{ "best_metric": null, "best_model_checkpoint": null, "epoch": 2.9333333333333336, "eval_steps": 100, "global_step": 66, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.4444444444444444, "grad_norm": 24.911563873291016, "learning_rate": 4.0000000000000003e-07, "loss": 0.6944, "step": 10 }, { "epoch": 0.8888888888888888, "grad_norm": 20.48420524597168, "learning_rate": 8.000000000000001e-07, "loss": 0.6317, "step": 20 }, { "epoch": 1.3333333333333333, "grad_norm": 21.66866111755371, "learning_rate": 1.2000000000000002e-06, "loss": 0.5401, "step": 30 }, { "epoch": 1.7777777777777777, "grad_norm": 12.989063262939453, "learning_rate": 1.6000000000000001e-06, "loss": 0.4192, "step": 40 }, { "epoch": 2.2222222222222223, "grad_norm": 8.98383617401123, "learning_rate": 2.0000000000000003e-06, "loss": 0.29, "step": 50 }, { "epoch": 2.6666666666666665, "grad_norm": 5.708517074584961, "learning_rate": 2.4000000000000003e-06, "loss": 0.2052, "step": 60 }, { "epoch": 2.9333333333333336, "step": 66, "total_flos": 3605999510880.0, "train_loss": 0.4347104240547527, "train_runtime": 17.5364, "train_samples_per_second": 121.12, "train_steps_per_second": 3.764 }, { "epoch": 2.9333333333333336, "eval_accuracy": 1.0, "eval_accuracy_label_bluesky": 1.0, "eval_accuracy_label_specific": 1.0, "eval_f1": 1.0, "eval_loss": 0.1264917105436325, "eval_precision": 1.0, "eval_recall": 1.0, "eval_runtime": 0.8403, "eval_samples_per_second": 361.76, "eval_steps_per_second": 22.61, "step": 66 } ], "logging_steps": 10, "max_steps": 66, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 1000, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 3605999510880.0, "train_batch_size": 16, "trial_name": null, "trial_params": null }