checkpoint-100/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.5384615384615383,
  "eval_steps": 50,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 38.090023040771484,
      "learning_rate": 7.75e-06,
      "loss": 3.4036,
      "step": 50
    },
    {
      "epoch": 0.7692307692307693,
      "eval_loss": 2.0667002201080322,
      "eval_runtime": 12.5463,
      "eval_samples_per_second": 10.441,
      "eval_steps_per_second": 1.355,
      "step": 50
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 76.00936889648438,
      "learning_rate": 5.300000000000001e-06,
      "loss": 2.1551,
      "step": 100
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 1.8017847537994385,
      "eval_runtime": 12.1789,
      "eval_samples_per_second": 10.756,
      "eval_steps_per_second": 1.396,
      "step": 100
    }
  ],
  "logging_steps": 50,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.82243166208e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}