{
  "best_metric": 0.49667900800704956,
  "best_model_checkpoint": "mikhail-panzo/fil_b64_le5_s4000/checkpoint-500",
  "epoch": 22.22222222222222,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 5.408687114715576,
      "learning_rate": 2.5000000000000004e-07,
      "loss": 0.8111,
      "step": 50
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 3.0843052864074707,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.7694,
      "step": 100
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 5.368913650512695,
      "learning_rate": 7.5e-07,
      "loss": 0.7562,
      "step": 150
    },
    {
      "epoch": 8.88888888888889,
      "grad_norm": 1.200212836265564,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.7184,
      "step": 200
    },
    {
      "epoch": 11.11111111111111,
      "grad_norm": 1.2329410314559937,
      "learning_rate": 1.25e-06,
      "loss": 0.7022,
      "step": 250
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 1.7799458503723145,
      "learning_rate": 1.5e-06,
      "loss": 0.6823,
      "step": 300
    },
    {
      "epoch": 15.555555555555555,
      "grad_norm": 2.0204672813415527,
      "learning_rate": 1.75e-06,
      "loss": 0.6635,
      "step": 350
    },
    {
      "epoch": 17.77777777777778,
      "grad_norm": 1.8028016090393066,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.6499,
      "step": 400
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.5485812425613403,
      "learning_rate": 2.25e-06,
      "loss": 0.6216,
      "step": 450
    },
    {
      "epoch": 22.22222222222222,
      "grad_norm": 1.0476795434951782,
      "learning_rate": 2.5e-06,
      "loss": 0.575,
      "step": 500
    },
    {
      "epoch": 22.22222222222222,
      "eval_loss": 0.49667900800704956,
      "eval_runtime": 8.3577,
      "eval_samples_per_second": 19.024,
      "eval_steps_per_second": 2.393,
      "step": 500
    }
  ],
  "logging_steps": 50,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 182,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6884761162771008.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}