{
  "best_metric": 1.1381473541259766,
  "best_model_checkpoint": "data/Llama-31-8B_task-3_180-samples_config-2_full/checkpoint-85",
  "epoch": 16.941176470588236,
  "eval_steps": 500,
  "global_step": 144,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 0.5176953077316284,
      "learning_rate": 2.5e-06,
      "loss": 1.7131,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 0.5487761497497559,
      "learning_rate": 5e-06,
      "loss": 1.7301,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 0.4699436128139496,
      "learning_rate": 1e-05,
      "loss": 1.7256,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 0.42270535230636597,
      "learning_rate": 1.5e-05,
      "loss": 1.6691,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 0.41488713026046753,
      "learning_rate": 2e-05,
      "loss": 1.695,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 1.6759352684020996,
      "eval_runtime": 31.8571,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.5180110335350037,
      "learning_rate": 2.5e-05,
      "loss": 1.6882,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 0.35783591866493225,
      "learning_rate": 3e-05,
      "loss": 1.6225,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 0.4034726619720459,
      "learning_rate": 3.5e-05,
      "loss": 1.5951,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 0.4607871472835541,
      "learning_rate": 4e-05,
      "loss": 1.5868,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.5454449653625488,
      "eval_runtime": 31.8528,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.2619361877441406,
      "learning_rate": 4.5e-05,
      "loss": 1.5575,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.23799021542072296,
      "learning_rate": 5e-05,
      "loss": 1.4926,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.2394094467163086,
      "learning_rate": 5.500000000000001e-05,
      "loss": 1.4593,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.255198210477829,
      "learning_rate": 6e-05,
      "loss": 1.473,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 1.4197732210159302,
      "eval_runtime": 31.8597,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.23705199360847473,
      "learning_rate": 6.500000000000001e-05,
      "loss": 1.3998,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.2624831199645996,
      "learning_rate": 7e-05,
      "loss": 1.3573,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.27011838555336,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.2873,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.2953531742095947,
      "learning_rate": 8e-05,
      "loss": 1.3048,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.2443869560956955,
      "learning_rate": 8.5e-05,
      "loss": 1.2035,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.2535014152526855,
      "eval_runtime": 31.8533,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.2208809107542038,
      "learning_rate": 9e-05,
      "loss": 1.1907,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.1921616941690445,
      "learning_rate": 9.5e-05,
      "loss": 1.1664,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.15039095282554626,
      "learning_rate": 0.0001,
      "loss": 1.1716,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.15168997645378113,
      "learning_rate": 9.999238475781957e-05,
      "loss": 1.1293,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 1.199088454246521,
      "eval_runtime": 31.8617,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.16245824098587036,
      "learning_rate": 9.99695413509548e-05,
      "loss": 1.0927,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.17652744054794312,
      "learning_rate": 9.99314767377287e-05,
      "loss": 1.1417,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.18893443048000336,
      "learning_rate": 9.987820251299122e-05,
      "loss": 1.1069,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.16078542172908783,
      "learning_rate": 9.980973490458728e-05,
      "loss": 1.1361,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.1733344793319702,
      "eval_runtime": 31.8555,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.1555337905883789,
      "learning_rate": 9.972609476841367e-05,
      "loss": 1.0797,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.14737282693386078,
      "learning_rate": 9.962730758206611e-05,
      "loss": 1.0578,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.16058345139026642,
      "learning_rate": 9.951340343707852e-05,
      "loss": 1.0911,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.14137417078018188,
      "learning_rate": 9.938441702975689e-05,
      "loss": 1.1333,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 1.1571494340896606,
      "eval_runtime": 31.8579,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.15388211607933044,
      "learning_rate": 9.924038765061042e-05,
      "loss": 1.0784,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.1491321623325348,
      "learning_rate": 9.908135917238321e-05,
      "loss": 1.07,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.17386263608932495,
      "learning_rate": 9.890738003669029e-05,
      "loss": 1.0382,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 0.17576885223388672,
      "learning_rate": 9.871850323926177e-05,
      "loss": 1.0377,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.1866431087255478,
      "learning_rate": 9.851478631379982e-05,
      "loss": 1.0612,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.1461663246154785,
      "eval_runtime": 31.8577,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.1761140078306198,
      "learning_rate": 9.829629131445342e-05,
      "loss": 1.0261,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 0.19430580735206604,
      "learning_rate": 9.806308479691595e-05,
      "loss": 1.0604,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 0.19356723129749298,
      "learning_rate": 9.781523779815179e-05,
      "loss": 1.0369,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 0.21120838820934296,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.9895,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 1.1391992568969727,
      "eval_runtime": 31.8668,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 0.16879647970199585,
      "learning_rate": 9.727592877996585e-05,
      "loss": 1.0034,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.1657697558403015,
      "learning_rate": 9.698463103929542e-05,
      "loss": 1.0109,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 0.19832511246204376,
