{
  "best_metric": 0.5402693748474121,
  "best_model_checkpoint": "data/Mistral-7B_task-2_180-samples_config-2/checkpoint-42",
  "epoch": 12.0,
  "eval_steps": 500,
  "global_step": 102,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11764705882352941,
      "grad_norm": 2.3155293464660645,
      "learning_rate": 2.5e-06,
      "loss": 1.0419,
      "step": 1
    },
    {
      "epoch": 0.23529411764705882,
      "grad_norm": 2.6562511920928955,
      "learning_rate": 5e-06,
      "loss": 1.0989,
      "step": 2
    },
    {
      "epoch": 0.47058823529411764,
      "grad_norm": 2.378835916519165,
      "learning_rate": 1e-05,
      "loss": 1.0469,
      "step": 4
    },
    {
      "epoch": 0.7058823529411765,
      "grad_norm": 2.085439682006836,
      "learning_rate": 1.5e-05,
      "loss": 0.9504,
      "step": 6
    },
    {
      "epoch": 0.9411764705882353,
      "grad_norm": 1.7603436708450317,
      "learning_rate": 2e-05,
      "loss": 0.8897,
      "step": 8
    },
    {
      "epoch": 0.9411764705882353,
      "eval_loss": 0.8554134964942932,
      "eval_runtime": 45.606,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.789,
      "step": 8
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 1.640693187713623,
      "learning_rate": 2.5e-05,
      "loss": 0.8518,
      "step": 10
    },
    {
      "epoch": 1.4117647058823528,
      "grad_norm": 0.8897013068199158,
      "learning_rate": 3e-05,
      "loss": 0.7464,
      "step": 12
    },
    {
      "epoch": 1.6470588235294117,
      "grad_norm": 0.5992234349250793,
      "learning_rate": 3.5e-05,
      "loss": 0.659,
      "step": 14
    },
    {
      "epoch": 1.8823529411764706,
      "grad_norm": 0.5794339776039124,
      "learning_rate": 4e-05,
      "loss": 0.6578,
      "step": 16
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6425343751907349,
      "eval_runtime": 45.6142,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.789,
      "step": 17
    },
    {
      "epoch": 2.1176470588235294,
      "grad_norm": 0.49187347292900085,
      "learning_rate": 4.5e-05,
      "loss": 0.6511,
      "step": 18
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.4880034327507019,
      "learning_rate": 5e-05,
      "loss": 0.5877,
      "step": 20
    },
    {
      "epoch": 2.588235294117647,
      "grad_norm": 0.4656303822994232,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.6074,
      "step": 22
    },
    {
      "epoch": 2.8235294117647056,
      "grad_norm": 0.4449208974838257,
      "learning_rate": 6e-05,
      "loss": 0.6205,
      "step": 24
    },
    {
      "epoch": 2.9411764705882355,
      "eval_loss": 0.595414400100708,
      "eval_runtime": 45.6824,
      "eval_samples_per_second": 0.788,
      "eval_steps_per_second": 0.788,
      "step": 25
    },
    {
      "epoch": 3.0588235294117645,
      "grad_norm": 0.4166175127029419,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.5775,
      "step": 26
    },
    {
      "epoch": 3.2941176470588234,
      "grad_norm": 0.3330927789211273,
      "learning_rate": 7e-05,
      "loss": 0.5532,
      "step": 28
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.37248340249061584,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.5553,
      "step": 30
    },
    {
      "epoch": 3.764705882352941,
      "grad_norm": 0.3898817002773285,
      "learning_rate": 8e-05,
      "loss": 0.5447,
      "step": 32
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.3806105852127075,
      "learning_rate": 8.5e-05,
      "loss": 0.4973,
      "step": 34
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.5606213212013245,
      "eval_runtime": 45.6824,
      "eval_samples_per_second": 0.788,
      "eval_steps_per_second": 0.788,
      "step": 34
    },
    {
      "epoch": 4.235294117647059,
      "grad_norm": 0.37909018993377686,
      "learning_rate": 9e-05,
      "loss": 0.5128,
      "step": 36
    },
    {
      "epoch": 4.470588235294118,
      "grad_norm": 0.39016708731651306,
      "learning_rate": 9.5e-05,
      "loss": 0.4783,
      "step": 38
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.34981122612953186,
      "learning_rate": 0.0001,
      "loss": 0.4516,
      "step": 40
    },
    {
      "epoch": 4.9411764705882355,
      "grad_norm": 0.40651199221611023,
      "learning_rate": 9.999238475781957e-05,
      "loss": 0.4478,
      "step": 42
    },
    {
      "epoch": 4.9411764705882355,
      "eval_loss": 0.5402693748474121,
      "eval_runtime": 45.6024,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.789,
      "step": 42
    },
    {
      "epoch": 5.176470588235294,
      "grad_norm": 0.38201889395713806,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.4217,
      "step": 44
    },
    {
      "epoch": 5.411764705882353,
      "grad_norm": 0.3878049850463867,
      "learning_rate": 9.99314767377287e-05,
      "loss": 0.4203,
      "step": 46
    },
    {
      "epoch": 5.647058823529412,
      "grad_norm": 0.463470458984375,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.3668,
      "step": 48
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.5280053019523621,
      "learning_rate": 9.980973490458728e-05,
      "loss": 0.3946,
      "step": 50
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.5508570075035095,
      "eval_runtime": 45.6036,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.789,
      "step": 51
    },
    {
      "epoch": 6.117647058823529,
      "grad_norm": 0.5039118528366089,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.344,
      "step": 52
    },
    {
