| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.941176470588236, | |
| "eval_steps": 500, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5882352941176471, | |
| "grad_norm": 0.9219610978922069, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 1.156, | |
| "mean_token_accuracy": 0.754171684384346, | |
| "num_tokens": 2505918.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.1176470588235294, | |
| "grad_norm": 0.7463648293161145, | |
| "learning_rate": 2.9166666666666666e-05, | |
| "loss": 1.0146, | |
| "mean_token_accuracy": 0.7723337577448951, | |
| "num_tokens": 4791423.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.7058823529411766, | |
| "grad_norm": 0.3716128310715938, | |
| "learning_rate": 2.7083333333333335e-05, | |
| "loss": 0.857, | |
| "mean_token_accuracy": 0.8031900465488434, | |
| "num_tokens": 7330272.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.235294117647059, | |
| "grad_norm": 0.41129716737447847, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.8299, | |
| "mean_token_accuracy": 0.8095726602607303, | |
| "num_tokens": 9618952.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 0.302155542036915, | |
| "learning_rate": 2.2916666666666667e-05, | |
| "loss": 0.7125, | |
| "mean_token_accuracy": 0.8311115294694901, | |
| "num_tokens": 12114377.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.3529411764705883, | |
| "grad_norm": 0.3598975874672238, | |
| "learning_rate": 2.0833333333333333e-05, | |
| "loss": 0.7004, | |
| "mean_token_accuracy": 0.8364552325672574, | |
| "num_tokens": 14409009.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.9411764705882355, | |
| "grad_norm": 1.629633563400701, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 0.6345, | |
| "mean_token_accuracy": 0.8474959045648575, | |
| "num_tokens": 16918082.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 0.32018154650999825, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.6297, | |
| "mean_token_accuracy": 0.8520832426018186, | |
| "num_tokens": 19217454.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.31022940523188686, | |
| "learning_rate": 1.4583333333333333e-05, | |
| "loss": 0.5247, | |
| "mean_token_accuracy": 0.8709487617015839, | |
| "num_tokens": 21475135.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 5.588235294117647, | |
| "grad_norm": 0.3433976279979851, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.535, | |
| "mean_token_accuracy": 0.8695783138275146, | |
| "num_tokens": 23999790.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 0.4335383103240699, | |
| "learning_rate": 1.0416666666666666e-05, | |
| "loss": 0.4735, | |
| "mean_token_accuracy": 0.8853772415055169, | |
| "num_tokens": 26264984.0, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 6.705882352941177, | |
| "grad_norm": 0.4472925347472103, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.3983, | |
| "mean_token_accuracy": 0.8990388303995133, | |
| "num_tokens": 28792279.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.235294117647059, | |
| "grad_norm": 0.4834252565897201, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.4992, | |
| "mean_token_accuracy": 0.8846225539843241, | |
| "num_tokens": 31085796.0, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 7.823529411764706, | |
| "grad_norm": 0.4452711526767909, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.3877, | |
| "mean_token_accuracy": 0.9059422463178635, | |
| "num_tokens": 33579925.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.352941176470589, | |
| "grad_norm": 0.37901061025745547, | |
| "learning_rate": 2.0833333333333334e-06, | |
| "loss": 0.4177, | |
| "mean_token_accuracy": 0.9035960800117917, | |
| "num_tokens": 35872214.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 0.2871495120735155, | |
| "learning_rate": 0.0, | |
| "loss": 0.3761, | |
| "mean_token_accuracy": 0.9080615401268005, | |
| "num_tokens": 38392486.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "step": 80, | |
| "total_flos": 57486051115008.0, | |
| "train_loss": 0.6341701850295067, | |
| "train_runtime": 762.1388, | |
| "train_samples_per_second": 7.072, | |
| "train_steps_per_second": 0.105 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 80, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 57486051115008.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |