{
  "best_metric": 0.8902506232261658,
  "best_model_checkpoint": "data/Gemma-2-2B_task-2_60-samples_config-1_full_auto/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": 0.40798550844192505,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.3007,
      "step": 1
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 0.4898965358734131,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.3435,
      "step": 2
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 0.41924849152565,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.2771,
      "step": 4
    },
    {
      "epoch": 0.8695652173913043,
      "eval_loss": 1.301425576210022,
      "eval_runtime": 5.4066,
      "eval_samples_per_second": 2.22,
      "eval_steps_per_second": 2.22,
      "step": 5
    },
    {
      "epoch": 1.0434782608695652,
      "grad_norm": 0.44152921438217163,
      "learning_rate": 2.4e-05,
      "loss": 1.2883,
      "step": 6
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 0.4159512519836426,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 1.2757,
      "step": 8
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 0.34436213970184326,
      "learning_rate": 4e-05,
      "loss": 1.2877,
      "step": 10
    },
    {
      "epoch": 1.9130434782608696,
      "eval_loss": 1.220075011253357,
      "eval_runtime": 5.171,
      "eval_samples_per_second": 2.321,
      "eval_steps_per_second": 2.321,
      "step": 11
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 0.3733598589897156,
      "learning_rate": 4.8e-05,
      "loss": 1.2311,
      "step": 12
    },
    {
      "epoch": 2.4347826086956523,
      "grad_norm": 0.2551831305027008,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 1.1287,
      "step": 14
    },
    {
      "epoch": 2.782608695652174,
      "grad_norm": 0.24580122530460358,
      "learning_rate": 6.400000000000001e-05,
      "loss": 1.1635,
      "step": 16
    },
    {
      "epoch": 2.9565217391304346,
      "eval_loss": 1.1143940687179565,
      "eval_runtime": 5.1607,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 2.325,
      "step": 17
    },
    {
      "epoch": 3.130434782608696,
      "grad_norm": 0.26780688762664795,
      "learning_rate": 7.2e-05,
      "loss": 1.0634,
      "step": 18
    },
    {
      "epoch": 3.4782608695652173,
      "grad_norm": 0.3455520570278168,
      "learning_rate": 8e-05,
      "loss": 1.0534,
      "step": 20
    },
    {
      "epoch": 3.8260869565217392,
      "grad_norm": 0.2788378894329071,
      "learning_rate": 8.800000000000001e-05,
      "loss": 1.0236,
      "step": 22
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.0384857654571533,
      "eval_runtime": 5.1617,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 2.325,
      "step": 23
    },
    {
      "epoch": 4.173913043478261,
      "grad_norm": 0.2767649292945862,
      "learning_rate": 9.6e-05,
      "loss": 1.0295,
      "step": 24
    },
    {
      "epoch": 4.521739130434782,
      "grad_norm": 0.2858169972896576,
      "learning_rate": 9.999512620046522e-05,
      "loss": 0.9799,
      "step": 26
    },
    {
      "epoch": 4.869565217391305,
      "grad_norm": 0.33711665868759155,
      "learning_rate": 9.995614150494293e-05,
      "loss": 0.9255,
      "step": 28
    },
    {
      "epoch": 4.869565217391305,
      "eval_loss": 0.9710798263549805,
      "eval_runtime": 5.162,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 2.325,
      "step": 28
    },
    {
      "epoch": 5.217391304347826,
      "grad_norm": 0.2209019809961319,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.8708,
      "step": 30
    },
    {
      "epoch": 5.565217391304348,
      "grad_norm": 0.20020312070846558,
      "learning_rate": 9.976136999909156e-05,
      "loss": 0.8939,
      "step": 32
    },
    {
      "epoch": 5.913043478260869,
      "grad_norm": 0.1745641827583313,
      "learning_rate": 9.96057350657239e-05,
      "loss": 0.8522,
      "step": 34
    },
    {
      "epoch": 5.913043478260869,
