trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 100,
  "global_step": 891,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.34,
      "learning_rate": 1e-05,
      "loss": 2.4387,
      "step": 100
    },
    {
      "epoch": 0.34,
      "eval_loss": 1.9587514400482178,
      "eval_runtime": 41.9011,
      "eval_samples_per_second": 5.966,
      "eval_steps_per_second": 0.764,
      "step": 100
    },
    {
      "epoch": 0.67,
      "learning_rate": 1e-05,
      "loss": 1.8451,
      "step": 200
    },
    {
      "epoch": 0.67,
      "eval_loss": 1.839536428451538,
      "eval_runtime": 42.0627,
      "eval_samples_per_second": 5.944,
      "eval_steps_per_second": 0.761,
      "step": 200
    },
    {
      "epoch": 1.01,
      "learning_rate": 1e-05,
      "loss": 1.7983,
      "step": 300
    },
    {
      "epoch": 1.01,
      "eval_loss": 1.8240922689437866,
      "eval_runtime": 41.9798,
      "eval_samples_per_second": 5.955,
      "eval_steps_per_second": 0.762,
      "step": 300
    },
    {
      "epoch": 1.35,
      "learning_rate": 1e-05,
      "loss": 1.7717,
      "step": 400
    },
    {
      "epoch": 1.35,
      "eval_loss": 1.8145651817321777,
      "eval_runtime": 41.9722,
      "eval_samples_per_second": 5.956,
      "eval_steps_per_second": 0.762,
      "step": 400
    },
    {
      "epoch": 1.68,
      "learning_rate": 1e-05,
      "loss": 1.7786,
      "step": 500
    },
    {
      "epoch": 1.68,
      "eval_loss": 1.8092119693756104,
      "eval_runtime": 41.9735,
      "eval_samples_per_second": 5.956,
      "eval_steps_per_second": 0.762,
      "step": 500
    },
    {
      "epoch": 2.02,
      "learning_rate": 1e-05,
      "loss": 1.7616,
      "step": 600
    },
    {
      "epoch": 2.02,
      "eval_loss": 1.803665280342102,
      "eval_runtime": 41.6297,
      "eval_samples_per_second": 6.005,
      "eval_steps_per_second": 0.769,
      "step": 600
    },
    {
      "epoch": 2.36,
      "learning_rate": 1e-05,
      "loss": 1.7515,
      "step": 700
    },
    {
      "epoch": 2.36,
      "eval_loss": 1.8004812002182007,
      "eval_runtime": 42.1099,
      "eval_samples_per_second": 5.937,
      "eval_steps_per_second": 0.76,
      "step": 700
    },
    {
      "epoch": 2.69,
      "learning_rate": 1e-05,
      "loss": 1.7505,
      "step": 800
    },
    {
      "epoch": 2.69,
      "eval_loss": 1.7984752655029297,
      "eval_runtime": 42.0116,
      "eval_samples_per_second": 5.951,
      "eval_steps_per_second": 0.762,
      "step": 800
    }
  ],
  "logging_steps": 100,
  "max_steps": 891,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2.541658938273792e+16,
  "trial_name": null,
  "trial_params": null
}