trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.999958205046044,
  "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": 2.2308,
      "step": 500
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001,
      "loss": 2.1659,
      "step": 1000
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001,
      "loss": 2.1386,
      "step": 1500
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001,
      "loss": 2.1317,
      "step": 2000
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 2.1235,
      "step": 2500
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001,
      "loss": 2.117,
      "step": 3000
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 2.1029,
      "step": 3500
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 2.0994,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001,
      "loss": 2.0893,
      "step": 4500
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001,
      "loss": 2.0724,
      "step": 5000
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001,
      "loss": 2.0689,
      "step": 5500
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001,
      "loss": 2.0622,
      "step": 6000
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001,
      "loss": 2.0587,
      "step": 6500
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001,
      "loss": 2.056,
      "step": 7000
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0001,
      "loss": 2.0505,
      "step": 7500
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0001,
      "loss": 2.048,
      "step": 8000
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0001,
      "loss": 2.0461,
      "step": 8500
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0001,
      "loss": 2.0387,
      "step": 9000
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0001,
      "loss": 2.0286,
      "step": 9500
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0001,
      "loss": 2.0268,
      "step": 10000
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0001,
      "loss": 2.0224,
      "step": 10500
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0001,
      "loss": 2.0139,
      "step": 11000
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0001,
      "loss": 2.0187,
      "step": 11500
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0001,
      "loss": 2.0135,
      "step": 12000
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0001,
      "loss": 2.0193,
      "step": 12500
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0001,
      "loss": 2.0137,
      "step": 13000
    },
    {
      "epoch": 3.0,
      "step": 13458,
      "total_flos": 2.6528400748483584e+17,
      "train_loss": 2.0691928055687137,
      "train_runtime": 37382.5344,
      "train_samples_per_second": 23.041,
      "train_steps_per_second": 0.36
    }
  ],
  "max_steps": 13458,
  "num_train_epochs": 3,
  "total_flos": 2.6528400748483584e+17,
  "trial_name": null,
  "trial_params": null
}