_rate": 1.2138364779874214e-05, | |
| "loss": 0.1154, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8974193548387097, | |
| "eval_loss": 0.05176570266485214, | |
| "eval_runtime": 5.7827, | |
| "eval_samples_per_second": 536.081, | |
| "eval_steps_per_second": 11.24, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.716981132075472, | |
| "grad_norm": 0.3152710497379303, | |
| "learning_rate": 8.207547169811321e-06, | |
| "loss": 0.0744, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9051612903225806, | |
| "eval_loss": 0.04294683784246445, | |
| "eval_runtime": 5.9586, | |
| "eval_samples_per_second": 520.253, | |
| "eval_steps_per_second": 10.909, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9087096774193548, | |
| "eval_loss": 0.03813477233052254, | |
| "eval_runtime": 5.6249, | |
| "eval_samples_per_second": 551.12, | |
| "eval_steps_per_second": 11.556, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.289308176100629, | |
| "grad_norm": 0.27830758690834045, | |
| "learning_rate": 4.276729559748428e-06, | |
| "loss": 0.0593, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9158064516129032, | |
| "eval_loss": 0.03578370437026024, | |
| "eval_runtime": 5.8612, | |
| "eval_samples_per_second": 528.901, | |
| "eval_steps_per_second": 11.09, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.861635220125786, | |
| "grad_norm": 0.28137683868408203, | |
| "learning_rate": 3.459119496855346e-07, | |
| "loss": 0.0533, | |
| "step": 2500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2544, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 651155886807636.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.10338086591389373, | |
| "num_train_epochs": 8, | |
| "temperature": 9 | |
| } | |
| } | |