{
  "best_metric": 0.927240788936615,
  "best_model_checkpoint": "data/Gemma-2-2B_task-1_120-samples_config-1_full_auto/checkpoint-66",
  "epoch": 13.0,
  "eval_steps": 500,
  "global_step": 143,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09090909090909091,
      "grad_norm": 1.3007901906967163,
      "learning_rate": 1.818181818181818e-06,
      "loss": 2.2296,
      "step": 1
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 1.2423328161239624,
      "learning_rate": 3.636363636363636e-06,
      "loss": 2.249,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 1.2909067869186401,
      "learning_rate": 7.272727272727272e-06,
      "loss": 2.3573,
      "step": 4
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 1.4190131425857544,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 2.3105,
      "step": 6
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 1.294917106628418,
      "learning_rate": 1.4545454545454545e-05,
      "loss": 2.2242,
      "step": 8
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 1.3501969575881958,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 2.3294,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.200763702392578,
      "eval_runtime": 2.1241,
      "eval_samples_per_second": 11.299,
      "eval_steps_per_second": 11.299,
      "step": 11
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 1.3492711782455444,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 2.2238,
      "step": 12
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 1.1932955980300903,
      "learning_rate": 2.5454545454545454e-05,
      "loss": 2.1806,
      "step": 14
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 1.1321687698364258,
      "learning_rate": 2.909090909090909e-05,
      "loss": 1.9931,
      "step": 16
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 1.0956549644470215,
      "learning_rate": 3.272727272727273e-05,
      "loss": 2.0067,
      "step": 18
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 1.1535120010375977,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 1.8463,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.817171037197113,
      "learning_rate": 4e-05,
      "loss": 1.6286,
      "step": 22
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.6357680559158325,
      "eval_runtime": 2.1558,
      "eval_samples_per_second": 11.133,
      "eval_steps_per_second": 11.133,
      "step": 22
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 0.8934890627861023,
      "learning_rate": 4.3636363636363636e-05,
      "loss": 1.556,
      "step": 24
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 0.8142895698547363,
      "learning_rate": 4.7272727272727275e-05,
      "loss": 1.5426,
      "step": 26
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.919063925743103,
      "learning_rate": 5.090909090909091e-05,
      "loss": 1.4179,
      "step": 28
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.9269520044326782,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 1.3439,
      "step": 30
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.9637172222137451,
      "learning_rate": 5.818181818181818e-05,
      "loss": 1.2359,
      "step": 32
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.1648658514022827,
      "eval_runtime": 2.1541,
      "eval_samples_per_second": 11.142,
      "eval_steps_per_second": 11.142,
      "step": 33
    },
    {
      "epoch": 3.090909090909091,
      "grad_norm": 0.793499767780304,
      "learning_rate": 6.181818181818182e-05,
      "loss": 1.1798,
      "step": 34
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 0.7822566032409668,
      "learning_rate": 6.545454545454546e-05,
      "loss": 1.0478,
      "step": 36
    },
    {
      "epoch": 3.4545454545454546,
      "grad_norm": 0.6562247276306152,
      "learning_rate": 6.90909090909091e-05,
      "loss": 1.0335,
      "step": 38
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.6146300435066223,
      "learning_rate": 7.272727272727273e-05,
      "loss": 1.0407,
      "step": 40
    },
    {
      "epoch": 3.8181818181818183,
      "grad_norm": 0.6027118563652039,
      "learning_rate": 7.636363636363637e-05,
      "loss": 1.0345,
      "step": 42
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.5640882849693298,
      "learning_rate": 8e-05,
      "loss": 0.9635,
      "step": 44
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.9961578249931335,
      "eval_runtime": 2.156,
      "eval_samples_per_second": 11.132,
      "eval_steps_per_second": 11.132,
      "step": 44
    },
    {
      "epoch": 4.181818181818182,
      "grad_norm": 0.5649698972702026,
      "learning_rate": 8.363636363636364e-05,
      "loss": 0.9352,
      "step": 46
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 0.6958158612251282,
      "learning_rate": 8.727272727272727e-05,
      "loss": 0.9637,
      "step": 48
