{ "best_global_step": null, "best_metric": null, "best_model_checkpoint": null, "epoch": 0.625, "eval_steps": 500, "global_step": 10, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "entropy": 0.7410857528448105, "epoch": 0.0625, "grad_norm": 7.1875, "learning_rate": 0.0, "loss": 1.2128111124038696, "mean_token_accuracy": 0.7650293260812759, "num_tokens": 738.0, "step": 1 }, { "entropy": 0.9062299728393555, "epoch": 0.125, "grad_norm": 8.25, "learning_rate": 0.0002, "loss": 1.3959347009658813, "mean_token_accuracy": 0.7537878900766373, "num_tokens": 1421.0, "step": 2 }, { "entropy": 0.6709113866090775, "epoch": 0.1875, "grad_norm": 3.796875, "learning_rate": 0.00019396926207859084, "loss": 0.5683731436729431, "mean_token_accuracy": 0.8942521065473557, "num_tokens": 2154.0, "step": 3 }, { "entropy": 0.5079795271158218, "epoch": 0.25, "grad_norm": 1.6328125, "learning_rate": 0.0001766044443118978, "loss": 0.41578274965286255, "mean_token_accuracy": 0.9333577752113342, "num_tokens": 2850.0, "step": 4 }, { "entropy": 0.318486500531435, "epoch": 0.3125, "grad_norm": 0.86328125, "learning_rate": 0.00015000000000000001, "loss": 0.25355467200279236, "mean_token_accuracy": 0.9582149684429169, "num_tokens": 3519.0, "step": 5 }, { "entropy": 0.2318822667002678, "epoch": 0.375, "grad_norm": 0.5390625, "learning_rate": 0.00011736481776669306, "loss": 0.24391978979110718, "mean_token_accuracy": 0.968939408659935, "num_tokens": 4197.0, "step": 6 }, { "entropy": 0.22734014317393303, "epoch": 0.4375, "grad_norm": 1.09375, "learning_rate": 8.263518223330697e-05, "loss": 0.3192989230155945, "mean_token_accuracy": 0.960879310965538, "num_tokens": 4881.0, "step": 7 }, { "entropy": 0.2031594216823578, "epoch": 0.5, "grad_norm": 0.9375, "learning_rate": 5.000000000000002e-05, "loss": 0.22030511498451233, "mean_token_accuracy": 0.9645396322011948, "num_tokens": 5613.0, "step": 8 }, { "entropy": 0.23354319110512733, "epoch": 0.5625, "grad_norm": 1.390625, "learning_rate": 2.339555568810221e-05, "loss": 0.24805954098701477, "mean_token_accuracy": 0.9575671851634979, "num_tokens": 6346.0, "step": 9 }, { "entropy": 0.21387554705142975, "epoch": 0.625, "grad_norm": 1.1171875, "learning_rate": 6.030737921409169e-06, "loss": 0.21693643927574158, "mean_token_accuracy": 0.966460719704628, "num_tokens": 7066.0, "step": 10 }, { "epoch": 0.625, "step": 10, "total_flos": 319957003517952.0, "train_loss": 0.5094976186752319, "train_runtime": 14.8579, "train_samples_per_second": 2.692, "train_steps_per_second": 0.673 } ], "logging_steps": 1, "max_steps": 10, "num_input_tokens_seen": 0, "num_train_epochs": 1, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": false, "should_training_stop": false }, "attributes": {} } }, "total_flos": 319957003517952.0, "train_batch_size": 1, "trial_name": null, "trial_params": null }