{
  "best_metric": 1.3272449970245361,
  "best_model_checkpoint": "data/Gemma-2-2B_task-1_120-samples_config-1/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": 2.7654104232788086,
      "learning_rate": 1.818181818181818e-06,
      "loss": 2.4749,
      "step": 1
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 2.517627000808716,
      "learning_rate": 3.636363636363636e-06,
      "loss": 2.5354,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 2.8291656970977783,
      "learning_rate": 7.272727272727272e-06,
      "loss": 2.4329,
      "step": 4
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 2.6653389930725098,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 2.3833,
      "step": 6
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 2.7465708255767822,
      "learning_rate": 1.4545454545454545e-05,
      "loss": 2.3999,
      "step": 8
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 2.671473503112793,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 2.311,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.237502098083496,
      "eval_runtime": 2.2179,
      "eval_samples_per_second": 10.821,
      "eval_steps_per_second": 10.821,
      "step": 11
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 2.5977914333343506,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 2.2169,
      "step": 12
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 2.3312153816223145,
      "learning_rate": 2.5454545454545454e-05,
      "loss": 2.1782,
      "step": 14
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 2.1294496059417725,
      "learning_rate": 2.909090909090909e-05,
      "loss": 2.0556,
      "step": 16
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 2.0936222076416016,
      "learning_rate": 3.272727272727273e-05,
      "loss": 1.8615,
      "step": 18
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 1.9292643070220947,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 1.7459,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.2698487043380737,
      "learning_rate": 4e-05,
      "loss": 1.6775,
      "step": 22
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.6417160034179688,
      "eval_runtime": 2.1987,
      "eval_samples_per_second": 10.916,
      "eval_steps_per_second": 10.916,
      "step": 22
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 1.4562098979949951,
      "learning_rate": 4.3636363636363636e-05,
      "loss": 1.6336,
      "step": 24
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 1.1218253374099731,
      "learning_rate": 4.7272727272727275e-05,
      "loss": 1.571,
      "step": 26
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 1.04109525680542,
      "learning_rate": 5.090909090909091e-05,
      "loss": 1.4032,
      "step": 28
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 1.1949294805526733,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 1.3756,
      "step": 30
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 1.2722885608673096,
      "learning_rate": 5.818181818181818e-05,
      "loss": 1.4086,
      "step": 32
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.4307798147201538,
      "eval_runtime": 2.2389,
      "eval_samples_per_second": 10.72,
      "eval_steps_per_second": 10.72,
      "step": 33
    },
    {
      "epoch": 3.090909090909091,
      "grad_norm": 1.1219353675842285,
      "learning_rate": 6.181818181818182e-05,
      "loss": 1.4083,
      "step": 34
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 0.9904714226722717,
      "learning_rate": 6.545454545454546e-05,
      "loss": 1.3082,
      "step": 36
    },
    {
      "epoch": 3.4545454545454546,
      "grad_norm": 1.2756577730178833,
      "learning_rate": 6.90909090909091e-05,
      "loss": 1.3081,
      "step": 38
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 1.146028995513916,
      "learning_rate": 7.272727272727273e-05,
      "loss": 1.2388,
      "step": 40
    },
    {
      "epoch": 3.8181818181818183,
      "grad_norm": 1.2157866954803467,
      "learning_rate": 7.636363636363637e-05,
      "loss": 1.1917,
      "step": 42
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.1651657819747925,
      "learning_rate": 8e-05,
      "loss": 1.1316,
      "step": 44
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.337244987487793,
      "eval_runtime": 2.1906,
      "eval_samples_per_second": 10.956,
      "eval_steps_per_second": 10.956,
      "step": 44
    },
    {
      "epoch": 4.181818181818182,
