{
  "best_metric": 0.8860878944396973,
  "best_model_checkpoint": "data/Llama-31-8B_task-1_120-samples_config-1_auto/checkpoint-44",
  "epoch": 11.0,
  "eval_steps": 500,
  "global_step": 121,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09090909090909091,
      "grad_norm": 0.9034837484359741,
      "learning_rate": 1.818181818181818e-06,
      "loss": 1.1489,
      "step": 1
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 0.8655723333358765,
      "learning_rate": 3.636363636363636e-06,
      "loss": 1.1738,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 0.7380895018577576,
      "learning_rate": 7.272727272727272e-06,
      "loss": 1.0975,
      "step": 4
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 0.9108906388282776,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 1.198,
      "step": 6
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 0.8552144765853882,
      "learning_rate": 1.4545454545454545e-05,
      "loss": 1.1239,
      "step": 8
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.9636886119842529,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 1.2238,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.1612118482589722,
      "eval_runtime": 9.8682,
      "eval_samples_per_second": 2.432,
      "eval_steps_per_second": 2.432,
      "step": 11
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 0.8112488985061646,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 1.1075,
      "step": 12
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 0.7141273021697998,
      "learning_rate": 2.5454545454545454e-05,
      "loss": 1.0238,
      "step": 14
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 0.601105809211731,
      "learning_rate": 2.909090909090909e-05,
      "loss": 0.9827,
      "step": 16
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 0.4828379452228546,
      "learning_rate": 3.272727272727273e-05,
      "loss": 1.0882,
      "step": 18
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.6637647747993469,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 1.0579,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6359596848487854,
      "learning_rate": 4e-05,
      "loss": 0.8979,
      "step": 22
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.0151863098144531,
      "eval_runtime": 9.8636,
      "eval_samples_per_second": 2.433,
      "eval_steps_per_second": 2.433,
      "step": 22
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 0.6983725428581238,
      "learning_rate": 4.3636363636363636e-05,
      "loss": 0.888,
      "step": 24
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 0.567716121673584,
      "learning_rate": 4.7272727272727275e-05,
      "loss": 0.9355,
      "step": 26
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.526594340801239,
      "learning_rate": 5.090909090909091e-05,
      "loss": 0.9666,
      "step": 28
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.48598650097846985,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 0.8928,
      "step": 30
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.592958927154541,
      "learning_rate": 5.818181818181818e-05,
      "loss": 0.8593,
      "step": 32
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.9378392696380615,
      "eval_runtime": 9.8953,
      "eval_samples_per_second": 2.425,
      "eval_steps_per_second": 2.425,
      "step": 33
    },
    {
      "epoch": 3.090909090909091,
      "grad_norm": 0.5815737843513489,
      "learning_rate": 6.181818181818182e-05,
      "loss": 0.8167,
      "step": 34
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 0.6120511293411255,
      "learning_rate": 6.545454545454546e-05,
      "loss": 0.7236,
      "step": 36
    },
    {
      "epoch": 3.4545454545454546,
      "grad_norm": 0.6058388948440552,
      "learning_rate": 6.90909090909091e-05,
      "loss": 0.8298,
      "step": 38
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.5814407467842102,
      "learning_rate": 7.272727272727273e-05,
      "loss": 0.7682,
      "step": 40
    },
    {
      "epoch": 3.8181818181818183,
      "grad_norm": 0.625813364982605,
      "learning_rate": 7.636363636363637e-05,
      "loss": 0.8129,
      "step": 42
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7839888334274292,
      "learning_rate": 8e-05,
      "loss": 0.7403,
      "step": 44
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.8860878944396973,
      "eval_runtime": 9.9057,
      "eval_samples_per_second": 2.423,
      "eval_steps_per_second": 2.423,
      "step": 44
    },
    {
      "epoch": 4.181818181818182,
      "grad_norm": 0.7345572710037231,
      "learning_rate": 8.363636363636364e-05,
      "loss": 0.6877,
      "step": 46
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 0.7543135285377502,
      "learning_rate": 8.727272727272727e-05,
      "loss": 0.7107,
      "step": 48
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 0.9157195687294006,
      "learning_rate": 9.090909090909092e-05,
      "loss": 0.673,
      "step": 50
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 0.9693506360054016,
      "learning_rate": 9.454545454545455e-05,
      "loss": 0.6431,
      "step": 52
    },
    {
      "epoch": 4.909090909090909,
      "grad_norm": 0.8752488493919373,
      "learning_rate": 9.818181818181818e-05,
      "loss": 0.5823,
      "step": 54
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.9115815162658691,
      "eval_runtime": 9.8961,
      "eval_samples_per_second": 2.425,
      "eval_steps_per_second": 2.425,
      "step": 55
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 0.8184692859649658,
      "learning_rate": 9.999899300364532e-05,
      "loss": 0.563,
      "step": 56
    },
    {
      "epoch": 5.2727272727272725,
      "grad_norm": 0.8796578645706177,
      "learning_rate": 9.99909372761763e-05,
      "loss": 0.5209,
      "step": 58
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 0.9785192012786865,
      "learning_rate": 9.997482711915927e-05,
      "loss": 0.4667,
      "step": 60
    },
    {
      "epoch": 5.636363636363637,
      "grad_norm": 1.2116471529006958,
      "learning_rate": 9.99506651282272e-05,
      "loss": 0.4563,
      "step": 62
