{
  "best_metric": 0.9736120104789734,
  "best_model_checkpoint": "data/Gemma-2-2B_task-1_120-samples_config-1_auto/checkpoint-55",
  "epoch": 12.0,
  "eval_steps": 500,
  "global_step": 132,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09090909090909091,
      "grad_norm": 1.896773099899292,
      "learning_rate": 1.818181818181818e-06,
      "loss": 1.6148,
      "step": 1
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 1.7315363883972168,
      "learning_rate": 3.636363636363636e-06,
      "loss": 1.5745,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 1.9249202013015747,
      "learning_rate": 7.272727272727272e-06,
      "loss": 1.6592,
      "step": 4
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 1.9421212673187256,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 1.6111,
      "step": 6
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 1.8695214986801147,
      "learning_rate": 1.4545454545454545e-05,
      "loss": 1.5167,
      "step": 8
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 2.2962398529052734,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 1.7508,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.5551661252975464,
      "eval_runtime": 2.1687,
      "eval_samples_per_second": 11.067,
      "eval_steps_per_second": 11.067,
      "step": 11
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 1.9042454957962036,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 1.543,
      "step": 12
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 1.5851277112960815,
      "learning_rate": 2.5454545454545454e-05,
      "loss": 1.3843,
      "step": 14
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 1.3642412424087524,
      "learning_rate": 2.909090909090909e-05,
      "loss": 1.279,
      "step": 16
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 1.492651343345642,
      "learning_rate": 3.272727272727273e-05,
      "loss": 1.3709,
      "step": 18
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 1.7590340375900269,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 1.2589,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9888437390327454,
      "learning_rate": 4e-05,
      "loss": 1.066,
      "step": 22
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.1328343152999878,
      "eval_runtime": 2.1687,
      "eval_samples_per_second": 11.066,
      "eval_steps_per_second": 11.066,
      "step": 22
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 1.20571768283844,
      "learning_rate": 4.3636363636363636e-05,
      "loss": 1.0086,
      "step": 24
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 0.9041438698768616,
      "learning_rate": 4.7272727272727275e-05,
      "loss": 1.0831,
      "step": 26
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.9103597402572632,
      "learning_rate": 5.090909090909091e-05,
      "loss": 1.071,
      "step": 28
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.870693564414978,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 1.0293,
      "step": 30
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.8961207866668701,
      "learning_rate": 5.818181818181818e-05,
      "loss": 0.9348,
      "step": 32
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.0277460813522339,
      "eval_runtime": 2.1645,
      "eval_samples_per_second": 11.088,
      "eval_steps_per_second": 11.088,
      "step": 33
    },
    {
      "epoch": 3.090909090909091,
      "grad_norm": 0.8866666555404663,
      "learning_rate": 6.181818181818182e-05,
      "loss": 0.9195,
      "step": 34
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 0.8282307982444763,
      "learning_rate": 6.545454545454546e-05,
      "loss": 0.8315,
      "step": 36
    },
    {
      "epoch": 3.4545454545454546,
      "grad_norm": 0.9073108434677124,
      "learning_rate": 6.90909090909091e-05,
      "loss": 0.9055,
      "step": 38
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.8472995758056641,
      "learning_rate": 7.272727272727273e-05,
      "loss": 0.8648,
      "step": 40
    },
    {
      "epoch": 3.8181818181818183,
      "grad_norm": 0.9500349164009094,
      "learning_rate": 7.636363636363637e-05,
      "loss": 0.8957,
      "step": 42
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.039790153503418,
      "learning_rate": 8e-05,
      "loss": 0.8176,
      "step": 44
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.9743786454200745,
      "eval_runtime": 2.1442,
      "eval_samples_per_second": 11.193,
      "eval_steps_per_second": 11.193,
      "step": 44
