coreset-selection/mix_diff_10

1{2  "best_global_step": null,3  "best_metric": null,4  "best_model_checkpoint": null,5  "epoch": 5.0,6  "eval_steps": 500,7  "global_step": 1300,8  "is_hyper_param_search": false,9  "is_local_process_zero": true,10  "is_world_process_zero": true,11  "log_history": [12    {13      "epoch": 1.9230769230769231,14      "grad_norm": 67.05139923095703,15      "learning_rate": 6.161538461538462e-06,16      "loss": -94.5512,17      "step": 50018    },19    {20      "epoch": 3.8461538461538463,21      "grad_norm": 98.70185852050781,22      "learning_rate": 2.3153846153846153e-06,23      "loss": -146.8585,24      "step": 100025    }26  ],27  "logging_steps": 500,28  "max_steps": 1300,29  "num_input_tokens_seen": 0,30  "num_train_epochs": 5,31  "save_steps": 500,32  "stateful_callbacks": {33    "TrainerControl": {34      "args": {35        "should_epoch_stop": false,36        "should_evaluate": false,37        "should_log": false,38        "should_save": true,39        "should_training_stop": true40      },41      "attributes": {}42    }43  },44  "total_flos": 0.0,45  "train_batch_size": 2,46  "trial_name": null,47  "trial_params": null48}49