{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 16,
  "global_step": 84,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.047619047619047616,
      "grad_norm": 0.6781264543533325,
      "learning_rate": 0.0,
      "loss": 1.26171875,
      "step": 1
    },
    {
      "epoch": 0.09523809523809523,
      "grad_norm": 0.6318029761314392,
      "learning_rate": 2e-05,
      "loss": 1.20166015625,
      "step": 2
    },
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 0.7097207903862,
      "learning_rate": 4e-05,
      "loss": 1.412139892578125,
      "step": 3
    },
    {
      "epoch": 0.19047619047619047,
      "grad_norm": 0.622149646282196,
      "learning_rate": 6e-05,
      "loss": 1.25537109375,
      "step": 4
    },
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 2.2328977584838867,
      "learning_rate": 8e-05,
      "loss": 1.1912841796875,
      "step": 5
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 0.3143213987350464,
      "learning_rate": 0.0001,
      "loss": 1.228759765625,
      "step": 6
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.2846517860889435,
      "learning_rate": 9.996442287522857e-05,
      "loss": 1.09326171875,
      "step": 7
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 0.3680817186832428,
      "learning_rate": 9.985774775566131e-05,
      "loss": 1.0126953125,
      "step": 8
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 0.5873160362243652,
      "learning_rate": 9.96801433165888e-05,
      "loss": 1.241455078125,
      "step": 9
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 0.3845677077770233,
      "learning_rate": 9.943189038713501e-05,
      "loss": 1.058349609375,
      "step": 10
    },
    {
      "epoch": 0.5238095238095238,
      "grad_norm": 0.262990802526474,
      "learning_rate": 9.911338150620863e-05,
      "loss": 1.056396484375,
      "step": 11
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 0.2172330915927887,
      "learning_rate": 9.872512030181845e-05,
      "loss": 1.0838623046875,
      "step": 12
    },
    {
      "epoch": 0.6190476190476191,
      "grad_norm": 0.23545600473880768,
      "learning_rate": 9.826772069473393e-05,
      "loss": 1.056396484375,
      "step": 13
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.18727780878543854,
      "learning_rate": 9.774190592775023e-05,
      "loss": 1.0078125,
      "step": 14
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 0.24253623187541962,
      "learning_rate": 9.714850742209276e-05,
      "loss": 1.090576171875,
      "step": 15
    },
    {
      "epoch": 0.7619047619047619,
      "grad_norm": 0.13259640336036682,
      "learning_rate": 9.648846346276929e-05,
      "loss": 0.7398681640625,
      "step": 16
    },
    {
      "epoch": 0.8095238095238095,
      "grad_norm": 0.20533913373947144,
      "learning_rate": 9.576281771494855e-05,
      "loss": 1.141845703125,
      "step": 17
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 0.18200747668743134,
      "learning_rate": 9.497271757371094e-05,
      "loss": 1.017333984375,
      "step": 18
    },
    {
      "epoch": 0.9047619047619048,
      "grad_norm": 0.19662703573703766,
      "learning_rate": 9.411941234978122e-05,
      "loss": 1.0162353515625,
      "step": 19
    },
    {
      "epoch": 0.9523809523809523,
      "grad_norm": 0.19710412621498108,
      "learning_rate": 9.320425129411139e-05,
      "loss": 1.0875244140625,
      "step": 20
    },
    {
      "epoch": 0.9523809523809523,
      "eval_loss": 0.916015625,
      "eval_runtime": 6.1898,
      "eval_samples_per_second": 0.646,
      "eval_steps_per_second": 0.162,
      "step": 20
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.13886402547359467,
      "learning_rate": 9.222868146443759e-05,
      "loss": 0.9620361328125,
      "step": 21
    },
    {
      "epoch": 1.0476190476190477,
      "grad_norm": 0.1303921639919281,
      "learning_rate": 9.119424543718442e-05,
      "loss": 0.9793701171875,
      "step": 22
    },
    {
      "epoch": 1.0952380952380953,
      "grad_norm": 0.14203043282032013,
      "learning_rate": 9.010257886833456e-05,
      "loss": 0.936279296875,
      "step": 23
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 0.19456878304481506,
      "learning_rate": 8.895540790712036e-05,
      "loss": 1.0773696899414062,
      "step": 24
    },
    {
      "epoch": 1.1904761904761905,
      "grad_norm": 0.19268886744976044,
      "learning_rate": 8.775454646662724e-05,
      "loss": 0.962890625,
      "step": 25
    },
    {
      "epoch": 1.2380952380952381,
      "grad_norm": 1.1448683738708496,
      "learning_rate": 8.650189335562414e-05,
      "loss": 1.0135498046875,
      "step": 26
    },
    {
      "epoch": 1.2857142857142856,
      "grad_norm": 0.17032912373542786,
      "learning_rate": 8.519942927615666e-05,
      "loss": 1.041259765625,
      "step": 27
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.16223372519016266,
