{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 100.0,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.355809211730957,
      "learning_rate": 0.0001,
      "loss": 1.3272,
      "step": 1
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.355809211730957,
      "learning_rate": 0.0002,
      "loss": 1.3272,
      "step": 2
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.4446067810058594,
      "learning_rate": 0.00019795918367346938,
      "loss": 1.1654,
      "step": 3
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.521039366722107,
      "learning_rate": 0.0001959183673469388,
      "loss": 0.9031,
      "step": 4
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.5087562799453735,
      "learning_rate": 0.00019387755102040816,
      "loss": 0.6727,
      "step": 5
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.60162353515625,
      "learning_rate": 0.00019183673469387756,
      "loss": 0.4641,
      "step": 6
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.0103554725646973,
      "learning_rate": 0.00018979591836734697,
      "loss": 0.2926,
      "step": 7
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.812863826751709,
      "learning_rate": 0.00018775510204081634,
      "loss": 0.2053,
      "step": 8
    },
    {
      "epoch": 9.0,
      "grad_norm": 1.5256155729293823,
      "learning_rate": 0.00018571428571428572,
      "loss": 0.1473,
      "step": 9
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.875736951828003,
      "learning_rate": 0.00018367346938775512,
      "loss": 0.1351,
      "step": 10
    },
    {
      "epoch": 11.0,
      "grad_norm": 0.4695473313331604,
      "learning_rate": 0.0001816326530612245,
      "loss": 0.1029,
      "step": 11
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.5592666864395142,
      "learning_rate": 0.0001795918367346939,
      "loss": 0.0982,
      "step": 12
    },
    {
      "epoch": 13.0,
      "grad_norm": 0.3424108922481537,
      "learning_rate": 0.00017755102040816327,
      "loss": 0.086,
      "step": 13
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.5507439374923706,
      "learning_rate": 0.00017551020408163265,
      "loss": 0.082,
      "step": 14
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.40908390283584595,
      "learning_rate": 0.00017346938775510205,
      "loss": 0.0727,
      "step": 15
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.5264757871627808,
      "learning_rate": 0.00017142857142857143,
      "loss": 0.0651,
      "step": 16
    },
    {
      "epoch": 17.0,
      "grad_norm": 0.5899124145507812,
      "learning_rate": 0.00016938775510204083,
      "loss": 0.0557,
      "step": 17
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.5028952956199646,
      "learning_rate": 0.00016734693877551023,
      "loss": 0.042,
      "step": 18
    },
    {
      "epoch": 19.0,
      "grad_norm": 0.6451049447059631,
      "learning_rate": 0.0001653061224489796,
      "loss": 0.0315,
      "step": 19
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.5155397057533264,
      "learning_rate": 0.00016326530612244898,
      "loss": 0.017,
      "step": 20
    },
    {
      "epoch": 21.0,
      "grad_norm": 0.29147273302078247,
      "learning_rate": 0.00016122448979591838,
      "loss": 0.009,
      "step": 21
    },
    {
      "epoch": 22.0,
      "grad_norm": 0.3708033263683319,
      "learning_rate": 0.00015918367346938776,
      "loss": 0.0076,
      "step": 22
    },
    {
      "epoch": 23.0,
      "grad_norm": 0.08893364667892456,
      "learning_rate": 0.00015714285714285716,
      "loss": 0.0057,
      "step": 23
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.1690322905778885,
      "learning_rate": 0.00015510204081632654,
      "loss": 0.0062,
      "step": 24
    },
    {
      "epoch": 25.0,
      "grad_norm": 0.20433245599269867,
      "learning_rate": 0.0001530612244897959,
      "loss": 0.0058,
      "step": 25
    },
    {
      "epoch": 26.0,
