trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.4922118380062304,
  "eval_steps": 500,
  "global_step": 800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1557632398753894,
      "grad_norm": 5.09375,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 2.6073,
      "step": 50
    },
    {
      "epoch": 0.3115264797507788,
      "grad_norm": 3.53125,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.8752,
      "step": 100
    },
    {
      "epoch": 0.4672897196261682,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001,
      "loss": 0.5161,
      "step": 150
    },
    {
      "epoch": 0.6230529595015576,
      "grad_norm": 2.3125,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.3785,
      "step": 200
    },
    {
      "epoch": 0.778816199376947,
      "grad_norm": 2.125,
      "learning_rate": 0.0001666666666666667,
      "loss": 0.293,
      "step": 250
    },
    {
      "epoch": 0.9345794392523364,
      "grad_norm": 2.0,
      "learning_rate": 0.0002,
      "loss": 0.252,
      "step": 300
    },
    {
      "epoch": 1.0903426791277258,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00018491704374057317,
      "loss": 0.2341,
      "step": 350
    },
    {
      "epoch": 1.2461059190031152,
      "grad_norm": 2.015625,
      "learning_rate": 0.0001698340874811463,
      "loss": 0.183,
      "step": 400
    },
    {
      "epoch": 1.4018691588785046,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00015475113122171948,
      "loss": 0.1659,
      "step": 450
    },
    {
      "epoch": 1.557632398753894,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001396681749622926,
      "loss": 0.1368,
      "step": 500
    },
    {
      "epoch": 1.7133956386292835,
      "grad_norm": 0.408203125,
      "learning_rate": 0.00012458521870286577,
      "loss": 0.104,
      "step": 550
    },
    {
      "epoch": 1.8691588785046729,
      "grad_norm": 2.671875,
      "learning_rate": 0.00010950226244343893,
      "loss": 0.088,
      "step": 600
    },
    {
      "epoch": 2.0249221183800623,
      "grad_norm": 0.91015625,
      "learning_rate": 9.441930618401207e-05,
      "loss": 0.0903,
      "step": 650
    },
    {
      "epoch": 2.1806853582554515,
      "grad_norm": 1.8046875,
      "learning_rate": 7.933634992458521e-05,
      "loss": 0.0713,
      "step": 700
    },
    {
      "epoch": 2.336448598130841,
      "grad_norm": 0.2392578125,
      "learning_rate": 6.425339366515838e-05,
      "loss": 0.061,
      "step": 750
    },
    {
      "epoch": 2.4922118380062304,
      "grad_norm": 1.0,
      "learning_rate": 4.917043740573152e-05,
      "loss": 0.0581,
      "step": 800
    }
  ],
  "logging_steps": 50,
  "max_steps": 963,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 200,
  "total_flos": 0.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}