{
  "best_global_step": 141,
  "best_metric": 2.384185791015625e-07,
  "best_model_checkpoint": "./star_trek_guard_finetuned/checkpoint-141",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 423,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0071111111111111115,
      "grad_norm": 863.2054443359375,
      "learning_rate": 0.0,
      "loss": 31.4624,
      "step": 1
    },
    {
      "epoch": 0.07111111111111111,
      "grad_norm": 975.65576171875,
      "learning_rate": 4.186046511627907e-05,
      "loss": 35.8397,
      "step": 10
    },
    {
      "epoch": 0.14222222222222222,
      "grad_norm": 198.86065673828125,
      "learning_rate": 8.837209302325582e-05,
      "loss": 8.8149,
      "step": 20
    },
    {
      "epoch": 0.21333333333333335,
      "grad_norm": 8.080985069274902,
      "learning_rate": 0.00013488372093023256,
      "loss": 0.5709,
      "step": 30
    },
    {
      "epoch": 0.28444444444444444,
      "grad_norm": 0.00014901161193847656,
      "learning_rate": 0.0001813953488372093,
      "loss": 0.1449,
      "step": 40
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 231.45864868164062,
      "learning_rate": 0.00019987699691483048,
      "loss": 2.3554,
      "step": 50
    },
    {
      "epoch": 0.4266666666666667,
      "grad_norm": 0.17169412970542908,
      "learning_rate": 0.00019912640693269752,
      "loss": 0.7842,
      "step": 60
    },
    {
      "epoch": 0.49777777777777776,
      "grad_norm": 0.0002152260858565569,
      "learning_rate": 0.00019769868307835994,
      "loss": 0.3842,
      "step": 70
    },
    {
      "epoch": 0.5688888888888889,
      "grad_norm": 0.7667607665061951,
      "learning_rate": 0.00019560357815343577,
      "loss": 0.1437,
      "step": 80
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.00011813640594482422,
      "learning_rate": 0.00019285540384897073,
      "loss": 0.1796,
      "step": 90
    },
    {
      "epoch": 0.7111111111111111,
      "grad_norm": 0.014821142889559269,
      "learning_rate": 0.00018947293298207635,
      "loss": 0.0,
      "step": 100
    },
    {
      "epoch": 0.7822222222222223,
      "grad_norm": 0.0,
      "learning_rate": 0.0001854792712585539,
      "loss": 0.0002,
      "step": 110
    },
    {
      "epoch": 0.8533333333333334,
      "grad_norm": 0.0,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 120
    },
    {
      "epoch": 0.9244444444444444,
      "grad_norm": 0.0003181487263645977,
      "learning_rate": 0.0001757714869760335,
      "loss": 0.4035,
      "step": 130
    },
    {
      "epoch": 0.9955555555555555,
      "grad_norm": 0.0,
      "learning_rate": 0.00017012367842724887,
      "loss": 0.3667,
      "step": 140
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.384185791015625e-07,
      "eval_runtime": 24.7269,
      "eval_samples_per_second": 20.221,
      "eval_steps_per_second": 5.055,
      "step": 141
    },
    {
      "epoch": 1.064,
      "grad_norm": 0.00018907367484644055,
      "learning_rate": 0.00016399685405033167,
      "loss": 0.0714,
      "step": 150
    },
    {
      "epoch": 1.1351111111111112,
      "grad_norm": 48.62416076660156,
      "learning_rate": 0.00015743286626829437,
      "loss": 0.0278,
      "step": 160
    },
    {
      "epoch": 1.2062222222222223,
      "grad_norm": 0.006157203111797571,
      "learning_rate": 0.0001504765537734844,
      "loss": 0.0,
      "step": 170
    },
    {
      "epoch": 1.2773333333333334,
      "grad_norm": 0.016050906851887703,
      "learning_rate": 0.00014317543523384928,
      "loss": 0.0,
      "step": 180
    },
    {
      "epoch": 1.3484444444444446,
      "grad_norm": 0.0,
      "learning_rate": 0.00013557938469225167,
      "loss": 0.0737,
      "step": 190
    },
    {
      "epoch": 1.4195555555555557,
      "grad_norm": 0.0,
      "learning_rate": 0.00012774029087618446,
      "loss": 0.0,
      "step": 200
    },
    {
      "epoch": 1.4906666666666666,
      "grad_norm": 1.4126300811767578e-05,
      "learning_rate": 0.00011971170274514802,
      "loss": 0.0,
      "step": 210
    },
    {
