trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.999745123027154,
  "global_step": 9561,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 2.5744,
      "step": 500
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001,
      "loss": 2.5289,
      "step": 1000
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001,
      "loss": 2.4993,
      "step": 1500
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001,
      "loss": 2.4688,
      "step": 2000
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 2.4584,
      "step": 2500
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001,
      "loss": 2.4395,
      "step": 3000
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001,
      "loss": 2.4191,
      "step": 3500
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001,
      "loss": 2.3905,
      "step": 4000
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001,
      "loss": 2.3809,
      "step": 4500
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0001,
      "loss": 2.3724,
      "step": 5000
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0001,
      "loss": 2.3675,
      "step": 5500
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0001,
      "loss": 2.3564,
      "step": 6000
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0001,
      "loss": 2.3443,
      "step": 6500
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0001,
      "loss": 2.3159,
      "step": 7000
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0001,
      "loss": 2.3183,
      "step": 7500
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0001,
      "loss": 2.301,
      "step": 8000
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0001,
      "loss": 2.3033,
      "step": 8500
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.0001,
      "loss": 2.3083,
      "step": 9000
    },
    {
      "epoch": 2.98,
      "learning_rate": 0.0001,
      "loss": 2.2948,
      "step": 9500
    },
    {
      "epoch": 3.0,
      "step": 9561,
      "total_flos": 1.7956612343026483e+17,
      "train_loss": 2.3911604716206503,
      "train_runtime": 21268.3554,
      "train_samples_per_second": 28.778,
      "train_steps_per_second": 0.45
    }
  ],
  "max_steps": 9561,
  "num_train_epochs": 3,
  "total_flos": 1.7956612343026483e+17,
  "trial_name": null,
  "trial_params": null
}