    },
    {
      "epoch": 4.181818181818182,
      "grad_norm": 1.096146821975708,
      "learning_rate": 8.363636363636364e-05,
      "loss": 0.7854,
      "step": 46
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 1.0144972801208496,
      "learning_rate": 8.727272727272727e-05,
      "loss": 0.795,
      "step": 48
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 1.1709052324295044,
      "learning_rate": 9.090909090909092e-05,
      "loss": 0.7459,
      "step": 50
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 1.1173423528671265,
      "learning_rate": 9.454545454545455e-05,
      "loss": 0.7177,
      "step": 52
    },
    {
      "epoch": 4.909090909090909,
      "grad_norm": 1.1534667015075684,
      "learning_rate": 9.818181818181818e-05,
      "loss": 0.6379,
      "step": 54
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.9736120104789734,
      "eval_runtime": 2.173,
      "eval_samples_per_second": 11.045,
      "eval_steps_per_second": 11.045,
      "step": 55
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 1.1661216020584106,
      "learning_rate": 9.999899300364532e-05,
      "loss": 0.6415,
      "step": 56
    },
    {
      "epoch": 5.2727272727272725,
      "grad_norm": 1.0937227010726929,
      "learning_rate": 9.99909372761763e-05,
      "loss": 0.5881,
      "step": 58
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 1.1940053701400757,
      "learning_rate": 9.997482711915927e-05,
      "loss": 0.5414,
      "step": 60
    },
    {
      "epoch": 5.636363636363637,
      "grad_norm": 1.3874033689498901,
      "learning_rate": 9.99506651282272e-05,
      "loss": 0.5359,
      "step": 62
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 1.7549492120742798,
      "learning_rate": 9.991845519630678e-05,
      "loss": 0.5691,
      "step": 64
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.616335153579712,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.497,
      "step": 66
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.0944799184799194,
      "eval_runtime": 2.1138,
      "eval_samples_per_second": 11.354,
      "eval_steps_per_second": 11.354,
      "step": 66
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 1.2271863222122192,
      "learning_rate": 9.982991356370404e-05,
      "loss": 0.3866,
      "step": 68
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 1.521119475364685,
      "learning_rate": 9.977359612865423e-05,
      "loss": 0.427,
      "step": 70
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 1.6232653856277466,
      "learning_rate": 9.970925928158274e-05,
      "loss": 0.407,
      "step": 72
    },
    {
      "epoch": 6.7272727272727275,
      "grad_norm": 1.5563262701034546,
      "learning_rate": 9.963691338830044e-05,
      "loss": 0.2937,
      "step": 74
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 1.8535608053207397,
      "learning_rate": 9.955657010501806e-05,
      "loss": 0.3367,
      "step": 76
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.3431849479675293,
      "eval_runtime": 2.2799,
      "eval_samples_per_second": 10.527,
      "eval_steps_per_second": 10.527,
      "step": 77
    },
    {
      "epoch": 7.090909090909091,
      "grad_norm": 1.7326850891113281,
      "learning_rate": 9.946824237646824e-05,
      "loss": 0.291,
      "step": 78
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 1.5598558187484741,
      "learning_rate": 9.937194443381972e-05,
      "loss": 0.2737,
      "step": 80
    },
    {
      "epoch": 7.454545454545454,
      "grad_norm": 1.730252742767334,
      "learning_rate": 9.926769179238466e-05,
      "loss": 0.1833,
      "step": 82
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 1.9764121770858765,
      "learning_rate": 9.915550124911866e-05,
      "loss": 0.1923,
      "step": 84
    },
    {
      "epoch": 7.818181818181818,
      "grad_norm": 2.5331475734710693,
      "learning_rate": 9.903539087991462e-05,
      "loss": 0.2406,
      "step": 86
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.1329357624053955,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.2147,
      "step": 88
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.6289167404174805,
      "eval_runtime": 2.1403,
      "eval_samples_per_second": 11.213,
      "eval_steps_per_second": 11.213,
      "step": 88
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 1.5905722379684448,
      "learning_rate": 9.877148934427037e-05,
      "loss": 0.1366,
      "step": 90
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 1.624558925628662,