      "eval_loss": 0.928705632686615,
      "eval_runtime": 5.1649,
      "eval_samples_per_second": 2.323,
      "eval_steps_per_second": 2.323,
      "step": 34
    },
    {
      "epoch": 6.260869565217392,
      "grad_norm": 0.18299494683742523,
      "learning_rate": 9.941141907232765e-05,
      "loss": 0.8272,
      "step": 36
    },
    {
      "epoch": 6.608695652173913,
      "grad_norm": 0.2237248420715332,
      "learning_rate": 9.917857354066931e-05,
      "loss": 0.8394,
      "step": 38
    },
    {
      "epoch": 6.956521739130435,
      "grad_norm": 0.17994999885559082,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.7873,
      "step": 40
    },
    {
      "epoch": 6.956521739130435,
      "eval_loss": 0.9061260223388672,
      "eval_runtime": 5.1612,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 2.325,
      "step": 40
    },
    {
      "epoch": 7.304347826086957,
      "grad_norm": 0.1722320318222046,
      "learning_rate": 9.859805002892732e-05,
      "loss": 0.8364,
      "step": 42
    },
    {
      "epoch": 7.6521739130434785,
      "grad_norm": 0.19566795229911804,
      "learning_rate": 9.825082472361557e-05,
      "loss": 0.7777,
      "step": 44
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.19303321838378906,
      "learning_rate": 9.786597487660337e-05,
      "loss": 0.7746,
      "step": 46
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.8943353295326233,
      "eval_runtime": 5.16,
      "eval_samples_per_second": 2.326,
      "eval_steps_per_second": 2.326,
      "step": 46
    },
    {
      "epoch": 8.347826086956522,
      "grad_norm": 0.167044997215271,
      "learning_rate": 9.744380058222483e-05,
      "loss": 0.7712,
      "step": 48
    },
    {
      "epoch": 8.695652173913043,
      "grad_norm": 0.18928879499435425,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.7645,
      "step": 50
    },
    {
      "epoch": 8.869565217391305,
      "eval_loss": 0.8905796408653259,
      "eval_runtime": 5.1612,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 2.325,
      "step": 51
    },
    {
      "epoch": 9.043478260869565,
      "grad_norm": 0.19295816123485565,
      "learning_rate": 9.648882429441257e-05,
      "loss": 0.752,
      "step": 52
    },
    {
      "epoch": 9.391304347826088,
      "grad_norm": 0.19442199170589447,
      "learning_rate": 9.595676696276172e-05,
      "loss": 0.7035,
      "step": 54
    },
    {
      "epoch": 9.73913043478261,
      "grad_norm": 0.20344549417495728,
      "learning_rate": 9.538887392664544e-05,
      "loss": 0.7497,
      "step": 56
    },
    {
      "epoch": 9.91304347826087,
      "eval_loss": 0.8902506232261658,
      "eval_runtime": 5.1624,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 2.325,
      "step": 57
    },
    {
      "epoch": 10.08695652173913,
      "grad_norm": 0.19857479631900787,
      "learning_rate": 9.478558801197065e-05,
      "loss": 0.7399,
      "step": 58
    },
    {
      "epoch": 10.434782608695652,
      "grad_norm": 0.21007387340068817,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.7034,
      "step": 60
    },
    {
      "epoch": 10.782608695652174,
      "grad_norm": 0.23670177161693573,
      "learning_rate": 9.347474647526095e-05,
      "loss": 0.677,
      "step": 62
    },
    {
      "epoch": 10.956521739130435,
      "eval_loss": 0.8981442451477051,
      "eval_runtime": 5.1692,
      "eval_samples_per_second": 2.321,
      "eval_steps_per_second": 2.321,
      "step": 63
    },
    {
      "epoch": 11.130434782608695,
      "grad_norm": 0.2575482726097107,
      "learning_rate": 9.276821300802534e-05,
      "loss": 0.655,
      "step": 64
    },
    {
      "epoch": 11.478260869565217,
      "grad_norm": 0.25422921776771545,
      "learning_rate": 9.202833017478422e-05,
      "loss": 0.7144,
      "step": 66
    },
    {
      "epoch": 11.826086956521738,
      "grad_norm": 0.27919456362724304,
      "learning_rate": 9.125567491391476e-05,
      "loss": 0.6337,
