{
  "best_metric": 0.43232032656669617,
  "best_model_checkpoint": "data/Llama-31-8B_task-3_120-samples_config-1/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": 3.9303781986236572,
      "learning_rate": 1.818181818181818e-06,
      "loss": 3.095,
      "step": 1
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 2.5306804180145264,
      "learning_rate": 3.636363636363636e-06,
      "loss": 2.5399,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 2.5905895233154297,
      "learning_rate": 7.272727272727272e-06,
      "loss": 2.5717,
      "step": 4
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 2.0816569328308105,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 2.399,
      "step": 6
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 3.414346218109131,
      "learning_rate": 1.4545454545454545e-05,
      "loss": 2.3702,
      "step": 8
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 2.974047899246216,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 2.1435,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.079517126083374,
      "eval_runtime": 20.998,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 11
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 2.398120403289795,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 2.0242,
      "step": 12
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 2.205050468444824,
      "learning_rate": 2.5454545454545454e-05,
      "loss": 1.8418,
      "step": 14
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 2.050658941268921,
      "learning_rate": 2.909090909090909e-05,
      "loss": 1.7493,
      "step": 16
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 1.9007184505462646,
      "learning_rate": 3.272727272727273e-05,
      "loss": 1.3743,
      "step": 18
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 1.4700392484664917,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 1.1207,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8176637887954712,
      "learning_rate": 4e-05,
      "loss": 0.7297,
      "step": 22
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8105940818786621,
      "eval_runtime": 20.999,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 22
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 0.9762272834777832,
      "learning_rate": 4.3636363636363636e-05,
      "loss": 0.6388,
      "step": 24
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 1.4088635444641113,
      "learning_rate": 4.7272727272727275e-05,
      "loss": 0.7878,
      "step": 26
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.6369071006774902,
      "learning_rate": 5.090909090909091e-05,
      "loss": 0.5585,
      "step": 28
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 1.0127092599868774,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 0.4625,
      "step": 30
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.42090320587158203,
      "learning_rate": 5.818181818181818e-05,
      "loss": 0.3105,
      "step": 32
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5279206037521362,
      "eval_runtime": 20.9926,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 33
    },
    {
      "epoch": 3.090909090909091,
      "grad_norm": 0.8818747401237488,
      "learning_rate": 6.181818181818182e-05,
      "loss": 0.6171,
      "step": 34
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 1.9256893396377563,
      "learning_rate": 6.545454545454546e-05,
      "loss": 0.1621,
      "step": 36
    },
    {
      "epoch": 3.4545454545454546,
      "grad_norm": 0.47599685192108154,
      "learning_rate": 6.90909090909091e-05,
      "loss": 0.4579,
      "step": 38
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.3722813129425049,
      "learning_rate": 7.272727272727273e-05,
      "loss": 0.4309,
      "step": 40
    },
    {
      "epoch": 3.8181818181818183,
      "grad_norm": 0.4472358524799347,
      "learning_rate": 7.636363636363637e-05,
      "loss": 0.5221,
      "step": 42
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.48660457134246826,
      "learning_rate": 8e-05,
      "loss": 0.4824,
      "step": 44
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.47061729431152344,
      "eval_runtime": 20.9956,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 44
    },
    {
      "epoch": 4.181818181818182,
      "grad_norm": 0.2496122121810913,
      "learning_rate": 8.363636363636364e-05,
      "loss": 0.2864,
      "step": 46
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 0.43212294578552246,
      "learning_rate": 8.727272727272727e-05,
      "loss": 0.4797,
      "step": 48
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 0.32961055636405945,