      "learning_rate": 9.667902132486009e-05,
      "loss": 1.003,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 0.219316765666008,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.9858,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.1381473541259766,
      "eval_runtime": 31.8595,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.2147984355688095,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.9601,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.21176777780056,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.9891,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 0.22931328415870667,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.9392,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.24265824258327484,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.939,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 1.1420247554779053,
      "eval_runtime": 31.8533,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.24943289160728455,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.9836,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 0.61884605884552,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.9779,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 0.3478313982486725,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.8764,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.31895795464515686,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.9216,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.32360953092575073,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.8747,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.1664069890975952,
      "eval_runtime": 31.8594,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 102
    },
    {
      "epoch": 12.235294117647058,
      "grad_norm": 0.38271987438201904,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.8447,
      "step": 104
    },
    {
      "epoch": 12.470588235294118,
      "grad_norm": 0.4208544194698334,
      "learning_rate": 9.193352839727121e-05,
      "loss": 0.8554,
      "step": 106
    },
    {
      "epoch": 12.705882352941176,
      "grad_norm": 0.3989667296409607,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.8843,
      "step": 108
    },
    {
      "epoch": 12.941176470588236,
      "grad_norm": 0.384284645318985,
      "learning_rate": 9.09576022144496e-05,
      "loss": 0.8694,
      "step": 110
    },
    {
      "epoch": 12.941176470588236,
      "eval_loss": 1.1779769659042358,
      "eval_runtime": 31.8607,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 110
    },
    {
      "epoch": 13.176470588235293,
      "grad_norm": 0.375377893447876,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.8364,
      "step": 112
    },
    {
      "epoch": 13.411764705882353,
      "grad_norm": 0.4887729585170746,
      "learning_rate": 8.993177550236464e-05,
      "loss": 0.7637,
      "step": 114
    },
    {
      "epoch": 13.647058823529411,
      "grad_norm": 0.48660561442375183,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.7966,
      "step": 116
    },
    {
      "epoch": 13.882352941176471,
      "grad_norm": 0.5761192440986633,
      "learning_rate": 8.885729807284856e-05,
      "loss": 0.8188,
      "step": 118
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.2246310710906982,
      "eval_runtime": 31.8527,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 119
    },
    {
      "epoch": 14.117647058823529,
      "grad_norm": 0.523510754108429,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.8017,
      "step": 120
    },
    {
      "epoch": 14.352941176470589,
      "grad_norm": 0.6305602192878723,
      "learning_rate": 8.773547901113862e-05,
      "loss": 0.6666,
      "step": 122
    },
    {
      "epoch": 14.588235294117647,
      "grad_norm": 0.5910943746566772,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.7669,
      "step": 124
    },
    {
      "epoch": 14.823529411764707,
      "grad_norm": 0.6196286082267761,
      "learning_rate": 8.656768508095853e-05,
      "loss": 0.697,
      "step": 126
    },
    {
      "epoch": 14.941176470588236,
      "eval_loss": 1.234824538230896,
      "eval_runtime": 31.858,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 127
    },
    {
      "epoch": 15.058823529411764,
      "grad_norm": 0.6877910494804382,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.7504,
      "step": 128
    },
    {
      "epoch": 15.294117647058824,
      "grad_norm": 1.5988885164260864,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.6642,
      "step": 130
    },
    {
      "epoch": 15.529411764705882,
      "grad_norm": 0.8035158514976501,
      "learning_rate": 8.473291852294987e-05,
      "loss": 0.6829,
      "step": 132
    },
    {
      "epoch": 15.764705882352942,
      "grad_norm": 0.7143783569335938,
      "learning_rate": 8.409991800312493e-05,
      "loss": 0.6849,
      "step": 134
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.6947578191757202,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.6048,
      "step": 136
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.310172200202942,
      "eval_runtime": 31.8619,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 136
    },
    {
      "epoch": 16.235294117647058,
      "grad_norm": 2.3495612144470215,
      "learning_rate": 8.280295144952536e-05,
      "loss": 0.552,
      "step": 138
    },
    {
      "epoch": 16.470588235294116,
      "grad_norm": 0.9533553719520569,
      "learning_rate": 8.213938048432697e-05,
      "loss": 0.6525,
      "step": 140
    },
    {
      "epoch": 16.705882352941178,
      "grad_norm": 0.9292975664138794,
      "learning_rate": 8.146601955249188e-05,
      "loss": 0.5725,
      "step": 142
    },
    {
      "epoch": 16.941176470588236,
      "grad_norm": 0.9577359557151794,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.5898,
      "step": 144
    },
    {
      "epoch": 16.941176470588236,
      "eval_loss": 1.3189616203308105,
      "eval_runtime": 31.8633,
      "eval_samples_per_second": 1.13,
      "eval_steps_per_second": 1.13,
      "step": 144
    },
    {
      "epoch": 16.941176470588236,
      "step": 144,
      "total_flos": 1.8132641623362765e+17,
      "train_loss": 1.0652225611524448,
      "train_runtime": 6655.309,
      "train_samples_per_second": 1.022,
      "train_steps_per_second": 0.06
    }
  ],
  "logging_steps": 2,
  "max_steps": 400,
  "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": 1.8132641623362765e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}