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 0.588526725769043,
      "learning_rate": 9.090909090909092e-05,
      "loss": 0.9095,
      "step": 50
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 0.5746359825134277,
      "learning_rate": 9.454545454545455e-05,
      "loss": 0.8988,
      "step": 52
    },
    {
      "epoch": 4.909090909090909,
      "grad_norm": 0.5816696882247925,
      "learning_rate": 9.818181818181818e-05,
      "loss": 0.8326,
      "step": 54
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.9410693049430847,
      "eval_runtime": 2.1521,
      "eval_samples_per_second": 11.152,
      "eval_steps_per_second": 11.152,
      "step": 55
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 0.5797222852706909,
      "learning_rate": 9.999899300364532e-05,
      "loss": 0.8265,
      "step": 56
    },
    {
      "epoch": 5.2727272727272725,
      "grad_norm": 0.6078121662139893,
      "learning_rate": 9.99909372761763e-05,
      "loss": 0.8295,
      "step": 58
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 0.5782092809677124,
      "learning_rate": 9.997482711915927e-05,
      "loss": 0.7497,
      "step": 60
    },
    {
      "epoch": 5.636363636363637,
      "grad_norm": 0.6757051944732666,
      "learning_rate": 9.99506651282272e-05,
      "loss": 0.843,
      "step": 62
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 0.71096271276474,
      "learning_rate": 9.991845519630678e-05,
      "loss": 0.7576,
      "step": 64
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.6752740144729614,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.7894,
      "step": 66
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.927240788936615,
      "eval_runtime": 2.1578,
      "eval_samples_per_second": 11.122,
      "eval_steps_per_second": 11.122,
      "step": 66
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 0.6642020344734192,
      "learning_rate": 9.982991356370404e-05,
      "loss": 0.6761,
      "step": 68
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 0.6479915380477905,
      "learning_rate": 9.977359612865423e-05,
      "loss": 0.7253,
      "step": 70
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 0.732330858707428,
      "learning_rate": 9.970925928158274e-05,
      "loss": 0.7279,
      "step": 72
    },
    {
      "epoch": 6.7272727272727275,
      "grad_norm": 0.7875674962997437,
      "learning_rate": 9.963691338830044e-05,
      "loss": 0.6863,
      "step": 74
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 0.9151310920715332,
      "learning_rate": 9.955657010501806e-05,
      "loss": 0.6386,
      "step": 76
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.9589347243309021,
      "eval_runtime": 2.1355,
      "eval_samples_per_second": 11.239,
      "eval_steps_per_second": 11.239,
      "step": 77
    },
    {
      "epoch": 7.090909090909091,
      "grad_norm": 0.8763688206672668,
      "learning_rate": 9.946824237646824e-05,
      "loss": 0.5939,
      "step": 78
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 0.8729435801506042,
      "learning_rate": 9.937194443381972e-05,
      "loss": 0.6269,
      "step": 80
    },
    {
      "epoch": 7.454545454545454,
      "grad_norm": 0.9867181777954102,
      "learning_rate": 9.926769179238466e-05,
      "loss": 0.5187,
      "step": 82
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 0.9073792099952698,
      "learning_rate": 9.915550124911866e-05,
      "loss": 0.5338,
      "step": 84
    },
    {
      "epoch": 7.818181818181818,
      "grad_norm": 1.1112242937088013,
      "learning_rate": 9.903539087991462e-05,
      "loss": 0.6073,
      "step": 86
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.1530464887619019,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.5299,
      "step": 88
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.014830470085144,
      "eval_runtime": 2.1202,
      "eval_samples_per_second": 11.319,
      "eval_steps_per_second": 11.319,
      "step": 88
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 1.020405888557434,
      "learning_rate": 9.877148934427037e-05,
      "loss": 0.4534,
      "step": 90
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 1.3011430501937866,
      "learning_rate": 9.862774069706346e-05,
      "loss": 0.4053,
      "step": 92
    },
    {
      "epoch": 8.545454545454545,
      "grad_norm": 1.2572883367538452,
      "learning_rate": 9.847615725553456e-05,
      "loss": 0.4027,
      "step": 94
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 1.2410905361175537,
      "learning_rate": 9.831676344247342e-05,
      "loss": 0.4549,
      "step": 96
    },
    {
      "epoch": 8.909090909090908,
      "grad_norm": 1.4010752439498901,
      "learning_rate": 9.814958493905963e-05,
      "loss": 0.4425,
      "step": 98
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.1573680639266968,
      "eval_runtime": 2.1992,
      "eval_samples_per_second": 10.913,
