last-checkpoint/trainer_state.json

{
  "best_metric": 2.4730148315429688,
  "best_model_checkpoint": "./outputs/checkpoint-500",
  "epoch": 0.36429872495446264,
  "eval_steps": 100,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 0.0002,
      "loss": 2.7202,
      "step": 100
    },
    {
      "epoch": 0.07,
      "eval_loss": 2.6173269748687744,
      "eval_runtime": 210.4944,
      "eval_samples_per_second": 29.806,
      "eval_steps_per_second": 3.729,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0002,
      "loss": 2.5807,
      "step": 200
    },
    {
      "epoch": 0.15,
      "eval_loss": 2.566875696182251,
      "eval_runtime": 206.05,
      "eval_samples_per_second": 30.449,
      "eval_steps_per_second": 3.81,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0002,
      "loss": 2.5403,
      "step": 300
    },
    {
      "epoch": 0.22,
      "eval_loss": 2.531456708908081,
      "eval_runtime": 205.7527,
      "eval_samples_per_second": 30.493,
      "eval_steps_per_second": 3.815,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0002,
      "loss": 2.5135,
      "step": 400
    },
    {
      "epoch": 0.29,
      "eval_loss": 2.4987542629241943,
      "eval_runtime": 205.7584,
      "eval_samples_per_second": 30.492,
      "eval_steps_per_second": 3.815,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0002,
      "loss": 2.4703,
      "step": 500
    },
    {
      "epoch": 0.36,
      "eval_loss": 2.4730148315429688,
      "eval_runtime": 205.8153,
      "eval_samples_per_second": 30.484,
      "eval_steps_per_second": 3.814,
      "step": 500
    }
  ],
  "logging_steps": 100,
  "max_steps": 4116,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.469294009229312e+16,
  "trial_name": null,
  "trial_params": null
}