{
  "best_global_step": 100,
  "best_metric": 1.198757290840149,
  "best_model_checkpoint": "/home/ricoiban/GEMMA/mnlp_chatsplaining/safety_model_output/checkpoint-100",
  "epoch": 2.0414507772020727,
  "eval_steps": 50,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.20725388601036268,
      "grad_norm": 21.424072265625,
      "learning_rate": 0.00019666666666666666,
      "loss": 11.0272,
      "mean_token_accuracy": 0.296875,
      "num_tokens": 21984.0,
      "step": 10
    },
    {
      "epoch": 0.41450777202072536,
      "grad_norm": 3.1715638637542725,
      "learning_rate": 0.00018833333333333335,
      "loss": 0.7306,
      "mean_token_accuracy": 0.63125,
      "num_tokens": 44111.0,
      "step": 20
    },
    {
      "epoch": 0.6217616580310881,
      "grad_norm": 6.497854232788086,
      "learning_rate": 0.00018,
      "loss": 0.6675,
      "mean_token_accuracy": 0.684375,
      "num_tokens": 68120.0,
      "step": 30
    },
    {
      "epoch": 0.8290155440414507,
      "grad_norm": 2.4520726203918457,
      "learning_rate": 0.00017166666666666667,
      "loss": 0.6535,
      "mean_token_accuracy": 0.68125,
      "num_tokens": 89669.0,
      "step": 40
    },
    {
      "epoch": 1.0207253886010363,
      "grad_norm": 4.587820529937744,
      "learning_rate": 0.00016333333333333334,
      "loss": 0.6026,
      "mean_token_accuracy": 0.7128378378378378,
      "num_tokens": 111297.0,
      "step": 50
    },
    {
      "epoch": 1.0207253886010363,
      "eval_loss": 1.4000602960586548,
      "eval_mean_token_accuracy": 0.5,
      "eval_num_tokens": 111297.0,
      "eval_runtime": 0.1262,
      "eval_samples_per_second": 7.921,
      "eval_steps_per_second": 7.921,
      "step": 50
    },
    {
      "epoch": 1.2279792746113989,
      "grad_norm": 2.7760419845581055,
      "learning_rate": 0.000155,
      "loss": 0.4794,
      "mean_token_accuracy": 0.815625,
      "num_tokens": 133338.0,
      "step": 60
    },
    {
      "epoch": 1.4352331606217616,
      "grad_norm": 7.210590362548828,
      "learning_rate": 0.00014666666666666666,
      "loss": 0.4987,
      "mean_token_accuracy": 0.79375,
      "num_tokens": 156589.0,
      "step": 70
    },
    {
      "epoch": 1.6424870466321244,
      "grad_norm": 3.8868515491485596,
      "learning_rate": 0.00013833333333333333,
      "loss": 0.4606,
      "mean_token_accuracy": 0.7875,
      "num_tokens": 178995.0,
      "step": 80
    },
    {
      "epoch": 1.849740932642487,
      "grad_norm": 2.4217886924743652,
      "learning_rate": 0.00013000000000000002,
      "loss": 0.5303,
      "mean_token_accuracy": 0.771875,
      "num_tokens": 203491.0,
      "step": 90
    },
    {
      "epoch": 2.0414507772020727,
      "grad_norm": 3.1198058128356934,
      "learning_rate": 0.00012166666666666667,
      "loss": 0.4854,
      "mean_token_accuracy": 0.8108108108108109,
      "num_tokens": 223362.0,
      "step": 100
    },
    {
      "epoch": 2.0414507772020727,
      "eval_loss": 1.198757290840149,
      "eval_mean_token_accuracy": 0.5,
      "eval_num_tokens": 223362.0,
      "eval_runtime": 0.1833,
      "eval_samples_per_second": 5.455,
      "eval_steps_per_second": 5.455,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 245,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 852188282880000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}