      "grad_norm": 1.262305498123169,
      "learning_rate": 8.363636363636364e-05,
      "loss": 1.0936,
      "step": 46
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 1.339064121246338,
      "learning_rate": 8.727272727272727e-05,
      "loss": 1.0332,
      "step": 48
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 1.416332721710205,
      "learning_rate": 9.090909090909092e-05,
      "loss": 1.0611,
      "step": 50
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 1.3856735229492188,
      "learning_rate": 9.454545454545455e-05,
      "loss": 1.035,
      "step": 52
    },
    {
      "epoch": 4.909090909090909,
      "grad_norm": 1.427396297454834,
      "learning_rate": 9.818181818181818e-05,
      "loss": 0.9745,
      "step": 54
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.3272449970245361,
      "eval_runtime": 2.2368,
      "eval_samples_per_second": 10.73,
      "eval_steps_per_second": 10.73,
      "step": 55
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 1.4408196210861206,
      "learning_rate": 9.999899300364532e-05,
      "loss": 0.9447,
      "step": 56
    },
    {
      "epoch": 5.2727272727272725,
      "grad_norm": 1.4390223026275635,
      "learning_rate": 9.99909372761763e-05,
      "loss": 0.7821,
      "step": 58
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 1.4396119117736816,
      "learning_rate": 9.997482711915927e-05,
      "loss": 0.7663,
      "step": 60
    },
    {
      "epoch": 5.636363636363637,
      "grad_norm": 1.6289987564086914,
      "learning_rate": 9.99506651282272e-05,
      "loss": 0.8362,
      "step": 62
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 1.8272826671600342,
      "learning_rate": 9.991845519630678e-05,
      "loss": 0.7933,
      "step": 64
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.8998042345046997,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.7135,
      "step": 66
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.3980852365493774,
      "eval_runtime": 2.1963,
      "eval_samples_per_second": 10.927,
      "eval_steps_per_second": 10.927,
      "step": 66
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 1.608075499534607,
      "learning_rate": 9.982991356370404e-05,
      "loss": 0.6363,
      "step": 68
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 1.8586798906326294,
      "learning_rate": 9.977359612865423e-05,
      "loss": 0.5902,
      "step": 70
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 1.8311554193496704,
      "learning_rate": 9.970925928158274e-05,
      "loss": 0.4982,
      "step": 72
    },
    {
      "epoch": 6.7272727272727275,
      "grad_norm": 2.392566442489624,
      "learning_rate": 9.963691338830044e-05,
      "loss": 0.548,
      "step": 74
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 2.681428909301758,
      "learning_rate": 9.955657010501806e-05,
      "loss": 0.4228,
      "step": 76
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.7707710266113281,
      "eval_runtime": 2.1916,
      "eval_samples_per_second": 10.951,
      "eval_steps_per_second": 10.951,
      "step": 77
    },
    {
      "epoch": 7.090909090909091,
      "grad_norm": 1.9553114175796509,
      "learning_rate": 9.946824237646824e-05,
      "loss": 0.3491,
      "step": 78
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 2.017584800720215,
      "learning_rate": 9.937194443381972e-05,
      "loss": 0.3254,
      "step": 80
    },
    {
      "epoch": 7.454545454545454,
      "grad_norm": 2.6024396419525146,
      "learning_rate": 9.926769179238466e-05,
      "loss": 0.2919,
      "step": 82
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 2.637301445007324,
      "learning_rate": 9.915550124911866e-05,
      "loss": 0.2924,
      "step": 84
    },
    {
      "epoch": 7.818181818181818,
      "grad_norm": 2.9900031089782715,
      "learning_rate": 9.903539087991462e-05,
      "loss": 0.3155,
      "step": 86
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.341743230819702,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.3143,
      "step": 88
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.152059316635132,
      "eval_runtime": 2.1986,
      "eval_samples_per_second": 10.916,
      "eval_steps_per_second": 10.916,
      "step": 88
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 1.73684561252594,
      "learning_rate": 9.877148934427037e-05,
      "loss": 0.1885,
      "step": 90
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 2.3624324798583984,