      "eval_steps_per_second": 10.913,
      "step": 99
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 1.2350188493728638,
      "learning_rate": 9.797464868072488e-05,
      "loss": 0.3804,
      "step": 100
    },
    {
      "epoch": 9.272727272727273,
      "grad_norm": 1.1127281188964844,
      "learning_rate": 9.779198285281325e-05,
      "loss": 0.2995,
      "step": 102
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 1.459039568901062,
      "learning_rate": 9.760161688604008e-05,
      "loss": 0.3257,
      "step": 104
    },
    {
      "epoch": 9.636363636363637,
      "grad_norm": 1.4079865217208862,
      "learning_rate": 9.740358145174998e-05,
      "loss": 0.2908,
      "step": 106
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 1.5733085870742798,
      "learning_rate": 9.719790845697533e-05,
      "loss": 0.2923,
      "step": 108
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.5562692880630493,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.306,
      "step": 110
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.396474838256836,
      "eval_runtime": 2.1275,
      "eval_samples_per_second": 11.281,
      "eval_steps_per_second": 11.281,
      "step": 110
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 1.1821430921554565,
      "learning_rate": 9.676378356149734e-05,
      "loss": 0.2324,
      "step": 112
    },
    {
      "epoch": 10.363636363636363,
      "grad_norm": 1.3496441841125488,
      "learning_rate": 9.653540160603956e-05,
      "loss": 0.2395,
      "step": 114
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 1.2889825105667114,
      "learning_rate": 9.629952196931901e-05,
      "loss": 0.1997,
      "step": 116
    },
    {
      "epoch": 10.727272727272727,
      "grad_norm": 1.8057669401168823,
      "learning_rate": 9.60561826557425e-05,
      "loss": 0.2273,
      "step": 118
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 1.588689923286438,
      "learning_rate": 9.580542287160348e-05,
      "loss": 0.2155,
      "step": 120
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.5442113876342773,
      "eval_runtime": 2.1219,
      "eval_samples_per_second": 11.31,
      "eval_steps_per_second": 11.31,
      "step": 121
    },
    {
      "epoch": 11.090909090909092,
      "grad_norm": 1.0042868852615356,
      "learning_rate": 9.554728301876526e-05,
      "loss": 0.1652,
      "step": 122
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 1.324501633644104,
      "learning_rate": 9.528180468815155e-05,
      "loss": 0.1557,
      "step": 124
    },
    {
      "epoch": 11.454545454545455,
      "grad_norm": 1.3630895614624023,
      "learning_rate": 9.50090306530454e-05,
      "loss": 0.1592,
      "step": 126
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 1.2809746265411377,
      "learning_rate": 9.472900486219769e-05,
      "loss": 0.1426,
      "step": 128
    },
    {
      "epoch": 11.818181818181818,
      "grad_norm": 1.5500458478927612,
      "learning_rate": 9.444177243274618e-05,
      "loss": 0.1525,
      "step": 130
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.8316224813461304,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.1928,
      "step": 132
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.7250471115112305,
      "eval_runtime": 2.1646,
      "eval_samples_per_second": 11.087,
      "eval_steps_per_second": 11.087,
      "step": 132
    },
    {
      "epoch": 12.181818181818182,
      "grad_norm": 0.9303521513938904,
      "learning_rate": 9.384587392471515e-05,
      "loss": 0.1101,
      "step": 134
    },
    {
      "epoch": 12.363636363636363,
      "grad_norm": 1.1953262090682983,
      "learning_rate": 9.353730385598887e-05,
      "loss": 0.1177,
      "step": 136
    },
    {
      "epoch": 12.545454545454545,
      "grad_norm": 1.2532438039779663,
      "learning_rate": 9.322171915289635e-05,
      "loss": 0.1201,
      "step": 138
    },
    {
      "epoch": 12.727272727272727,
      "grad_norm": 1.1903181076049805,
      "learning_rate": 9.289917066174886e-05,
      "loss": 0.1252,
      "step": 140
    },
    {
      "epoch": 12.909090909090908,
      "grad_norm": 1.360329508781433,
      "learning_rate": 9.256971035084785e-05,
      "loss": 0.1191,
      "step": 142
    },
    {
      "epoch": 13.0,
      "eval_loss": 1.8247795104980469,
      "eval_runtime": 2.1576,
      "eval_samples_per_second": 11.124,
      "eval_steps_per_second": 11.124,
      "step": 143
    },
    {
      "epoch": 13.0,
      "step": 143,
      "total_flos": 4963046713720832.0,
      "train_loss": 0.8354059737045448,
      "train_runtime": 502.0783,
      "train_samples_per_second": 8.764,
      "train_steps_per_second": 1.095
    }
  ],
  "logging_steps": 2,
  "max_steps": 550,
  "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": 4963046713720832.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}