trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "global_step": 4481,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11,
      "learning_rate": 2.665253291675965e-05,
      "loss": 2.8266,
      "step": 500
    },
    {
      "epoch": 0.22,
      "learning_rate": 2.3305065833519305e-05,
      "loss": 1.8389,
      "step": 1000
    },
    {
      "epoch": 0.33,
      "learning_rate": 1.9957598750278953e-05,
      "loss": 1.6042,
      "step": 1500
    },
    {
      "epoch": 0.45,
      "learning_rate": 1.661013166703861e-05,
      "loss": 1.5406,
      "step": 2000
    },
    {
      "epoch": 0.56,
      "learning_rate": 1.326266458379826e-05,
      "loss": 1.4462,
      "step": 2500
    },
    {
      "epoch": 0.67,
      "learning_rate": 9.915197500557913e-06,
      "loss": 1.4261,
      "step": 3000
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.567730417317563e-06,
      "loss": 1.3561,
      "step": 3500
    },
    {
      "epoch": 0.89,
      "learning_rate": 3.220263334077215e-06,
      "loss": 1.3676,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "step": 4481,
      "total_flos": 1.0536269121865728e+16,
      "train_loss": 1.6402719195893076,
      "train_runtime": 2258.7241,
      "train_samples_per_second": 23.803,
      "train_steps_per_second": 1.984
    }
  ],
  "max_steps": 4481,
  "num_train_epochs": 1,
  "total_flos": 1.0536269121865728e+16,
  "trial_name": null,
  "trial_params": null
}