last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.029607698001480384,
  "eval_steps": 2,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005921539600296077,
      "grad_norm": 0.7750168442726135,
      "learning_rate": 1e-05,
      "loss": 0.6543,
      "step": 1
    },
    {
      "epoch": 0.005921539600296077,
      "eval_loss": 0.6377593874931335,
      "eval_runtime": 73.9398,
      "eval_samples_per_second": 3.854,
      "eval_steps_per_second": 1.934,
      "step": 1
    },
    {
      "epoch": 0.011843079200592153,
      "grad_norm": 0.7847601175308228,
      "learning_rate": 2e-05,
      "loss": 0.7245,
      "step": 2
    },
    {
      "epoch": 0.011843079200592153,
      "eval_loss": 0.6308547258377075,
      "eval_runtime": 10.3795,
      "eval_samples_per_second": 27.458,
      "eval_steps_per_second": 13.777,
      "step": 2
    },
    {
      "epoch": 0.01776461880088823,
      "grad_norm": 0.8490102887153625,
      "learning_rate": 3e-05,
      "loss": 0.7567,
      "step": 3
    },
    {
      "epoch": 0.023686158401184307,
      "grad_norm": 0.7016004920005798,
      "learning_rate": 4e-05,
      "loss": 0.5979,
      "step": 4
    },
    {
      "epoch": 0.023686158401184307,
      "eval_loss": 0.614459753036499,
      "eval_runtime": 10.408,
      "eval_samples_per_second": 27.383,
      "eval_steps_per_second": 13.739,
      "step": 4
    },
    {
      "epoch": 0.029607698001480384,
      "grad_norm": 0.6877695322036743,
      "learning_rate": 5e-05,
      "loss": 0.5787,
      "step": 5
    }
  ],
  "logging_steps": 1,
  "max_steps": 5,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1957911173529600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}