      "epoch": 6.352941176470588,
      "grad_norm": 0.5101126432418823,
      "learning_rate": 9.962730758206611e-05,
      "loss": 0.2839,
      "step": 54
    },
    {
      "epoch": 6.588235294117647,
      "grad_norm": 0.7359801530838013,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.2958,
      "step": 56
    },
    {
      "epoch": 6.823529411764706,
      "grad_norm": 0.7432470321655273,
      "learning_rate": 9.938441702975689e-05,
      "loss": 0.3109,
      "step": 58
    },
    {
      "epoch": 6.9411764705882355,
      "eval_loss": 0.595815896987915,
      "eval_runtime": 45.6043,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.789,
      "step": 59
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.7509456276893616,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.2704,
      "step": 60
    },
    {
      "epoch": 7.294117647058823,
      "grad_norm": 0.7170895338058472,
      "learning_rate": 9.908135917238321e-05,
      "loss": 0.2012,
      "step": 62
    },
    {
      "epoch": 7.529411764705882,
      "grad_norm": 0.9948111176490784,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.1713,
      "step": 64
    },
    {
      "epoch": 7.764705882352941,
      "grad_norm": 1.1264708042144775,
      "learning_rate": 9.871850323926177e-05,
      "loss": 0.1708,
      "step": 66
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.9287075996398926,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.182,
      "step": 68
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.6715656518936157,
      "eval_runtime": 45.6085,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.789,
      "step": 68
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.8156751990318298,
      "learning_rate": 9.829629131445342e-05,
      "loss": 0.115,
      "step": 70
    },
    {
      "epoch": 8.470588235294118,
      "grad_norm": 1.2089699506759644,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.1075,
      "step": 72
    },
    {
      "epoch": 8.705882352941176,
      "grad_norm": 1.3467602729797363,
      "learning_rate": 9.781523779815179e-05,
      "loss": 0.0964,
      "step": 74
    },
    {
      "epoch": 8.941176470588236,
      "grad_norm": 0.9090074300765991,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0682,
      "step": 76
    },
    {
      "epoch": 8.941176470588236,
      "eval_loss": 0.7786654233932495,
      "eval_runtime": 45.6063,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.789,
      "step": 76
    },
    {
      "epoch": 9.176470588235293,
      "grad_norm": 1.0487481355667114,
      "learning_rate": 9.727592877996585e-05,
      "loss": 0.066,
      "step": 78
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 1.0376051664352417,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0467,
      "step": 80
    },
    {
      "epoch": 9.647058823529411,
      "grad_norm": 1.2137938737869263,
      "learning_rate": 9.667902132486009e-05,
      "loss": 0.0452,
      "step": 82
    },
    {
      "epoch": 9.882352941176471,
      "grad_norm": 1.0914534330368042,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.0461,
      "step": 84
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.9119131565093994,
      "eval_runtime": 45.6034,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.789,
      "step": 85
    },
    {
      "epoch": 10.117647058823529,
      "grad_norm": 0.5998616218566895,
      "learning_rate": 9.602524267262203e-05,
      "loss": 0.0361,
      "step": 86
    },
    {
      "epoch": 10.352941176470589,
      "grad_norm": 0.7274846434593201,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0236,
      "step": 88
    },
    {
      "epoch": 10.588235294117647,
      "grad_norm": 0.8589937090873718,
      "learning_rate": 9.53153893518325e-05,
      "loss": 0.029,
      "step": 90
    },
    {
      "epoch": 10.823529411764707,
      "grad_norm": 0.9843662977218628,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.03,
      "step": 92
    },
    {
      "epoch": 10.941176470588236,
      "eval_loss": 0.989643394947052,
      "eval_runtime": 45.6021,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.789,
      "step": 93
    },
    {
      "epoch": 11.058823529411764,
      "grad_norm": 0.5862740874290466,
      "learning_rate": 9.45503262094184e-05,
      "loss": 0.0274,
      "step": 94
    },
    {
      "epoch": 11.294117647058824,
      "grad_norm": 0.5251568555831909,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.012,
      "step": 96
    },
    {
      "epoch": 11.529411764705882,
      "grad_norm": 0.4932527542114258,
      "learning_rate": 9.373098535696979e-05,
      "loss": 0.0133,
      "step": 98
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.8276335597038269,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0138,
      "step": 100
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.488267183303833,
      "learning_rate": 9.285836503510562e-05,
      "loss": 0.0116,
      "step": 102
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.0824575424194336,
      "eval_runtime": 45.6139,
      "eval_samples_per_second": 0.789,
      "eval_steps_per_second": 0.789,
      "step": 102
    },
    {
      "epoch": 12.0,
      "step": 102,
      "total_flos": 1.9338296750925414e+17,
      "train_loss": 0.3643010138333136,
      "train_runtime": 7015.4881,
      "train_samples_per_second": 0.969,
      "train_steps_per_second": 0.057
    }
  ],
  "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": 1.9338296750925414e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}