      "learning_rate": 9.862774069706346e-05,
      "loss": 0.111,
      "step": 92
    },
    {
      "epoch": 8.545454545454545,
      "grad_norm": 1.9237786531448364,
      "learning_rate": 9.847615725553456e-05,
      "loss": 0.1213,
      "step": 94
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 1.8648244142532349,
      "learning_rate": 9.831676344247342e-05,
      "loss": 0.1816,
      "step": 96
    },
    {
      "epoch": 8.909090909090908,
      "grad_norm": 3.522146224975586,
      "learning_rate": 9.814958493905963e-05,
      "loss": 0.1552,
      "step": 98
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.8935288190841675,
      "eval_runtime": 2.1658,
      "eval_samples_per_second": 11.082,
      "eval_steps_per_second": 11.082,
      "step": 99
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 1.4460781812667847,
      "learning_rate": 9.797464868072488e-05,
      "loss": 0.0982,
      "step": 100
    },
    {
      "epoch": 9.272727272727273,
      "grad_norm": 1.1425362825393677,
      "learning_rate": 9.779198285281325e-05,
      "loss": 0.0741,
      "step": 102
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 1.5075608491897583,
      "learning_rate": 9.760161688604008e-05,
      "loss": 0.0819,
      "step": 104
    },
    {
      "epoch": 9.636363636363637,
      "grad_norm": 1.9946374893188477,
      "learning_rate": 9.740358145174998e-05,
      "loss": 0.0879,
      "step": 106
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 2.043844223022461,
      "learning_rate": 9.719790845697533e-05,
      "loss": 0.0921,
      "step": 108
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.4709323644638062,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0668,
      "step": 110
    },
    {
      "epoch": 10.0,
      "eval_loss": 2.018200635910034,
      "eval_runtime": 2.1785,
      "eval_samples_per_second": 11.017,
      "eval_steps_per_second": 11.017,
      "step": 110
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 1.0680981874465942,
      "learning_rate": 9.676378356149734e-05,
      "loss": 0.0426,
      "step": 112
    },
    {
      "epoch": 10.363636363636363,
      "grad_norm": 0.9540239572525024,
      "learning_rate": 9.653540160603956e-05,
      "loss": 0.047,
      "step": 114
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 1.3698723316192627,
      "learning_rate": 9.629952196931901e-05,
      "loss": 0.0501,
      "step": 116
    },
    {
      "epoch": 10.727272727272727,
      "grad_norm": 1.75531005859375,
      "learning_rate": 9.60561826557425e-05,
      "loss": 0.0535,
      "step": 118
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 1.7834439277648926,
      "learning_rate": 9.580542287160348e-05,
      "loss": 0.0425,
      "step": 120
    },
    {
      "epoch": 11.0,
      "eval_loss": 2.3609249591827393,
      "eval_runtime": 2.1793,
      "eval_samples_per_second": 11.013,
      "eval_steps_per_second": 11.013,
      "step": 121
    },
    {
      "epoch": 11.090909090909092,
      "grad_norm": 1.7064998149871826,
      "learning_rate": 9.554728301876526e-05,
      "loss": 0.0459,
      "step": 122
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 1.7905004024505615,
      "learning_rate": 9.528180468815155e-05,
      "loss": 0.0385,
      "step": 124
    },
    {
      "epoch": 11.454545454545455,
      "grad_norm": 1.7331182956695557,
      "learning_rate": 9.50090306530454e-05,
      "loss": 0.0615,
      "step": 126
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 1.467869758605957,
      "learning_rate": 9.472900486219769e-05,
      "loss": 0.0444,
      "step": 128
    },
    {
      "epoch": 11.818181818181818,
      "grad_norm": 1.434717059135437,
      "learning_rate": 9.444177243274618e-05,
      "loss": 0.04,
      "step": 130
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.624537467956543,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0507,
      "step": 132
    },
    {
      "epoch": 12.0,
      "eval_loss": 2.3114547729492188,
      "eval_runtime": 2.3476,
      "eval_samples_per_second": 10.223,
      "eval_steps_per_second": 10.223,
      "step": 132
    },
    {
      "epoch": 12.0,
      "step": 132,
      "total_flos": 4581273894912000.0,
      "train_loss": 0.5821334374221888,
      "train_runtime": 471.3358,
      "train_samples_per_second": 9.335,
      "train_steps_per_second": 1.167
    }
  ],
  "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": 4581273894912000.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}