      "learning_rate": 9.090909090909092e-05,
      "loss": 0.316,
      "step": 50
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 0.3680732250213623,
      "learning_rate": 9.454545454545455e-05,
      "loss": 0.3711,
      "step": 52
    },
    {
      "epoch": 4.909090909090909,
      "grad_norm": 0.4210279881954193,
      "learning_rate": 9.818181818181818e-05,
      "loss": 0.3517,
      "step": 54
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.47503742575645447,
      "eval_runtime": 20.9931,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 55
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 0.39646658301353455,
      "learning_rate": 9.999899300364532e-05,
      "loss": 0.3798,
      "step": 56
    },
    {
      "epoch": 5.2727272727272725,
      "grad_norm": 0.4636838734149933,
      "learning_rate": 9.99909372761763e-05,
      "loss": 0.3494,
      "step": 58
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 2.6332552433013916,
      "learning_rate": 9.997482711915927e-05,
      "loss": 0.2778,
      "step": 60
    },
    {
      "epoch": 5.636363636363637,
      "grad_norm": 0.3376973271369934,
      "learning_rate": 9.99506651282272e-05,
      "loss": 0.3365,
      "step": 62
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 0.6036357283592224,
      "learning_rate": 9.991845519630678e-05,
      "loss": 0.2182,
      "step": 64
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.6817249059677124,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.3732,
      "step": 66
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.43232032656669617,
      "eval_runtime": 20.9934,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 66
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 0.3437633514404297,
      "learning_rate": 9.982991356370404e-05,
      "loss": 0.2202,
      "step": 68
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 0.3820938766002655,
      "learning_rate": 9.977359612865423e-05,
      "loss": 0.1994,
      "step": 70
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 0.6148548126220703,
      "learning_rate": 9.970925928158274e-05,
      "loss": 0.3668,
      "step": 72
    },
    {
      "epoch": 6.7272727272727275,
      "grad_norm": 0.48272010684013367,
      "learning_rate": 9.963691338830044e-05,
      "loss": 0.2936,
      "step": 74
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 0.4327438175678253,
      "learning_rate": 9.955657010501806e-05,
      "loss": 0.2019,
      "step": 76
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.48579612374305725,
      "eval_runtime": 20.9903,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 77
    },
    {
      "epoch": 7.090909090909091,
      "grad_norm": 0.7238974571228027,
      "learning_rate": 9.946824237646824e-05,
      "loss": 0.1178,
      "step": 78
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 0.7832289934158325,
      "learning_rate": 9.937194443381972e-05,
      "loss": 0.228,
      "step": 80
    },
    {
      "epoch": 7.454545454545454,
      "grad_norm": 0.5383899211883545,
      "learning_rate": 9.926769179238466e-05,
      "loss": 0.0936,
      "step": 82
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 0.40380972623825073,
      "learning_rate": 9.915550124911866e-05,
      "loss": 0.1099,
      "step": 84
    },
    {
      "epoch": 7.818181818181818,
      "grad_norm": 0.5559152364730835,
      "learning_rate": 9.903539087991462e-05,
      "loss": 0.1891,
      "step": 86
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.0801582336425781,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.1766,
      "step": 88
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.4911709129810333,
      "eval_runtime": 20.9948,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 88
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 0.5906036496162415,
      "learning_rate": 9.877148934427037e-05,
      "loss": 0.1261,
      "step": 90
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 0.7344932556152344,
      "learning_rate": 9.862774069706346e-05,
      "loss": 0.0682,
      "step": 92
    },
    {
      "epoch": 8.545454545454545,
      "grad_norm": 0.9448034167289734,
      "learning_rate": 9.847615725553456e-05,
      "loss": 0.037,
      "step": 94
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 0.8265174031257629,
      "learning_rate": 9.831676344247342e-05,
      "loss": 0.0889,
      "step": 96
    },
    {
      "epoch": 8.909090909090908,
      "grad_norm": 0.9721383452415466,
      "learning_rate": 9.814958493905963e-05,
      "loss": 0.1,
      "step": 98
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.5915277600288391,