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 1.3979865312576294,
      "learning_rate": 9.991845519630678e-05,
      "loss": 0.4748,
      "step": 64
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.3544435501098633,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.4269,
      "step": 66
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.032470941543579,
      "eval_runtime": 9.8941,
      "eval_samples_per_second": 2.426,
      "eval_steps_per_second": 2.426,
      "step": 66
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 1.0311542749404907,
      "learning_rate": 9.982991356370404e-05,
      "loss": 0.3077,
      "step": 68
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 1.0964933633804321,
      "learning_rate": 9.977359612865423e-05,
      "loss": 0.3093,
      "step": 70
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 1.4322137832641602,
      "learning_rate": 9.970925928158274e-05,
      "loss": 0.3127,
      "step": 72
    },
    {
      "epoch": 6.7272727272727275,
      "grad_norm": 1.3604410886764526,
      "learning_rate": 9.963691338830044e-05,
      "loss": 0.2649,
      "step": 74
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 1.505393624305725,
      "learning_rate": 9.955657010501806e-05,
      "loss": 0.2688,
      "step": 76
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.0846952199935913,
      "eval_runtime": 9.9072,
      "eval_samples_per_second": 2.422,
      "eval_steps_per_second": 2.422,
      "step": 77
    },
    {
      "epoch": 7.090909090909091,
      "grad_norm": 1.0734570026397705,
      "learning_rate": 9.946824237646824e-05,
      "loss": 0.2259,
      "step": 78
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 1.4121390581130981,
      "learning_rate": 9.937194443381972e-05,
      "loss": 0.1921,
      "step": 80
    },
    {
      "epoch": 7.454545454545454,
      "grad_norm": 1.5402942895889282,
      "learning_rate": 9.926769179238466e-05,
      "loss": 0.1367,
      "step": 82
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 1.5536012649536133,
      "learning_rate": 9.915550124911866e-05,
      "loss": 0.1337,
      "step": 84
    },
    {
      "epoch": 7.818181818181818,
      "grad_norm": 1.4846868515014648,
      "learning_rate": 9.903539087991462e-05,
      "loss": 0.1779,
      "step": 86
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.4603244066238403,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.1625,
      "step": 88
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.2474989891052246,
      "eval_runtime": 9.9003,
      "eval_samples_per_second": 2.424,
      "eval_steps_per_second": 2.424,
      "step": 88
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 0.9843153953552246,
      "learning_rate": 9.877148934427037e-05,
      "loss": 0.0988,
      "step": 90
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 0.8816183805465698,
      "learning_rate": 9.862774069706346e-05,
      "loss": 0.0893,
      "step": 92
    },
    {
      "epoch": 8.545454545454545,
      "grad_norm": 0.8757835030555725,
      "learning_rate": 9.847615725553456e-05,
      "loss": 0.092,
      "step": 94
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 0.8291594982147217,
      "learning_rate": 9.831676344247342e-05,
      "loss": 0.1048,
      "step": 96
    },
    {
      "epoch": 8.909090909090908,
      "grad_norm": 1.4149295091629028,
      "learning_rate": 9.814958493905963e-05,
      "loss": 0.0921,
      "step": 98
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.4958215951919556,
      "eval_runtime": 9.8965,
      "eval_samples_per_second": 2.425,
      "eval_steps_per_second": 2.425,
      "step": 99
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 0.8341209888458252,
      "learning_rate": 9.797464868072488e-05,
      "loss": 0.07,
      "step": 100
    },
    {
      "epoch": 9.272727272727273,
      "grad_norm": 1.030566930770874,
      "learning_rate": 9.779198285281325e-05,
      "loss": 0.0663,
      "step": 102
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 1.3842285871505737,
      "learning_rate": 9.760161688604008e-05,
      "loss": 0.0644,
      "step": 104
    },
    {
      "epoch": 9.636363636363637,
      "grad_norm": 0.6888823509216309,
      "learning_rate": 9.740358145174998e-05,
      "loss": 0.0581,
      "step": 106
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 1.4410877227783203,
      "learning_rate": 9.719790845697533e-05,
      "loss": 0.0972,
      "step": 108
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.2411458492279053,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0629,
      "step": 110
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.5465604066848755,
      "eval_runtime": 9.9036,
      "eval_samples_per_second": 2.423,
      "eval_steps_per_second": 2.423,
      "step": 110
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 0.8786189556121826,
      "learning_rate": 9.676378356149734e-05,
      "loss": 0.0381,
      "step": 112
    },
    {
      "epoch": 10.363636363636363,
      "grad_norm": 1.1501296758651733,
      "learning_rate": 9.653540160603956e-05,
      "loss": 0.0429,
      "step": 114
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 0.7037285566329956,
      "learning_rate": 9.629952196931901e-05,
      "loss": 0.041,
      "step": 116
    },
    {
      "epoch": 10.727272727272727,
      "grad_norm": 1.9075895547866821,
      "learning_rate": 9.60561826557425e-05,
      "loss": 0.0659,
      "step": 118
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 1.667096734046936,
      "learning_rate": 9.580542287160348e-05,
      "loss": 0.0518,
      "step": 120
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.649419903755188,
      "eval_runtime": 9.8927,
      "eval_samples_per_second": 2.426,
      "eval_steps_per_second": 2.426,
      "step": 121
    },
    {
      "epoch": 11.0,
      "step": 121,
      "total_flos": 1.6108834355412992e+16,
      "train_loss": 0.5131793877433154,
      "train_runtime": 1169.4049,
      "train_samples_per_second": 3.763,
      "train_steps_per_second": 0.47
    }
  ],
  "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": 1.6108834355412992e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}