{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.1902044698050404,
  "eval_steps": 100,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 5e-05,
      "loss": 1.8229,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 4.9999950203346446e-05,
      "loss": 1.8352,
      "step": 2
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.999980081358417e-05,
      "loss": 1.742,
      "step": 3
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.999955183130829e-05,
      "loss": 1.6315,
      "step": 4
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.999920325751068e-05,
      "loss": 1.6204,
      "step": 5
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.999875509357998e-05,
      "loss": 1.5485,
      "step": 6
    },
    {
      "epoch": 0.01,
      "learning_rate": 4.999820734130155e-05,
      "loss": 1.4713,
      "step": 7
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.999756000285748e-05,
      "loss": 1.4675,
      "step": 8
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.9996813080826606e-05,
      "loss": 1.5091,
      "step": 9
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.999596657818445e-05,
      "loss": 1.4009,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.9995020498303275e-05,
      "loss": 1.2992,
      "step": 11
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.999397484495198e-05,
      "loss": 1.2111,
      "step": 12
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.99928296222962e-05,
      "loss": 1.2281,
      "step": 13
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.999158483489818e-05,
      "loss": 1.2099,
      "step": 14
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.99902404877168e-05,
      "loss": 1.1259,
      "step": 15
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.998879658610761e-05,
      "loss": 1.022,
      "step": 16
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.998725313582272e-05,
      "loss": 1.1276,
      "step": 17
    },
    {
      "epoch": 0.03,
      "learning_rate": 4.998561014301081e-05,
      "loss": 1.1276,
      "step": 18
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.998386761421714e-05,
      "loss": 1.2404,
      "step": 19
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.998202555638346e-05,
      "loss": 1.1622,
      "step": 20
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.998008397684806e-05,
      "loss": 1.1335,
      "step": 21
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.997804288334565e-05,
      "loss": 1.2187,
      "step": 22
    },
    {
      "epoch": 0.04,
      "learning_rate": 4.9975902284007405e-05,
      "loss": 1.0678,
      "step": 23
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.997366218736091e-05,
      "loss": 1.0195,
      "step": 24
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.99713226023301e-05,
      "loss": 1.0138,
      "step": 25
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.9968883538235264e-05,
      "loss": 1.0706,
      "step": 26
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.996634500479297e-05,
      "loss": 0.9617,
      "step": 27
    },
    {
      "epoch": 0.05,
      "learning_rate": 4.996370701211607e-05,
      "loss": 1.034,
      "step": 28
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.996096957071361e-05,
      "loss": 0.9952,
      "step": 29
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.9958132691490816e-05,
      "loss": 0.9598,
      "step": 30
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.995519638574907e-05,
      "loss": 1.0342,
      "step": 31
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.9952160665185825e-05,
      "loss": 0.9147,
      "step": 32
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.994902554189458e-05,
      "loss": 0.924,
      "step": 33
    },
    {
      "epoch": 0.06,
      "learning_rate": 4.994579102836482e-05,
      "loss": 0.9021,
      "step": 34
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.994245713748198e-05,
      "loss": 1.0732,
      "step": 35
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.993902388252739e-05,
      "loss": 0.926,
      "step": 36
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.9935491277178236e-05,
      "loss": 0.9082,
      "step": 37
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.993185933550745e-05,
      "loss": 0.8735,
      "step": 38
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.992812807198372e-05,
      "loss": 1.0212,
      "step": 39
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.9924297501471414e-05,
      "loss": 0.9495,
      "step": 40
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.9920367639230483e-05,
      "loss": 0.9101,
      "step": 41
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.991633850091645e-05,
      "loss": 1.0533,
      "step": 42
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.991221010258034e-05,
      "loss": 0.8759,
      "step": 43
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.990798246066856e-05,
      "loss": 0.969,
      "step": 44
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.9903655592022916e-05,
      "loss": 0.8463,
      "step": 45
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.9899229513880494e-05,
      "loss": 0.844,
      "step": 46
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.989470424387361e-05,
      "loss": 0.8386,
      "step": 47
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.989007980002973e-05,
      "loss": 0.93,
      "step": 48
    },
    {
      "epoch": 0.09,
      "learning_rate": 4.988535620077138e-05,
      "loss": 0.8551,
      "step": 49
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.988053346491613e-05,
      "loss": 0.8633,
      "step": 50
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.987561161167648e-05,
      "loss": 0.9056,
      "step": 51
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.9870590660659755e-05,