      "eval_runtime": 20.9947,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 99
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 0.6191742420196533,
      "learning_rate": 9.797464868072488e-05,
      "loss": 0.0701,
      "step": 100
    },
    {
      "epoch": 9.272727272727273,
      "grad_norm": 0.10014607012271881,
      "learning_rate": 9.779198285281325e-05,
      "loss": 0.0196,
      "step": 102
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 1.462838888168335,
      "learning_rate": 9.760161688604008e-05,
      "loss": 0.077,
      "step": 104
    },
    {
      "epoch": 9.636363636363637,
      "grad_norm": 0.6379705667495728,
      "learning_rate": 9.740358145174998e-05,
      "loss": 0.0315,
      "step": 106
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 0.5921292304992676,
      "learning_rate": 9.719790845697533e-05,
      "loss": 0.0293,
      "step": 108
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.1837440729141235,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0326,
      "step": 110
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.6782414317131042,
      "eval_runtime": 21.0001,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 110
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 0.5487864017486572,
      "learning_rate": 9.676378356149734e-05,
      "loss": 0.0091,
      "step": 112
    },
    {
      "epoch": 10.363636363636363,
      "grad_norm": 0.4271487891674042,
      "learning_rate": 9.653540160603956e-05,
      "loss": 0.0222,
      "step": 114
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 0.7169693112373352,
      "learning_rate": 9.629952196931901e-05,
      "loss": 0.016,
      "step": 116
    },
    {
      "epoch": 10.727272727272727,
      "grad_norm": 1.3038582801818848,
      "learning_rate": 9.60561826557425e-05,
      "loss": 0.0375,
      "step": 118
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 0.32070016860961914,
      "learning_rate": 9.580542287160348e-05,
      "loss": 0.0057,
      "step": 120
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.692686140537262,
      "eval_runtime": 21.0014,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 121
    },
    {
      "epoch": 11.090909090909092,
      "grad_norm": 0.09818615019321442,
      "learning_rate": 9.554728301876526e-05,
      "loss": 0.009,
      "step": 122
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 0.7170988321304321,
      "learning_rate": 9.528180468815155e-05,
      "loss": 0.0104,
      "step": 124
    },
    {
      "epoch": 11.454545454545455,
      "grad_norm": 0.20836713910102844,
      "learning_rate": 9.50090306530454e-05,
      "loss": 0.0066,
      "step": 126
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 0.30115947127342224,
      "learning_rate": 9.472900486219769e-05,
      "loss": 0.0114,
      "step": 128
    },
    {
      "epoch": 11.818181818181818,
      "grad_norm": 1.4105948209762573,
      "learning_rate": 9.444177243274618e-05,
      "loss": 0.0106,
      "step": 130
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.8289133310317993,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0107,
      "step": 132
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.8774659037590027,
      "eval_runtime": 21.0027,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 132
    },
    {
      "epoch": 12.181818181818182,
      "grad_norm": 0.2651340365409851,
      "learning_rate": 9.384587392471515e-05,
      "loss": 0.0038,
      "step": 134
    },
    {
      "epoch": 12.363636363636363,
      "grad_norm": 0.0375906303524971,
      "learning_rate": 9.353730385598887e-05,
      "loss": 0.0007,
      "step": 136
    },
    {
      "epoch": 12.545454545454545,
      "grad_norm": 0.5629562735557556,
      "learning_rate": 9.322171915289635e-05,
      "loss": 0.0049,
      "step": 138
    },
    {
      "epoch": 12.727272727272727,
      "grad_norm": 0.6287766098976135,
      "learning_rate": 9.289917066174886e-05,
      "loss": 0.008,
      "step": 140
    },
    {
      "epoch": 12.909090909090908,
      "grad_norm": 0.1741022914648056,
      "learning_rate": 9.256971035084785e-05,
      "loss": 0.0022,
      "step": 142
    },
    {
      "epoch": 13.0,
      "eval_loss": 0.8793678283691406,
      "eval_runtime": 20.997,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 143
    },
    {
      "epoch": 13.0,
      "step": 143,
      "total_flos": 8.929318065825382e+16,
      "train_loss": 0.46659035206010396,
      "train_runtime": 3291.7176,
      "train_samples_per_second": 1.337,
      "train_steps_per_second": 0.167
    }
  ],
  "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": 8.929318065825382e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}