      "epoch": 1.561777777777778,
      "grad_norm": 0.00016976980259642005,
      "learning_rate": 0.00011154846369695863,
      "loss": 0.0,
      "step": 220
    },
    {
      "epoch": 1.6328888888888888,
      "grad_norm": 0.00011539459228515625,
      "learning_rate": 0.00010330633693173082,
      "loss": 0.0,
      "step": 230
    },
    {
      "epoch": 1.704,
      "grad_norm": 0.0006262153037823737,
      "learning_rate": 9.504162453267777e-05,
      "loss": 0.0001,
      "step": 240
    },
    {
      "epoch": 1.775111111111111,
      "grad_norm": 3.820657730102539e-05,
      "learning_rate": 8.681078286579311e-05,
      "loss": 0.0,
      "step": 250
    },
    {
      "epoch": 1.8462222222222222,
      "grad_norm": 0.00015928725770208985,
      "learning_rate": 7.867003692562534e-05,
      "loss": 0.0,
      "step": 260
    },
    {
      "epoch": 1.9173333333333333,
      "grad_norm": 0.0005349747953005135,
      "learning_rate": 7.067499626155354e-05,
      "loss": 0.0,
      "step": 270
    },
    {
      "epoch": 1.9884444444444445,
      "grad_norm": 0.0,
      "learning_rate": 6.28802751081779e-05,
      "loss": 0.0,
      "step": 280
    },
    {
      "epoch": 2.0,
      "eval_loss": 3.0994415283203125e-06,
      "eval_runtime": 24.7326,
      "eval_samples_per_second": 20.216,
      "eval_steps_per_second": 5.054,
      "step": 282
    },
    {
      "epoch": 2.056888888888889,
      "grad_norm": 1.4424324035644531e-05,
      "learning_rate": 5.533911931471936e-05,
      "loss": 0.0,
      "step": 290
    },
    {
      "epoch": 2.128,
      "grad_norm": 2.6166439056396484e-05,
      "learning_rate": 4.810304262187852e-05,
      "loss": 0.0,
      "step": 300
    },
    {
      "epoch": 2.1991111111111112,
      "grad_norm": 9.179115295410156e-05,
      "learning_rate": 4.12214747707527e-05,
      "loss": 0.0,
      "step": 310
    },
    {
      "epoch": 2.2702222222222224,
      "grad_norm": 0.0,
      "learning_rate": 3.4741423847583134e-05,
      "loss": 0.0,
      "step": 320
    },
    {
      "epoch": 2.3413333333333335,
      "grad_norm": 2.7120113372802734e-05,
      "learning_rate": 2.87071551708603e-05,
      "loss": 0.0,
      "step": 330
    },
    {
      "epoch": 2.4124444444444446,
      "grad_norm": 6.794929504394531e-05,
      "learning_rate": 2.315988891431412e-05,
      "loss": 0.0,
      "step": 340
    },
    {
      "epoch": 2.4835555555555557,
      "grad_norm": 1.519918441772461e-05,
      "learning_rate": 1.8137518531330767e-05,
      "loss": 0.0,
      "step": 350
    },
    {
      "epoch": 2.554666666666667,
      "grad_norm": 84.58100891113281,
      "learning_rate": 1.3674351904242611e-05,
      "loss": 0.0247,
      "step": 360
    },
    {
      "epoch": 2.6257777777777775,
      "grad_norm": 0.0,
      "learning_rate": 9.80087698670411e-06,
      "loss": 0.0,
      "step": 370
    },
    {
      "epoch": 2.696888888888889,
      "grad_norm": 0.0,
      "learning_rate": 6.543553540053926e-06,
      "loss": 0.0,
      "step": 380
    },
    {
      "epoch": 2.768,
      "grad_norm": 0.0002226971264462918,
      "learning_rate": 3.924632386315186e-06,
      "loss": 0.0,
      "step": 390
    },
    {
      "epoch": 2.8391111111111114,
      "grad_norm": 0.00040970483678393066,
      "learning_rate": 1.9620034125190644e-06,
      "loss": 0.0,
      "step": 400
    },
    {
      "epoch": 2.910222222222222,
      "grad_norm": 1.4424324035644531e-05,
      "learning_rate": 6.690733646361857e-07,
      "loss": 0.0,
      "step": 410
    },
    {
      "epoch": 2.981333333333333,
      "grad_norm": 0.0001682293659541756,
      "learning_rate": 5.467426590739511e-08,
      "loss": 0.0,
      "step": 420
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.980232238769531e-07,
      "eval_runtime": 24.7472,
      "eval_samples_per_second": 20.204,
      "eval_steps_per_second": 5.051,
      "step": 423
    }
  ],
  "logging_steps": 10,
  "max_steps": 423,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 1.5205925781504e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}