{
  "best_metric": 0.04085824266076088,
  "best_model_checkpoint": "./teapotllm/checkpoint-4212",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 10530,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.11193174868822098,
      "learning_rate": 4e-05,
      "loss": 0.984,
      "step": 2106
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.043082017451524734,
      "eval_runtime": 23.1834,
      "eval_samples_per_second": 45.464,
      "eval_steps_per_second": 5.694,
      "step": 2106
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.14396128058433533,
      "learning_rate": 3e-05,
      "loss": 0.0419,
      "step": 4212
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.04085824266076088,
      "eval_runtime": 23.1567,
      "eval_samples_per_second": 45.516,
      "eval_steps_per_second": 5.7,
      "step": 4212
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.3134700059890747,
      "learning_rate": 2e-05,
      "loss": 0.0354,
      "step": 6318
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.04090132564306259,
      "eval_runtime": 23.171,
      "eval_samples_per_second": 45.488,
      "eval_steps_per_second": 5.697,
      "step": 6318
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.06102537363767624,
      "learning_rate": 1e-05,
      "loss": 0.0313,
      "step": 8424
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.041253600269556046,
      "eval_runtime": 23.148,
      "eval_samples_per_second": 45.533,
      "eval_steps_per_second": 5.702,
      "step": 8424
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3205956816673279,
      "learning_rate": 0.0,
      "loss": 0.029,
      "step": 10530
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.041361190378665924,
      "eval_runtime": 23.1783,
      "eval_samples_per_second": 45.473,
      "eval_steps_per_second": 5.695,
      "step": 10530
    }
  ],
  "logging_steps": 500,
  "max_steps": 10530,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.442098984255488e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}