      "grad_norm": 0.16720539331436157,
      "learning_rate": 0.0001510204081632653,
      "loss": 0.0056,
      "step": 26
    },
    {
      "epoch": 27.0,
      "grad_norm": 0.17153231799602509,
      "learning_rate": 0.00014897959183673472,
      "loss": 0.0056,
      "step": 27
    },
    {
      "epoch": 28.0,
      "grad_norm": 0.1037655845284462,
      "learning_rate": 0.0001469387755102041,
      "loss": 0.0059,
      "step": 28
    },
    {
      "epoch": 29.0,
      "grad_norm": 0.14174027740955353,
      "learning_rate": 0.0001448979591836735,
      "loss": 0.0055,
      "step": 29
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.011748074553906918,
      "learning_rate": 0.00014285714285714287,
      "loss": 0.0053,
      "step": 30
    },
    {
      "epoch": 31.0,
      "grad_norm": 0.09256725758314133,
      "learning_rate": 0.00014081632653061224,
      "loss": 0.0054,
      "step": 31
    },
    {
      "epoch": 32.0,
      "grad_norm": 0.09271395206451416,
      "learning_rate": 0.00013877551020408165,
      "loss": 0.0054,
      "step": 32
    },
    {
      "epoch": 33.0,
      "grad_norm": 0.10514255613088608,
      "learning_rate": 0.00013673469387755102,
      "loss": 0.0049,
      "step": 33
    },
    {
      "epoch": 34.0,
      "grad_norm": 0.14201119542121887,
      "learning_rate": 0.0001346938775510204,
      "loss": 0.0055,
      "step": 34
    },
    {
      "epoch": 35.0,
      "grad_norm": 0.05419391393661499,
      "learning_rate": 0.0001326530612244898,
      "loss": 0.0049,
      "step": 35
    },
    {
      "epoch": 36.0,
      "grad_norm": 0.17390480637550354,
      "learning_rate": 0.00013061224489795917,
      "loss": 0.0056,
      "step": 36
    },
    {
      "epoch": 37.0,
      "grad_norm": 0.01634000428020954,
      "learning_rate": 0.00012857142857142858,
      "loss": 0.0053,
      "step": 37
    },
    {
      "epoch": 38.0,
      "grad_norm": 0.14033198356628418,
      "learning_rate": 0.00012653061224489798,
      "loss": 0.0055,
      "step": 38
    },
    {
      "epoch": 39.0,
      "grad_norm": 0.06495083123445511,
      "learning_rate": 0.00012448979591836735,
      "loss": 0.0053,
      "step": 39
    },
    {
      "epoch": 40.0,
      "grad_norm": 0.014886317774653435,
      "learning_rate": 0.00012244897959183676,
      "loss": 0.0053,
      "step": 40
    },
    {
      "epoch": 41.0,
      "grad_norm": 0.09001646935939789,
      "learning_rate": 0.00012040816326530613,
      "loss": 0.0054,
      "step": 41
    },
    {
      "epoch": 42.0,
      "grad_norm": 0.05034321919083595,
      "learning_rate": 0.00011836734693877552,
      "loss": 0.0048,
      "step": 42
    },
    {
      "epoch": 43.0,
      "grad_norm": 0.012698125094175339,
      "learning_rate": 0.0001163265306122449,
      "loss": 0.0053,
      "step": 43
    },
    {
      "epoch": 44.0,
      "grad_norm": 0.06450632214546204,
      "learning_rate": 0.00011428571428571428,
      "loss": 0.0053,
      "step": 44
    },
    {
      "epoch": 45.0,
      "grad_norm": 0.06438387930393219,
      "learning_rate": 0.00011224489795918367,
      "loss": 0.0053,
      "step": 45
    },
    {
      "epoch": 46.0,
      "grad_norm": 0.010477419942617416,
      "learning_rate": 0.00011020408163265306,
      "loss": 0.0053,
      "step": 46
    },
    {
      "epoch": 47.0,
      "grad_norm": 0.010764668695628643,
      "learning_rate": 0.00010816326530612246,
      "loss": 0.0053,
      "step": 47
    },
    {
      "epoch": 48.0,
      "grad_norm": 0.010250415652990341,
      "learning_rate": 0.00010612244897959185,
      "loss": 0.0053,
      "step": 48
    },
    {
      "epoch": 49.0,
      "grad_norm": 0.01072185579687357,
      "learning_rate": 0.00010408163265306123,
      "loss": 0.0053,
      "step": 49
    },
    {
      "epoch": 50.0,
      "grad_norm": 0.04789392277598381,
      "learning_rate": 0.00010204081632653062,
      "loss": 0.0058,
      "step": 50
    },
    {
      "epoch": 51.0,
      "grad_norm": 0.046109408140182495,
