{
  "best_metric": 0.8952568173408508,
  "best_model_checkpoint": "data/Gemma-2-2B_task-2_60-samples_config-1_full/checkpoint-57",
  "epoch": 16.869565217391305,
  "eval_steps": 500,
  "global_step": 97,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17391304347826086,
      "grad_norm": 0.4329001009464264,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.3173,
      "step": 1
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 0.5157751441001892,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.3553,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 0.44170382618904114,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.2928,
      "step": 4
    },
    {
      "epoch": 0.8695652173913043,
      "eval_loss": 1.3068894147872925,
      "eval_runtime": 5.3386,
      "eval_samples_per_second": 2.248,
      "eval_steps_per_second": 2.248,
      "step": 5
    },
    {
      "epoch": 1.0434782608695652,
      "grad_norm": 0.46065086126327515,
      "learning_rate": 2.4e-05,
      "loss": 1.305,
      "step": 6
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 0.4375300109386444,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 1.2895,
      "step": 8
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 0.35335251688957214,
      "learning_rate": 4e-05,
      "loss": 1.2992,
      "step": 10
    },
    {
      "epoch": 1.9130434782608696,
      "eval_loss": 1.2224934101104736,
      "eval_runtime": 5.0742,
      "eval_samples_per_second": 2.365,
      "eval_steps_per_second": 2.365,
      "step": 11
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 0.39072930812835693,
      "learning_rate": 4.8e-05,
      "loss": 1.2392,
      "step": 12
    },
    {
      "epoch": 2.4347826086956523,
      "grad_norm": 0.26168668270111084,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 1.1324,
      "step": 14
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 0.25940123200416565,
      "learning_rate": 6.400000000000001e-05,
      "loss": 1.1839,
      "step": 16
    },
    {
      "epoch": 2.9565217391304346,
      "eval_loss": 1.1169761419296265,
      "eval_runtime": 5.0686,
      "eval_samples_per_second": 2.368,
      "eval_steps_per_second": 2.368,
      "step": 17
    },
    {
      "epoch": 3.130434782608696,
      "grad_norm": 0.2605189383029938,
      "learning_rate": 7.2e-05,
      "loss": 1.0695,
      "step": 18
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.32380104064941406,
      "learning_rate": 8e-05,
      "loss": 1.0633,
      "step": 20
    },
    {
      "epoch": 3.8260869565217392,
      "grad_norm": 0.2902792692184448,
      "learning_rate": 8.800000000000001e-05,
      "loss": 1.0415,
      "step": 22
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.0407634973526,
      "eval_runtime": 5.0625,
      "eval_samples_per_second": 2.37,
      "eval_steps_per_second": 2.37,
      "step": 23
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 0.2631775438785553,
      "learning_rate": 9.6e-05,
      "loss": 1.0335,
      "step": 24
    },
    {
      "epoch": 4.521739130434782,
      "grad_norm": 0.2896063029766083,
      "learning_rate": 9.999512620046522e-05,
      "loss": 0.9847,
      "step": 26
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.3440544009208679,
      "learning_rate": 9.995614150494293e-05,
      "loss": 0.9451,
      "step": 28
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 0.9725618958473206,
      "eval_runtime": 5.0667,
      "eval_samples_per_second": 2.368,
      "eval_steps_per_second": 2.368,
      "step": 28
    },
    {
      "epoch": 5.217391304347826,
      "grad_norm": 0.22444793581962585,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.8814,
      "step": 30
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.19753581285476685,
      "learning_rate": 9.976136999909156e-05,
      "loss": 0.9001,
      "step": 32
    },
    {
      "epoch": 5.913043478260869,
      "grad_norm": 0.1629607230424881,
      "learning_rate": 9.96057350657239e-05,
      "loss": 0.8679,
      "step": 34
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 0.9312413334846497,
      "eval_runtime": 5.0133,
      "eval_samples_per_second": 2.394,
      "eval_steps_per_second": 2.394,
      "step": 34
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.17535679042339325,
      "learning_rate": 9.941141907232765e-05,
      "loss": 0.8387,
      "step": 36
    },
    {
      "epoch": 6.608695652173913,
      "grad_norm": 0.21582040190696716,
      "learning_rate": 9.917857354066931e-05,
      "loss": 0.854,
      "step": 38
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.19856537878513336,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.8006,
      "step": 40
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 0.9110665917396545,
      "eval_runtime": 5.028,
      "eval_samples_per_second": 2.387,
      "eval_steps_per_second": 2.387,
      "step": 40
    },
    {
      "epoch": 7.304347826086957,
      "grad_norm": 0.16799543797969818,
      "learning_rate": 9.859805002892732e-05,
      "loss": 0.8478,
      "step": 42
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.16270701587200165,
      "learning_rate": 9.825082472361557e-05,
      "loss": 0.7928,
      "step": 44
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.19171243906021118,
      "learning_rate": 9.786597487660337e-05,
      "loss": 0.7924,
      "step": 46
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.9004728198051453,
      "eval_runtime": 5.0154,
      "eval_samples_per_second": 2.393,
      "eval_steps_per_second": 2.393,
      "step": 46
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.16195151209831238,
      "learning_rate": 9.744380058222483e-05,
      "loss": 0.7896,
      "step": 48
    },
    {
      "epoch": 8.695652173913043,
      "grad_norm": 0.19511835277080536,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.7768,
      "step": 50
    },
    {
      "epoch": 8.869565217391305,
      "eval_loss": 0.89654541015625,
      "eval_runtime": 5.0165,
