trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.364797081623347,
  "eval_steps": 500,
  "global_step": 800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.022799817601459188,
      "grad_norm": 1.421875,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 5.51,
      "step": 50
    },
    {
      "epoch": 0.045599635202918376,
      "grad_norm": 3.5625,
      "learning_rate": 6.666666666666667e-05,
      "loss": 3.5808,
      "step": 100
    },
    {
      "epoch": 0.06839945280437756,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001,
      "loss": 0.4537,
      "step": 150
    },
    {
      "epoch": 0.09119927040583675,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.2916,
      "step": 200
    },
    {
      "epoch": 0.11399908800729594,
      "grad_norm": 1.71875,
      "learning_rate": 0.0001666666666666667,
      "loss": 0.2631,
      "step": 250
    },
    {
      "epoch": 0.13679890560875513,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0002,
      "loss": 0.1738,
      "step": 300
    },
    {
      "epoch": 0.15959872321021432,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00019988560970029743,
      "loss": 0.169,
      "step": 350
    },
    {
      "epoch": 0.1823985408116735,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00019977121940059484,
      "loss": 0.1567,
      "step": 400
    },
    {
      "epoch": 0.2051983584131327,
      "grad_norm": 0.859375,
      "learning_rate": 0.00019965682910089226,
      "loss": 0.1632,
      "step": 450
    },
    {
      "epoch": 0.22799817601459188,
      "grad_norm": 1.046875,
      "learning_rate": 0.00019954243880118967,
      "loss": 0.1487,
      "step": 500
    },
    {
      "epoch": 0.2507979936160511,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0001994280485014871,
      "loss": 0.1939,
      "step": 550
    },
    {
      "epoch": 0.27359781121751026,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00019931365820178448,
      "loss": 0.1556,
      "step": 600
    },
    {
      "epoch": 0.29639762881896947,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019919926790208192,
      "loss": 0.1534,
      "step": 650
    },
    {
      "epoch": 0.31919744642042863,
      "grad_norm": 1.703125,
      "learning_rate": 0.00019908487760237934,
      "loss": 0.1342,
      "step": 700
    },
    {
      "epoch": 0.34199726402188785,
      "grad_norm": 1.421875,
      "learning_rate": 0.00019897048730267673,
      "loss": 0.1341,
      "step": 750
    },
    {
      "epoch": 0.364797081623347,
      "grad_norm": 0.875,
      "learning_rate": 0.00019885609700297417,
      "loss": 0.1588,
      "step": 800
    }
  ],
  "logging_steps": 50,
  "max_steps": 87720,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 200,
  "total_flos": 0.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}