      "step": 68
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.9132208228111267,
      "eval_runtime": 5.1591,
      "eval_samples_per_second": 2.326,
      "eval_steps_per_second": 2.326,
      "step": 69
    },
    {
      "epoch": 12.173913043478262,
      "grad_norm": 0.26951611042022705,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.641,
      "step": 70
    },
    {
      "epoch": 12.521739130434783,
      "grad_norm": 0.31187334656715393,
      "learning_rate": 8.961448216775954e-05,
      "loss": 0.5881,
      "step": 72
    },
    {
      "epoch": 12.869565217391305,
      "grad_norm": 0.3737937808036804,
      "learning_rate": 8.874722443520899e-05,
      "loss": 0.637,
      "step": 74
    },
    {
      "epoch": 12.869565217391305,
      "eval_loss": 0.9354825615882874,
      "eval_runtime": 5.1674,
      "eval_samples_per_second": 2.322,
      "eval_steps_per_second": 2.322,
      "step": 74
    },
    {
      "epoch": 13.217391304347826,
      "grad_norm": 0.374449223279953,
      "learning_rate": 8.784975278258783e-05,
      "loss": 0.5865,
      "step": 76
    },
    {
      "epoch": 13.565217391304348,
      "grad_norm": 0.3786736726760864,
      "learning_rate": 8.692276703129421e-05,
      "loss": 0.5924,
      "step": 78
    },
    {
      "epoch": 13.91304347826087,
      "grad_norm": 0.5075408220291138,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.5524,
      "step": 80
    },
    {
      "epoch": 13.91304347826087,
      "eval_loss": 0.9834840893745422,
      "eval_runtime": 5.161,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 2.325,
      "step": 80
    },
    {
      "epoch": 14.26086956521739,
      "grad_norm": 0.4344421923160553,
      "learning_rate": 8.498316702566828e-05,
      "loss": 0.5021,
      "step": 82
    },
    {
      "epoch": 14.608695652173914,
      "grad_norm": 0.5073829889297485,
      "learning_rate": 8.397206521307584e-05,
      "loss": 0.5039,
      "step": 84
    },
    {
      "epoch": 14.956521739130435,
      "grad_norm": 0.5042697191238403,
      "learning_rate": 8.293447300593402e-05,
      "loss": 0.5137,
      "step": 86
    },
    {
      "epoch": 14.956521739130435,
      "eval_loss": 1.0105966329574585,
      "eval_runtime": 5.1608,
      "eval_samples_per_second": 2.325,
      "eval_steps_per_second": 2.325,
      "step": 86
    },
    {
      "epoch": 15.304347826086957,
      "grad_norm": 0.5449661016464233,
      "learning_rate": 8.18711994874345e-05,
      "loss": 0.4789,
      "step": 88
    },
    {
      "epoch": 15.652173913043478,
      "grad_norm": 0.5908651947975159,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.3888,
      "step": 90
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.5828967094421387,
      "learning_rate": 7.967094433018508e-05,
      "loss": 0.5166,
      "step": 92
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.093218445777893,
      "eval_runtime": 5.1626,
      "eval_samples_per_second": 2.324,
      "eval_steps_per_second": 2.324,
      "step": 92
    },
    {
      "epoch": 16.347826086956523,
      "grad_norm": 0.5814247727394104,
      "learning_rate": 7.85356783842216e-05,
      "loss": 0.4109,
      "step": 94
    },
    {
      "epoch": 16.695652173913043,
      "grad_norm": 1.3556056022644043,
      "learning_rate": 7.737816117462752e-05,
      "loss": 0.3388,
      "step": 96
    },
    {
      "epoch": 16.869565217391305,
      "eval_loss": 1.132517695426941,
      "eval_runtime": 5.1642,
      "eval_samples_per_second": 2.324,
      "eval_steps_per_second": 2.324,
      "step": 97
    },
    {
      "epoch": 16.869565217391305,
      "step": 97,
      "total_flos": 2.672993926905856e+16,
      "train_loss": 0.8043659771840597,
      "train_runtime": 1160.4735,
      "train_samples_per_second": 1.982,
      "train_steps_per_second": 0.215
    }
  ],
  "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": 2.672993926905856e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}