      "learning_rate": 0.0001,
      "loss": 0.0048,
      "step": 51
    },
    {
      "epoch": 52.0,
      "grad_norm": 0.06478489935398102,
      "learning_rate": 9.79591836734694e-05,
      "loss": 0.0053,
      "step": 52
    },
    {
      "epoch": 53.0,
      "grad_norm": 0.06432071328163147,
      "learning_rate": 9.591836734693878e-05,
      "loss": 0.0053,
      "step": 53
    },
    {
      "epoch": 54.0,
      "grad_norm": 0.02735786698758602,
      "learning_rate": 9.387755102040817e-05,
      "loss": 0.0058,
      "step": 54
    },
    {
      "epoch": 55.0,
      "grad_norm": 0.04583168402314186,
      "learning_rate": 9.183673469387756e-05,
      "loss": 0.0048,
      "step": 55
    },
    {
      "epoch": 56.0,
      "grad_norm": 0.08271433413028717,
      "learning_rate": 8.979591836734695e-05,
      "loss": 0.0053,
      "step": 56
    },
    {
      "epoch": 57.0,
      "grad_norm": 0.010216895490884781,
      "learning_rate": 8.775510204081632e-05,
      "loss": 0.0053,
      "step": 57
    },
    {
      "epoch": 58.0,
      "grad_norm": 0.06349101662635803,
      "learning_rate": 8.571428571428571e-05,
      "loss": 0.0053,
      "step": 58
    },
    {
      "epoch": 59.0,
      "grad_norm": 0.06379684805870056,
      "learning_rate": 8.367346938775511e-05,
      "loss": 0.0053,
      "step": 59
    },
    {
      "epoch": 60.0,
      "grad_norm": 0.010050535202026367,
      "learning_rate": 8.163265306122449e-05,
      "loss": 0.0053,
      "step": 60
    },
    {
      "epoch": 61.0,
      "grad_norm": 0.08275067806243896,
      "learning_rate": 7.959183673469388e-05,
      "loss": 0.0053,
      "step": 61
    },
    {
      "epoch": 62.0,
      "grad_norm": 0.010069911368191242,
      "learning_rate": 7.755102040816327e-05,
      "loss": 0.0053,
      "step": 62
    },
    {
      "epoch": 63.0,
      "grad_norm": 0.00969759002327919,
      "learning_rate": 7.551020408163266e-05,
      "loss": 0.0053,
      "step": 63
    },
    {
      "epoch": 64.0,
      "grad_norm": 0.06372305005788803,
      "learning_rate": 7.346938775510205e-05,
      "loss": 0.0053,
      "step": 64
    },
    {
      "epoch": 65.0,
      "grad_norm": 0.027780011296272278,
      "learning_rate": 7.142857142857143e-05,
      "loss": 0.0048,
      "step": 65
    },
    {
      "epoch": 66.0,
      "grad_norm": 0.009718519635498524,
      "learning_rate": 6.938775510204082e-05,
      "loss": 0.0053,
      "step": 66
    },
    {
      "epoch": 67.0,
      "grad_norm": 0.04677029326558113,
      "learning_rate": 6.73469387755102e-05,
      "loss": 0.0058,
      "step": 67
    },
    {
      "epoch": 68.0,
      "grad_norm": 0.00952499732375145,
      "learning_rate": 6.530612244897959e-05,
      "loss": 0.0053,
      "step": 68
    },
    {
      "epoch": 69.0,
      "grad_norm": 0.009604902006685734,
      "learning_rate": 6.326530612244899e-05,
      "loss": 0.0053,
      "step": 69
    },
    {
      "epoch": 70.0,
      "grad_norm": 0.06420488655567169,
      "learning_rate": 6.122448979591838e-05,
      "loss": 0.0053,
      "step": 70
    },
    {
      "epoch": 71.0,
      "grad_norm": 0.026840027421712875,
      "learning_rate": 5.918367346938776e-05,
      "loss": 0.0058,
      "step": 71
    },
    {
      "epoch": 72.0,
      "grad_norm": 0.009567616507411003,
      "learning_rate": 5.714285714285714e-05,
      "loss": 0.0053,
      "step": 72
    },
    {
      "epoch": 73.0,
      "grad_norm": 0.045753173530101776,
      "learning_rate": 5.510204081632653e-05,
      "loss": 0.0048,
      "step": 73
    },
    {
      "epoch": 74.0,
      "grad_norm": 0.009476774372160435,
      "learning_rate": 5.3061224489795926e-05,
      "loss": 0.0053,
      "step": 74
    },
    {
      "epoch": 75.0,
      "grad_norm": 0.04571187496185303,
      "learning_rate": 5.102040816326531e-05,
      "loss": 0.0048,
      "step": 75
    },
    {
      "epoch": 76.0,
      "grad_norm": 0.009730060584843159,
