trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 9465,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 4.7358689910195457e-05,
      "loss": 2.4071,
      "step": 500
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.471737982039092e-05,
      "loss": 1.9175,
      "step": 1000
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.207606973058637e-05,
      "loss": 1.7769,
      "step": 1500
    },
    {
      "epoch": 0.63,
      "learning_rate": 3.943475964078183e-05,
      "loss": 1.6681,
      "step": 2000
    },
    {
      "epoch": 0.79,
      "learning_rate": 3.679344955097729e-05,
      "loss": 1.6051,
      "step": 2500
    },
    {
      "epoch": 0.95,
      "learning_rate": 3.415213946117274e-05,
      "loss": 1.5318,
      "step": 3000
    },
    {
      "epoch": 1.11,
      "learning_rate": 3.15108293713682e-05,
      "loss": 1.3432,
      "step": 3500
    },
    {
      "epoch": 1.27,
      "learning_rate": 2.8869519281563656e-05,
      "loss": 1.2927,
      "step": 4000
    },
    {
      "epoch": 1.43,
      "learning_rate": 2.6228209191759113e-05,
      "loss": 1.2615,
      "step": 4500
    },
    {
      "epoch": 1.58,
      "learning_rate": 2.358689910195457e-05,
      "loss": 1.254,
      "step": 5000
    },
    {
      "epoch": 1.74,
      "learning_rate": 2.094558901215003e-05,
      "loss": 1.2243,
      "step": 5500
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.8304278922345483e-05,
      "loss": 1.2075,
      "step": 6000
    },
    {
      "epoch": 2.06,
      "learning_rate": 1.5662968832540943e-05,
      "loss": 1.1144,
      "step": 6500
    },
    {
      "epoch": 2.22,
      "learning_rate": 1.3021658742736398e-05,
      "loss": 1.0082,
      "step": 7000
    },
    {
      "epoch": 2.38,
      "learning_rate": 1.0380348652931855e-05,
      "loss": 1.0019,
      "step": 7500
    },
    {
      "epoch": 2.54,
      "learning_rate": 7.739038563127311e-06,
      "loss": 0.9838,
      "step": 8000
    },
    {
      "epoch": 2.69,
      "learning_rate": 5.097728473322768e-06,
      "loss": 0.9748,
      "step": 8500
    },
    {
      "epoch": 2.85,
      "learning_rate": 2.456418383518225e-06,
      "loss": 0.967,
      "step": 9000
    },
    {
      "epoch": 3.0,
      "step": 9465,
      "total_flos": 1.953871571882803e+16,
      "train_loss": 1.3438021781264444,
      "train_runtime": 3123.1765,
      "train_samples_per_second": 30.297,
      "train_steps_per_second": 3.031
    }
  ],
  "max_steps": 9465,
  "num_train_epochs": 3,
  "total_flos": 1.953871571882803e+16,
  "trial_name": null,
  "trial_params": null
}