{
  "best_metric": 0.9430673122406006,
  "best_model_checkpoint": "data/Gemma-2-2B_task-3_120-samples_config-1_full/checkpoint-77",
  "epoch": 14.0,
  "eval_steps": 500,
  "global_step": 154,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09090909090909091,
      "grad_norm": 0.4401099681854248,
      "learning_rate": 1.818181818181818e-06,
      "loss": 1.2873,
      "step": 1
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 0.555508553981781,
      "learning_rate": 3.636363636363636e-06,
      "loss": 1.5506,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 0.42541971802711487,
      "learning_rate": 7.272727272727272e-06,
      "loss": 1.3483,
      "step": 4
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 0.4697648286819458,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 1.3864,
      "step": 6
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 0.4494505822658539,
      "learning_rate": 1.4545454545454545e-05,
      "loss": 1.4567,
      "step": 8
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.4147140383720398,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 1.3447,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.366329550743103,
      "eval_runtime": 7.6918,
      "eval_samples_per_second": 3.12,
      "eval_steps_per_second": 3.12,
      "step": 11
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 0.4472775161266327,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 1.4277,
      "step": 12
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 0.37309274077415466,
      "learning_rate": 2.5454545454545454e-05,
      "loss": 1.3125,
      "step": 14
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 0.436692476272583,
      "learning_rate": 2.909090909090909e-05,
      "loss": 1.3597,
      "step": 16
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 0.36430731415748596,
      "learning_rate": 3.272727272727273e-05,
      "loss": 1.3537,
      "step": 18
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.29779544472694397,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 1.2595,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.262285977602005,
      "learning_rate": 4e-05,
      "loss": 1.1652,
      "step": 22
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.2193524837493896,
      "eval_runtime": 7.4578,
      "eval_samples_per_second": 3.218,
      "eval_steps_per_second": 3.218,
      "step": 22
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 0.2341645061969757,
      "learning_rate": 4.3636363636363636e-05,
      "loss": 1.1997,
      "step": 24
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 0.2721911668777466,
      "learning_rate": 4.7272727272727275e-05,
      "loss": 1.1882,
      "step": 26
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.25244641304016113,
      "learning_rate": 5.090909090909091e-05,
      "loss": 1.1256,
      "step": 28
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.2625352740287781,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 1.1495,
      "step": 30
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.26791104674339294,
      "learning_rate": 5.818181818181818e-05,
      "loss": 1.1036,
      "step": 32
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.0959328413009644,
      "eval_runtime": 7.4582,
      "eval_samples_per_second": 3.218,
      "eval_steps_per_second": 3.218,
      "step": 33
    },
    {
      "epoch": 3.090909090909091,
      "grad_norm": 0.26914045214653015,
      "learning_rate": 6.181818181818182e-05,
      "loss": 1.112,
      "step": 34
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 0.29382359981536865,
      "learning_rate": 6.545454545454546e-05,
      "loss": 1.0432,
      "step": 36
    },
    {
      "epoch": 3.4545454545454546,
      "grad_norm": 0.32928666472435,
      "learning_rate": 6.90909090909091e-05,
      "loss": 1.0381,
      "step": 38
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.24698656797409058,
      "learning_rate": 7.272727272727273e-05,
      "loss": 0.9541,
      "step": 40
    },
    {
      "epoch": 3.8181818181818183,
      "grad_norm": 0.31348344683647156,
      "learning_rate": 7.636363636363637e-05,
      "loss": 0.93,
      "step": 42
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.24753938615322113,
      "learning_rate": 8e-05,
      "loss": 1.021,
      "step": 44
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.9976680874824524,
      "eval_runtime": 7.472,
      "eval_samples_per_second": 3.212,
      "eval_steps_per_second": 3.212,
      "step": 44
    },
    {
      "epoch": 4.181818181818182,
      "grad_norm": 0.24109996855258942,
      "learning_rate": 8.363636363636364e-05,
      "loss": 0.9201,
      "step": 46
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 0.25494322180747986,
      "learning_rate": 8.727272727272727e-05,
      "loss": 0.9535,
      "step": 48
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 0.1965021938085556,
      "learning_rate": 9.090909090909092e-05,
      "loss": 0.9022,
      "step": 50
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 0.24913202226161957,
      "learning_rate": 9.454545454545455e-05,
      "loss": 0.9235,
      "step": 52
    },
    {
      "epoch": 4.909090909090909,
      "grad_norm": 0.18288737535476685,