      "learning_rate": 9.862774069706346e-05,
      "loss": 0.159,
      "step": 92
    },
    {
      "epoch": 8.545454545454545,
      "grad_norm": 2.342822551727295,
      "learning_rate": 9.847615725553456e-05,
      "loss": 0.123,
      "step": 94
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 2.090047836303711,
      "learning_rate": 9.831676344247342e-05,
      "loss": 0.1231,
      "step": 96
    },
    {
      "epoch": 8.909090909090908,
      "grad_norm": 3.0720016956329346,
      "learning_rate": 9.814958493905963e-05,
      "loss": 0.1454,
      "step": 98
    },
    {
      "epoch": 9.0,
      "eval_loss": 2.689652442932129,
      "eval_runtime": 2.1925,
      "eval_samples_per_second": 10.947,
      "eval_steps_per_second": 10.947,
      "step": 99
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 2.5275495052337646,
      "learning_rate": 9.797464868072488e-05,
      "loss": 0.11,
      "step": 100
    },
    {
      "epoch": 9.272727272727273,
      "grad_norm": 2.512793779373169,
      "learning_rate": 9.779198285281325e-05,
      "loss": 0.0796,
      "step": 102
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 3.505725383758545,
      "learning_rate": 9.760161688604008e-05,
      "loss": 0.0786,
      "step": 104
    },
    {
      "epoch": 9.636363636363637,
      "grad_norm": 3.8895909786224365,
      "learning_rate": 9.740358145174998e-05,
      "loss": 0.0944,
      "step": 106
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 2.5016000270843506,
      "learning_rate": 9.719790845697533e-05,
      "loss": 0.0798,
      "step": 108
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.5594393014907837,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0857,
      "step": 110
    },
    {
      "epoch": 10.0,
      "eval_loss": 2.6891584396362305,
      "eval_runtime": 2.2175,
      "eval_samples_per_second": 10.823,
      "eval_steps_per_second": 10.823,
      "step": 110
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 1.6706209182739258,
      "learning_rate": 9.676378356149734e-05,
      "loss": 0.0373,
      "step": 112
    },
    {
      "epoch": 10.363636363636363,
      "grad_norm": 1.8536970615386963,
      "learning_rate": 9.653540160603956e-05,
      "loss": 0.0545,
      "step": 114
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 2.2033114433288574,
      "learning_rate": 9.629952196931901e-05,
      "loss": 0.0719,
      "step": 116
    },
    {
      "epoch": 10.727272727272727,
      "grad_norm": 2.138977527618408,
      "learning_rate": 9.60561826557425e-05,
      "loss": 0.0587,
      "step": 118
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 3.207634210586548,
      "learning_rate": 9.580542287160348e-05,
      "loss": 0.06,
      "step": 120
    },
    {
      "epoch": 11.0,
      "eval_loss": 2.8942203521728516,
      "eval_runtime": 2.2274,
      "eval_samples_per_second": 10.775,
      "eval_steps_per_second": 10.775,
      "step": 121
    },
    {
      "epoch": 11.090909090909092,
      "grad_norm": 2.0782763957977295,
      "learning_rate": 9.554728301876526e-05,
      "loss": 0.0442,
      "step": 122
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 1.5070956945419312,
      "learning_rate": 9.528180468815155e-05,
      "loss": 0.0264,
      "step": 124
    },
    {
      "epoch": 11.454545454545455,
      "grad_norm": 2.6724226474761963,
      "learning_rate": 9.50090306530454e-05,
      "loss": 0.0485,
      "step": 126
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 2.257985830307007,
      "learning_rate": 9.472900486219769e-05,
      "loss": 0.0457,
      "step": 128
    },
    {
      "epoch": 11.818181818181818,
      "grad_norm": 2.384425163269043,
      "learning_rate": 9.444177243274618e-05,
      "loss": 0.0572,
      "step": 130
    },
    {
      "epoch": 12.0,
      "grad_norm": 3.0404624938964844,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0671,
      "step": 132
    },
    {
      "epoch": 12.0,
      "eval_loss": 2.884262800216675,
      "eval_runtime": 2.1852,
      "eval_samples_per_second": 10.983,
      "eval_steps_per_second": 10.983,
      "step": 132
    },
    {
      "epoch": 12.0,
      "step": 132,
      "total_flos": 4216951435952128.0,
      "train_loss": 0.836406874667966,
      "train_runtime": 459.9231,
      "train_samples_per_second": 9.567,
      "train_steps_per_second": 1.196
    }
  ],
  "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": 4216951435952128.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}