trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.999965170629088,
  "global_step": 13458,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11,
      "learning_rate": 0.0001,
      "loss": 1.7126,
      "step": 500
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001,
      "loss": 1.659,
      "step": 1000
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001,
      "loss": 1.6436,
      "step": 1500
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001,
      "loss": 1.6407,
      "step": 2000
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 1.6362,
      "step": 2500
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001,
      "loss": 1.6323,
      "step": 3000
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 1.6257,
      "step": 3500
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 1.6234,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001,
      "loss": 1.616,
      "step": 4500
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001,
      "loss": 1.5855,
      "step": 5000
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001,
      "loss": 1.5857,
      "step": 5500
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001,
      "loss": 1.5813,
      "step": 6000
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001,
      "loss": 1.5828,
      "step": 6500
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001,
      "loss": 1.5814,
      "step": 7000
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0001,
      "loss": 1.5781,
      "step": 7500
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0001,
      "loss": 1.5807,
      "step": 8000
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0001,
      "loss": 1.5803,
      "step": 8500
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0001,
      "loss": 1.5741,
      "step": 9000
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0001,
      "loss": 1.5449,
      "step": 9500
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0001,
      "loss": 1.5479,
      "step": 10000
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0001,
      "loss": 1.5442,
      "step": 10500
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0001,
      "loss": 1.539,
      "step": 11000
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0001,
      "loss": 1.547,
      "step": 11500
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0001,
      "loss": 1.5444,
      "step": 12000
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0001,
      "loss": 1.5503,
      "step": 12500
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0001,
      "loss": 1.5464,
      "step": 13000
    },
    {
      "epoch": 3.0,
      "step": 13458,
      "total_flos": 9.452936982497219e+17,
      "train_loss": 1.5899839068188995,
      "train_runtime": 94528.3189,
      "train_samples_per_second": 9.112,
      "train_steps_per_second": 0.142
    }
  ],
  "max_steps": 13458,
  "num_train_epochs": 3,
  "total_flos": 9.452936982497219e+17,
  "trial_name": null,
  "trial_params": null
}