      "loss": 0.9366,
      "step": 52
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.986547063186808e-05,
      "loss": 1.0469,
      "step": 53
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.98602515456983e-05,
      "loss": 0.8723,
      "step": 54
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.985493342294184e-05,
      "loss": 0.9484,
      "step": 55
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.9849516284784676e-05,
      "loss": 0.8346,
      "step": 56
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.984400015280724e-05,
      "loss": 0.789,
      "step": 57
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.983838504898433e-05,
      "loss": 0.8986,
      "step": 58
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.9832670995685006e-05,
      "loss": 0.8799,
      "step": 59
    },
    {
      "epoch": 0.11,
      "learning_rate": 4.9826858015672536e-05,
      "loss": 0.9028,
      "step": 60
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.982094613210428e-05,
      "loss": 0.8908,
      "step": 61
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.9814935368531586e-05,
      "loss": 0.8273,
      "step": 62
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.980882574889973e-05,
      "loss": 0.784,
      "step": 63
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.980261729754781e-05,
      "loss": 0.9336,
      "step": 64
    },
    {
      "epoch": 0.12,
      "learning_rate": 4.9796310039208623e-05,
      "loss": 0.8943,
      "step": 65
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.978990399900861e-05,
      "loss": 1.1,
      "step": 66
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.9783399202467714e-05,
      "loss": 0.877,
      "step": 67
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.9776795675499296e-05,
      "loss": 0.8291,
      "step": 68
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.9770093444410046e-05,
      "loss": 0.7997,
      "step": 69
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.9763292535899866e-05,
      "loss": 0.7911,
      "step": 70
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.9756392977061736e-05,
      "loss": 0.8996,
      "step": 71
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.974939479538166e-05,
      "loss": 0.8631,
      "step": 72
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.974229801873854e-05,
      "loss": 0.8303,
      "step": 73
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.9735102675404004e-05,
      "loss": 0.8708,
      "step": 74
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.972780879404239e-05,
      "loss": 0.998,
      "step": 75
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.972041640371056e-05,
      "loss": 1.1049,
      "step": 76
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.971292553385783e-05,
      "loss": 0.774,
      "step": 77
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.970533621432581e-05,
      "loss": 0.8142,
      "step": 78
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.969764847534832e-05,
      "loss": 0.9968,
      "step": 79
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.9689862347551255e-05,
      "loss": 0.9622,
      "step": 80
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.9681977861952464e-05,
      "loss": 0.8342,
      "step": 81
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.967399504996162e-05,
      "loss": 0.9,
      "step": 82
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.966591394338012e-05,
      "loss": 0.8515,
      "step": 83
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.965773457440092e-05,
      "loss": 0.7937,
      "step": 84
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.964945697560844e-05,
      "loss": 0.9577,
      "step": 85
    },
    {
      "epoch": 0.16,
      "learning_rate": 4.9641081179978424e-05,
      "loss": 0.8233,
      "step": 86
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.963260722087779e-05,
      "loss": 0.9769,
      "step": 87
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.9624035132064526e-05,
      "loss": 0.8927,
      "step": 88
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.961536494768754e-05,
      "loss": 0.9383,
      "step": 89
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.960659670228652e-05,
      "loss": 0.9177,
      "step": 90
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.959773043079181e-05,
      "loss": 0.8026,
      "step": 91
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.958876616852427e-05,
      "loss": 0.9647,
      "step": 92
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.9579703951195113e-05,
      "loss": 0.7545,
      "step": 93
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.957054381490579e-05,
      "loss": 0.7836,
      "step": 94
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.9561285796147824e-05,
      "loss": 0.8803,
      "step": 95
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.955192993180269e-05,
      "loss": 0.8988,
      "step": 96
    },
    {
      "epoch": 0.18,
      "learning_rate": 4.9542476259141656e-05,
      "loss": 0.8418,
      "step": 97
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.953292481582561e-05,
      "loss": 0.9004,
      "step": 98
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.9523275639904944e-05,
      "loss": 0.9159,
      "step": 99
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.951352876981939e-05,
      "loss": 0.7605,
      "step": 100
    },
    {
      "epoch": 0.19,
      "eval_loss": 0.7751120924949646,
      "eval_runtime": 2.7398,
      "eval_samples_per_second": 3.285,
      "eval_steps_per_second": 0.73,
      "step": 100
    }
  ],
  "logging_steps": 1,
  "max_steps": 1575,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.406258997362688e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}