      "learning_rate": 9.818181818181818e-05,
      "loss": 0.9434,
      "step": 54
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.9632444977760315,
      "eval_runtime": 7.4608,
      "eval_samples_per_second": 3.217,
      "eval_steps_per_second": 3.217,
      "step": 55
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 0.19076839089393616,
      "learning_rate": 9.999899300364532e-05,
      "loss": 0.8894,
      "step": 56
    },
    {
      "epoch": 5.2727272727272725,
      "grad_norm": 0.20267784595489502,
      "learning_rate": 9.99909372761763e-05,
      "loss": 0.9071,
      "step": 58
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 0.2087169587612152,
      "learning_rate": 9.997482711915927e-05,
      "loss": 0.8532,
      "step": 60
    },
    {
      "epoch": 5.636363636363637,
      "grad_norm": 0.3292751610279083,
      "learning_rate": 9.99506651282272e-05,
      "loss": 0.8854,
      "step": 62
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 0.21600216627120972,
      "learning_rate": 9.991845519630678e-05,
      "loss": 0.8516,
      "step": 64
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.21119904518127441,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.8982,
      "step": 66
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.9466676115989685,
      "eval_runtime": 7.4625,
      "eval_samples_per_second": 3.216,
      "eval_steps_per_second": 3.216,
      "step": 66
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 0.2550293505191803,
      "learning_rate": 9.982991356370404e-05,
      "loss": 0.8157,
      "step": 68
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 0.23336729407310486,
      "learning_rate": 9.977359612865423e-05,
      "loss": 0.8606,
      "step": 70
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 0.2545364201068878,
      "learning_rate": 9.970925928158274e-05,
      "loss": 0.7977,
      "step": 72
    },
    {
      "epoch": 6.7272727272727275,
      "grad_norm": 0.259082168340683,
      "learning_rate": 9.963691338830044e-05,
      "loss": 0.8332,
      "step": 74
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 0.2226342260837555,
      "learning_rate": 9.955657010501806e-05,
      "loss": 0.8566,
      "step": 76
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.9430673122406006,
      "eval_runtime": 7.464,
      "eval_samples_per_second": 3.215,
      "eval_steps_per_second": 3.215,
      "step": 77
    },
    {
      "epoch": 7.090909090909091,
      "grad_norm": 0.2310396134853363,
      "learning_rate": 9.946824237646824e-05,
      "loss": 0.8999,
      "step": 78
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 0.3359197676181793,
      "learning_rate": 9.937194443381972e-05,
      "loss": 0.7571,
      "step": 80
    },
    {
      "epoch": 7.454545454545454,
      "grad_norm": 0.27689051628112793,
      "learning_rate": 9.926769179238466e-05,
      "loss": 0.754,
      "step": 82
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 0.29372766613960266,
      "learning_rate": 9.915550124911866e-05,
      "loss": 0.7961,
      "step": 84
    },
    {
      "epoch": 7.818181818181818,
      "grad_norm": 0.3483748733997345,
      "learning_rate": 9.903539087991462e-05,
      "loss": 0.788,
      "step": 86
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.2859647870063782,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.8274,
      "step": 88
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.9471516013145447,
      "eval_runtime": 7.4534,
      "eval_samples_per_second": 3.22,
      "eval_steps_per_second": 3.22,
      "step": 88
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 0.31532877683639526,
      "learning_rate": 9.877148934427037e-05,
      "loss": 0.7785,
      "step": 90
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 0.3497365117073059,
      "learning_rate": 9.862774069706346e-05,
      "loss": 0.7253,
      "step": 92
    },
    {
      "epoch": 8.545454545454545,
      "grad_norm": 0.3729742765426636,
      "learning_rate": 9.847615725553456e-05,
      "loss": 0.759,
      "step": 94
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 0.4153083860874176,
      "learning_rate": 9.831676344247342e-05,
      "loss": 0.6846,
      "step": 96
    },
    {
      "epoch": 8.909090909090908,
      "grad_norm": 0.42387697100639343,
      "learning_rate": 9.814958493905963e-05,
      "loss": 0.75,
      "step": 98
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.9634823799133301,
      "eval_runtime": 7.4619,
      "eval_samples_per_second": 3.216,
      "eval_steps_per_second": 3.216,
      "step": 99
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 0.4489431381225586,
      "learning_rate": 9.797464868072488e-05,
      "loss": 0.7566,
      "step": 100
    },
    {
      "epoch": 9.272727272727273,
      "grad_norm": 0.4480694830417633,
      "learning_rate": 9.779198285281325e-05,
      "loss": 0.6584,
      "step": 102
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 0.7039919495582581,
      "learning_rate": 9.760161688604008e-05,
      "loss": 0.6571,
      "step": 104
    },
    {
      "epoch": 9.636363636363637,
      "grad_norm": 0.5547890663146973,
      "learning_rate": 9.740358145174998e-05,
      "loss": 0.737,
      "step": 106
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 0.5048365592956543,