      "learning_rate": 4.89795918367347e-05,
      "loss": 0.0053,
      "step": 76
    },
    {
      "epoch": 77.0,
      "grad_norm": 0.028138577938079834,
      "learning_rate": 4.6938775510204086e-05,
      "loss": 0.0048,
      "step": 77
    },
    {
      "epoch": 78.0,
      "grad_norm": 0.06447193771600723,
      "learning_rate": 4.4897959183673474e-05,
      "loss": 0.0053,
      "step": 78
    },
    {
      "epoch": 79.0,
      "grad_norm": 0.06447941064834595,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 0.0053,
      "step": 79
    },
    {
      "epoch": 80.0,
      "grad_norm": 0.028326738625764847,
      "learning_rate": 4.0816326530612245e-05,
      "loss": 0.0048,
      "step": 80
    },
    {
      "epoch": 81.0,
      "grad_norm": 0.08396949619054794,
      "learning_rate": 3.8775510204081634e-05,
      "loss": 0.0053,
      "step": 81
    },
    {
      "epoch": 82.0,
      "grad_norm": 0.04610452800989151,
      "learning_rate": 3.673469387755102e-05,
      "loss": 0.0048,
      "step": 82
    },
    {
      "epoch": 83.0,
      "grad_norm": 0.046097833663225174,
      "learning_rate": 3.469387755102041e-05,
      "loss": 0.0048,
      "step": 83
    },
    {
      "epoch": 84.0,
      "grad_norm": 0.009920083917677402,
      "learning_rate": 3.265306122448979e-05,
      "loss": 0.0053,
      "step": 84
    },
    {
      "epoch": 85.0,
      "grad_norm": 0.02849722094833851,
      "learning_rate": 3.061224489795919e-05,
      "loss": 0.0048,
      "step": 85
    },
    {
      "epoch": 86.0,
      "grad_norm": 0.00956336036324501,
      "learning_rate": 2.857142857142857e-05,
      "loss": 0.0053,
      "step": 86
    },
    {
      "epoch": 87.0,
      "grad_norm": 0.06530480086803436,
      "learning_rate": 2.6530612244897963e-05,
      "loss": 0.0053,
      "step": 87
    },
    {
      "epoch": 88.0,
      "grad_norm": 0.028555840253829956,
      "learning_rate": 2.448979591836735e-05,
      "loss": 0.0048,
      "step": 88
    },
    {
      "epoch": 89.0,
      "grad_norm": 0.009486482478678226,
      "learning_rate": 2.2448979591836737e-05,
      "loss": 0.0053,
      "step": 89
    },
    {
      "epoch": 90.0,
      "grad_norm": 0.028627220541238785,
      "learning_rate": 2.0408163265306123e-05,
      "loss": 0.0048,
      "step": 90
    },
    {
      "epoch": 91.0,
      "grad_norm": 0.009754066355526447,
      "learning_rate": 1.836734693877551e-05,
      "loss": 0.0053,
      "step": 91
    },
    {
      "epoch": 92.0,
      "grad_norm": 0.009760790504515171,
      "learning_rate": 1.6326530612244897e-05,
      "loss": 0.0053,
      "step": 92
    },
    {
      "epoch": 93.0,
      "grad_norm": 0.02760305069386959,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.0058,
      "step": 93
    },
    {
      "epoch": 94.0,
      "grad_norm": 0.009988558478653431,
      "learning_rate": 1.2244897959183674e-05,
      "loss": 0.0053,
      "step": 94
    },
    {
      "epoch": 95.0,
      "grad_norm": 0.04651060700416565,
      "learning_rate": 1.0204081632653061e-05,
      "loss": 0.0048,
      "step": 95
    },
    {
      "epoch": 96.0,
      "grad_norm": 0.010227691382169724,
      "learning_rate": 8.163265306122448e-06,
      "loss": 0.0053,
      "step": 96
    },
    {
      "epoch": 97.0,
      "grad_norm": 0.010372389107942581,
      "learning_rate": 6.122448979591837e-06,
      "loss": 0.0053,
      "step": 97
    },
    {
      "epoch": 98.0,
      "grad_norm": 0.009753258898854256,
      "learning_rate": 4.081632653061224e-06,
      "loss": 0.0053,
      "step": 98
    },
    {
      "epoch": 99.0,
      "grad_norm": 0.010054420679807663,
      "learning_rate": 2.040816326530612e-06,
      "loss": 0.0053,
      "step": 99
    },
    {
      "epoch": 100.0,
      "grad_norm": 0.01012208592146635,
      "learning_rate": 0.0,
      "loss": 0.0053,
      "step": 100
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "total_flos": 625210570752000.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}