{ "best_metric": null, "best_model_checkpoint": null, "epoch": 9.142857142857142, "eval_steps": 500, "global_step": 80, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.1428571428571428, "grad_norm": 0.9526768326759338, "learning_rate": 0.0001923879532511287, "loss": 1.1925, "step": 10 }, { "epoch": 2.2857142857142856, "grad_norm": 0.0799003466963768, "learning_rate": 0.00017071067811865476, "loss": 0.86, "step": 20 }, { "epoch": 3.4285714285714284, "grad_norm": 0.04628448933362961, "learning_rate": 0.000138268343236509, "loss": 0.7218, "step": 30 }, { "epoch": 4.571428571428571, "grad_norm": 0.04921531677246094, "learning_rate": 0.0001, "loss": 0.6714, "step": 40 }, { "epoch": 5.714285714285714, "grad_norm": 0.08176400512456894, "learning_rate": 6.173165676349103e-05, "loss": 0.6623, "step": 50 }, { "epoch": 6.857142857142857, "grad_norm": 0.056269895285367966, "learning_rate": 2.9289321881345254e-05, "loss": 0.6394, "step": 60 }, { "epoch": 8.0, "grad_norm": 0.04735175147652626, "learning_rate": 7.612046748871327e-06, "loss": 0.6334, "step": 70 }, { "epoch": 9.142857142857142, "grad_norm": 0.04113277792930603, "learning_rate": 0.0, "loss": 0.6326, "step": 80 }, { "epoch": 9.142857142857142, "step": 80, "total_flos": 3.28122019872768e+16, "train_loss": 0.7516828298568725, "train_runtime": 7940.6726, "train_samples_per_second": 0.344, "train_steps_per_second": 0.01 } ], "logging_steps": 10, "max_steps": 80, "num_input_tokens_seen": 0, "num_train_epochs": 10, "save_steps": 500, "total_flos": 3.28122019872768e+16, "train_batch_size": 8, "trial_name": null, "trial_params": null }