      "learning_rate": 9.719790845697533e-05,
      "loss": 0.6767,
      "step": 108
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.6568930149078369,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.6436,
      "step": 110
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.0182775259017944,
      "eval_runtime": 7.4594,
      "eval_samples_per_second": 3.217,
      "eval_steps_per_second": 3.217,
      "step": 110
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 0.5760776996612549,
      "learning_rate": 9.676378356149734e-05,
      "loss": 0.5993,
      "step": 112
    },
    {
      "epoch": 10.363636363636363,
      "grad_norm": 0.6874850392341614,
      "learning_rate": 9.653540160603956e-05,
      "loss": 0.5933,
      "step": 114
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 0.6967574954032898,
      "learning_rate": 9.629952196931901e-05,
      "loss": 0.5981,
      "step": 116
    },
    {
      "epoch": 10.727272727272727,
      "grad_norm": 0.7388347387313843,
      "learning_rate": 9.60561826557425e-05,
      "loss": 0.59,
      "step": 118
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 0.6717241406440735,
      "learning_rate": 9.580542287160348e-05,
      "loss": 0.6415,
      "step": 120
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.060663104057312,
      "eval_runtime": 7.4556,
      "eval_samples_per_second": 3.219,
      "eval_steps_per_second": 3.219,
      "step": 121
    },
    {
      "epoch": 11.090909090909092,
      "grad_norm": 0.8145678043365479,
      "learning_rate": 9.554728301876526e-05,
      "loss": 0.5758,
      "step": 122
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 1.2217453718185425,
      "learning_rate": 9.528180468815155e-05,
      "loss": 0.5195,
      "step": 124
    },
    {
      "epoch": 11.454545454545455,
      "grad_norm": 0.7760822176933289,
      "learning_rate": 9.50090306530454e-05,
      "loss": 0.5365,
      "step": 126
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 0.7117080092430115,
      "learning_rate": 9.472900486219769e-05,
      "loss": 0.5693,
      "step": 128
    },
    {
      "epoch": 11.818181818181818,
      "grad_norm": 0.9114668965339661,
      "learning_rate": 9.444177243274618e-05,
      "loss": 0.539,
      "step": 130
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.9518639445304871,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.5077,
      "step": 132
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.1553831100463867,
      "eval_runtime": 7.4567,
      "eval_samples_per_second": 3.219,
      "eval_steps_per_second": 3.219,
      "step": 132
    },
    {
      "epoch": 12.181818181818182,
      "grad_norm": 0.995181143283844,
      "learning_rate": 9.384587392471515e-05,
      "loss": 0.4883,
      "step": 134
    },
    {
      "epoch": 12.363636363636363,
      "grad_norm": 1.2349698543548584,
      "learning_rate": 9.353730385598887e-05,
      "loss": 0.4358,
      "step": 136
    },
    {
      "epoch": 12.545454545454545,
      "grad_norm": 1.0227618217468262,
      "learning_rate": 9.322171915289635e-05,
      "loss": 0.4565,
      "step": 138
    },
    {
      "epoch": 12.727272727272727,
      "grad_norm": 0.9845717549324036,
      "learning_rate": 9.289917066174886e-05,
      "loss": 0.445,
      "step": 140
    },
    {
      "epoch": 12.909090909090908,
      "grad_norm": 1.2052007913589478,
      "learning_rate": 9.256971035084785e-05,
      "loss": 0.5122,
      "step": 142
    },
    {
      "epoch": 13.0,
      "eval_loss": 1.2326768636703491,
      "eval_runtime": 7.4634,
      "eval_samples_per_second": 3.216,
      "eval_steps_per_second": 3.216,
      "step": 143
    },
    {
      "epoch": 13.090909090909092,
      "grad_norm": 0.9715532064437866,
      "learning_rate": 9.223339130211192e-05,
      "loss": 0.362,
      "step": 144
    },
    {
      "epoch": 13.272727272727273,
      "grad_norm": 1.35349702835083,
      "learning_rate": 9.189026770252436e-05,
      "loss": 0.4128,
      "step": 146
    },
    {
      "epoch": 13.454545454545455,
      "grad_norm": 1.1257082223892212,
      "learning_rate": 9.154039483540273e-05,
      "loss": 0.4465,
      "step": 148
    },
    {
      "epoch": 13.636363636363637,
      "grad_norm": 1.064841389656067,
      "learning_rate": 9.118382907149165e-05,
      "loss": 0.3993,
      "step": 150
    },
    {
      "epoch": 13.818181818181818,
      "grad_norm": 1.4757120609283447,
      "learning_rate": 9.082062785988049e-05,
      "loss": 0.3703,
      "step": 152
    },
    {
      "epoch": 14.0,
      "grad_norm": 1.2708607912063599,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.3314,
      "step": 154
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.3615187406539917,
      "eval_runtime": 7.4601,
      "eval_samples_per_second": 3.217,
      "eval_steps_per_second": 3.217,
      "step": 154
    },
    {
      "epoch": 14.0,
      "step": 154,
      "total_flos": 3.310644127544115e+16,
      "train_loss": 0.8359614691951058,
      "train_runtime": 1404.4143,
      "train_samples_per_second": 3.133,
      "train_steps_per_second": 0.392
    }
  ],
  "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": 3.310644127544115e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}