      "eval_samples_per_second": 2.392,
      "eval_steps_per_second": 2.392,
      "step": 51
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.18172132968902588,
      "learning_rate": 9.648882429441257e-05,
      "loss": 0.7658,
      "step": 52
    },
    {
      "epoch": 9.391304347826088,
      "grad_norm": 0.18167728185653687,
      "learning_rate": 9.595676696276172e-05,
      "loss": 0.7237,
      "step": 54
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.1947912722826004,
      "learning_rate": 9.538887392664544e-05,
      "loss": 0.7665,
      "step": 56
    },
    {
      "epoch": 9.91304347826087,
      "eval_loss": 0.8952568173408508,
      "eval_runtime": 5.0154,
      "eval_samples_per_second": 2.393,
      "eval_steps_per_second": 2.393,
      "step": 57
    },
    {
      "epoch": 10.08695652173913,
      "grad_norm": 0.18926352262496948,
      "learning_rate": 9.478558801197065e-05,
      "loss": 0.7526,
      "step": 58
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.19661836326122284,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.7222,
      "step": 60
    },
    {
      "epoch": 10.782608695652174,
      "grad_norm": 0.2336426079273224,
      "learning_rate": 9.347474647526095e-05,
      "loss": 0.6893,
      "step": 62
    },
    {
      "epoch": 10.956521739130435,
      "eval_loss": 0.9023671746253967,
      "eval_runtime": 5.0215,
      "eval_samples_per_second": 2.39,
      "eval_steps_per_second": 2.39,
      "step": 63
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.2892545163631439,
      "learning_rate": 9.276821300802534e-05,
      "loss": 0.6798,
      "step": 64
    },
    {
      "epoch": 11.478260869565217,
      "grad_norm": 0.23700620234012604,
      "learning_rate": 9.202833017478422e-05,
      "loss": 0.7339,
      "step": 66
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.2582704722881317,
      "learning_rate": 9.125567491391476e-05,
      "loss": 0.6515,
      "step": 68
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.9144725799560547,
      "eval_runtime": 5.016,
      "eval_samples_per_second": 2.392,
      "eval_steps_per_second": 2.392,
      "step": 69
    },
    {
      "epoch": 12.173913043478262,
      "grad_norm": 0.26315754652023315,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.66,
      "step": 70
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 0.30446359515190125,
      "learning_rate": 8.961448216775954e-05,
      "loss": 0.6077,
      "step": 72
    },
    {
      "epoch": 12.869565217391305,
      "grad_norm": 0.37722519040107727,
      "learning_rate": 8.874722443520899e-05,
      "loss": 0.6535,
      "step": 74
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 0.9321113228797913,
      "eval_runtime": 5.0166,
      "eval_samples_per_second": 2.392,
      "eval_steps_per_second": 2.392,
      "step": 74
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 0.362015962600708,
      "learning_rate": 8.784975278258783e-05,
      "loss": 0.6117,
      "step": 76
    },
    {
      "epoch": 13.565217391304348,
      "grad_norm": 0.3912771940231323,
      "learning_rate": 8.692276703129421e-05,
      "loss": 0.6129,
      "step": 78
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.5708993673324585,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.5737,
      "step": 80
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 0.972251832485199,
      "eval_runtime": 5.0143,
      "eval_samples_per_second": 2.393,
      "eval_steps_per_second": 2.393,
      "step": 80
    },
    {
      "epoch": 14.26086956521739,
      "grad_norm": 0.43917930126190186,
      "learning_rate": 8.498316702566828e-05,
      "loss": 0.5203,
      "step": 82
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.5976142287254333,
      "learning_rate": 8.397206521307584e-05,
      "loss": 0.5259,
      "step": 84
    },
    {
      "epoch": 14.956521739130435,
      "grad_norm": 0.4828043580055237,
      "learning_rate": 8.293447300593402e-05,
      "loss": 0.5352,
      "step": 86
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 0.9994226098060608,
      "eval_runtime": 5.0252,
      "eval_samples_per_second": 2.388,
      "eval_steps_per_second": 2.388,
      "step": 86
    },
    {
      "epoch": 15.304347826086957,
      "grad_norm": 0.5200834274291992,
      "learning_rate": 8.18711994874345e-05,
      "loss": 0.4989,
      "step": 88
    },
    {
      "epoch": 15.652173913043478,
      "grad_norm": 0.9574995040893555,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.4066,
      "step": 90
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.6054365038871765,
      "learning_rate": 7.967094433018508e-05,
      "loss": 0.5419,
      "step": 92
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.0519646406173706,
      "eval_runtime": 5.0124,
      "eval_samples_per_second": 2.394,
      "eval_steps_per_second": 2.394,
      "step": 92
    },
    {
      "epoch": 16.347826086956523,
      "grad_norm": 0.6740284562110901,
      "learning_rate": 7.85356783842216e-05,
      "loss": 0.4381,
      "step": 94
    },
    {
      "epoch": 16.695652173913043,
      "grad_norm": 1.5324946641921997,
      "learning_rate": 7.737816117462752e-05,
      "loss": 0.3587,
      "step": 96
    },
    {
      "epoch": 16.869565217391305,
      "eval_loss": 1.134497046470642,
      "eval_runtime": 5.0179,
      "eval_samples_per_second": 2.391,
      "eval_steps_per_second": 2.391,
      "step": 97
    },
    {
      "epoch": 16.869565217391305,
      "step": 97,
      "total_flos": 2.5909231861366784e+16,
      "train_loss": 0.8205352984752852,
      "train_runtime": 1125.2953,
      "train_samples_per_second": 2.044,
      "train_steps_per_second": 0.222
    }
  ],
  "logging_steps": 2,
  "max_steps": 250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 25,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 7,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.5909231861366784e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}