{
  "best_global_step": 100,
  "best_metric": 0.0743941143155098,
  "best_model_checkpoint": "/content/models/gemma_qlora_lmh_inst/checkpoint-100",
  "epoch": 2.0,
  "eval_steps": 20,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.200686830282211,
      "epoch": 0.40404040404040403,
      "grad_norm": 2.258388042449951,
      "learning_rate": 8.1e-06,
      "loss": 0.1306,
      "mean_token_accuracy": 0.9350446417927742,
      "num_tokens": 237653.0,
      "step": 20
    },
    {
      "epoch": 0.40404040404040403,
      "eval_entropy": 2.201928762289194,
      "eval_loss": 0.0838593915104866,
      "eval_mean_token_accuracy": 0.9620879109089191,
      "eval_num_tokens": 237653.0,
      "eval_runtime": 5.7826,
      "eval_samples_per_second": 34.068,
      "eval_steps_per_second": 2.248,
      "step": 20
    },
    {
      "entropy": 2.203989064693451,
      "epoch": 0.8080808080808081,
      "grad_norm": 3.0667307376861572,
      "learning_rate": 6.1e-06,
      "loss": 0.0797,
      "mean_token_accuracy": 0.9622767880558968,
      "num_tokens": 473622.0,
      "step": 40
    },
    {
      "epoch": 0.8080808080808081,
      "eval_entropy": 2.2022548638857327,
      "eval_loss": 0.07604048401117325,
      "eval_mean_token_accuracy": 0.9581043995343722,
      "eval_num_tokens": 473622.0,
      "eval_runtime": 5.7858,
      "eval_samples_per_second": 34.049,
      "eval_steps_per_second": 2.247,
      "step": 40
    },
    {
      "entropy": 2.2022441717294545,
      "epoch": 1.202020202020202,
      "grad_norm": 2.590899705886841,
      "learning_rate": 4.1e-06,
      "loss": 0.0656,
      "mean_token_accuracy": 0.9664224661313571,
      "num_tokens": 702354.0,
      "step": 60
    },
    {
      "epoch": 1.202020202020202,
      "eval_entropy": 2.1788861201359677,
      "eval_loss": 0.08057427406311035,
      "eval_mean_token_accuracy": 0.9635989024088933,
      "eval_num_tokens": 702354.0,
      "eval_runtime": 5.7897,
      "eval_samples_per_second": 34.026,
      "eval_steps_per_second": 2.245,
      "step": 60
    },
    {
      "entropy": 2.19284747838974,
      "epoch": 1.606060606060606,
      "grad_norm": 2.1625924110412598,
      "learning_rate": 2.1000000000000002e-06,
      "loss": 0.0579,
      "mean_token_accuracy": 0.9754464238882065,
      "num_tokens": 937654.0,
      "step": 80
    },
    {
      "epoch": 1.606060606060606,
      "eval_entropy": 2.178425770539504,
      "eval_loss": 0.07602041214704514,
      "eval_mean_token_accuracy": 0.965659343279325,
      "eval_num_tokens": 937654.0,
      "eval_runtime": 5.786,
      "eval_samples_per_second": 34.047,
      "eval_steps_per_second": 2.247,
      "step": 80
    },
    {
      "entropy": 2.1867658969683523,
      "epoch": 2.0,
      "grad_norm": 1.2462379932403564,
      "learning_rate": 1.0000000000000001e-07,
      "loss": 0.0512,
      "mean_token_accuracy": 0.9764194106444334,
      "num_tokens": 1165538.0,
      "step": 100
    },
    {
      "epoch": 2.0,
      "eval_entropy": 2.181581442172711,
      "eval_loss": 0.0743941143155098,
      "eval_mean_token_accuracy": 0.963598906993866,
      "eval_num_tokens": 1165538.0,
      "eval_runtime": 5.7817,
      "eval_samples_per_second": 34.073,
      "eval_steps_per_second": 2.248,
      "step": 100
    }
  ],
  "logging_steps": 20,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 20,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.9940993720561664e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}