last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.021052631578947368,
  "eval_steps": 3,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004210526315789474,
      "grad_norm": 6.040051460266113,
      "learning_rate": 2e-05,
      "loss": 5.2097,
      "step": 1
    },
    {
      "epoch": 0.004210526315789474,
      "eval_loss": 5.1479387283325195,
      "eval_runtime": 33.4641,
      "eval_samples_per_second": 2.988,
      "eval_steps_per_second": 1.494,
      "step": 1
    },
    {
      "epoch": 0.008421052631578947,
      "grad_norm": 6.306079387664795,
      "learning_rate": 4e-05,
      "loss": 5.1825,
      "step": 2
    },
    {
      "epoch": 0.01263157894736842,
      "grad_norm": 6.255680561065674,
      "learning_rate": 6e-05,
      "loss": 5.2103,
      "step": 3
    },
    {
      "epoch": 0.01263157894736842,
      "eval_loss": 5.074653625488281,
      "eval_runtime": 33.571,
      "eval_samples_per_second": 2.979,
      "eval_steps_per_second": 1.489,
      "step": 3
    },
    {
      "epoch": 0.016842105263157894,
      "grad_norm": 6.213514804840088,
      "learning_rate": 8e-05,
      "loss": 5.0592,
      "step": 4
    },
    {
      "epoch": 0.021052631578947368,
      "grad_norm": 6.902180194854736,
      "learning_rate": 0.0001,
      "loss": 4.7818,
      "step": 5
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2818744422236160.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}