      "learning_rate": 8.384921369164991e-05,
      "loss": 0.947998046875,
      "step": 28
    },
    {
      "epoch": 1.380952380952381,
      "grad_norm": 0.13596029579639435,
      "learning_rate": 8.24533815704736e-05,
      "loss": 0.8992919921875,
      "step": 29
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 0.19097673892974854,
      "learning_rate": 8.101414001011817e-05,
      "loss": 1.0830078125,
      "step": 30
    },
    {
      "epoch": 1.4761904761904763,
      "grad_norm": 0.1492234319448471,
      "learning_rate": 7.953376474732003e-05,
      "loss": 0.9466552734375,
      "step": 31
    },
    {
      "epoch": 1.5238095238095237,
      "grad_norm": 0.14599502086639404,
      "learning_rate": 7.801459655965401e-05,
      "loss": 0.957763671875,
      "step": 32
    },
    {
      "epoch": 1.5714285714285714,
      "grad_norm": 0.13878127932548523,
      "learning_rate": 7.645903756428279e-05,
      "loss": 0.9761962890625,
      "step": 33
    },
    {
      "epoch": 1.619047619047619,
      "grad_norm": 0.15830132365226746,
      "learning_rate": 7.486954741971603e-05,
      "loss": 0.957275390625,
      "step": 34
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.13619418442249298,
      "learning_rate": 7.324863943658474e-05,
      "loss": 0.92626953125,
      "step": 35
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 0.20511189103126526,
      "learning_rate": 7.159887660358036e-05,
      "loss": 0.980224609375,
      "step": 36
    },
    {
      "epoch": 1.7142857142857144,
      "eval_loss": 0.8896484375,
      "eval_runtime": 5.9036,
      "eval_samples_per_second": 0.678,
      "eval_steps_per_second": 0.169,
      "step": 36
    },
    {
      "epoch": 1.7619047619047619,
      "grad_norm": 0.11510299891233444,
      "learning_rate": 6.992286753484306e-05,
      "loss": 0.6959228515625,
      "step": 37
    },
    {
      "epoch": 1.8095238095238095,
      "grad_norm": 0.17241714894771576,
      "learning_rate": 6.822326234520645e-05,
      "loss": 1.042236328125,
      "step": 38
    },
    {
      "epoch": 1.8571428571428572,
      "grad_norm": 0.14825452864170074,
      "learning_rate": 6.650274845982136e-05,
      "loss": 0.93603515625,
      "step": 39
    },
    {
      "epoch": 1.9047619047619047,
      "grad_norm": 0.1515885442495346,
      "learning_rate": 6.476404636478432e-05,
      "loss": 0.948974609375,
      "step": 40
    },
    {
      "epoch": 1.9523809523809523,
      "grad_norm": 0.16863621771335602,
      "learning_rate": 6.300990530548979e-05,
      "loss": 0.9949951171875,
      "step": 41
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.4026670455932617,
      "learning_rate": 6.124309893950809e-05,
      "loss": 0.869873046875,
      "step": 42
    },
    {
      "epoch": 2.0476190476190474,
      "grad_norm": 0.16099941730499268,
      "learning_rate": 5.946642095086278e-05,
      "loss": 0.8929443359375,
      "step": 43
    },
    {
      "epoch": 2.0952380952380953,
      "grad_norm": 0.13304688036441803,
      "learning_rate": 5.7682680632641896e-05,
      "loss": 0.873046875,
      "step": 44
    },
    {
      "epoch": 2.142857142857143,
      "grad_norm": 0.18484529852867126,
      "learning_rate": 5.589469844492816e-05,
      "loss": 0.9822769165039062,
      "step": 45
    },
    {
      "epoch": 2.1904761904761907,
      "grad_norm": 0.17451068758964539,
      "learning_rate": 5.410530155507184e-05,
      "loss": 0.8817138671875,
      "step": 46
    },
    {
      "epoch": 2.238095238095238,
      "grad_norm": 0.17454630136489868,
      "learning_rate": 5.231731936735811e-05,
      "loss": 0.9676513671875,
      "step": 47
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 0.19139355421066284,
      "learning_rate": 5.053357904913724e-05,
      "loss": 0.9560546875,
      "step": 48
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.18285448849201202,
      "learning_rate": 4.875690106049193e-05,
      "loss": 0.8807373046875,
      "step": 49
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 0.16837579011917114,
      "learning_rate": 4.6990094694510245e-05,
      "loss": 0.8505859375,
      "step": 50
    },
    {
      "epoch": 2.4285714285714284,
      "grad_norm": 0.21485020220279694,
      "learning_rate": 4.5235953635215686e-05,
      "loss": 0.994140625,
      "step": 51
    },
    {
      "epoch": 2.4761904761904763,
      "grad_norm": 0.1551942676305771,
      "learning_rate": 4.349725154017864e-05,
      "loss": 0.88671875,
      "step": 52
    },
    {
      "epoch": 2.4761904761904763,
      "eval_loss": 0.876953125,
      "eval_runtime": 5.9086,
      "eval_samples_per_second": 0.677,
      "eval_steps_per_second": 0.169,
      "step": 52
    },
    {
      "epoch": 2.5238095238095237,
      "grad_norm": 0.16076764464378357,
      "learning_rate": 4.177673765479357e-05,
      "loss": 0.9061279296875,
      "step": 53
    },
    {
      "epoch": 2.571428571428571,
      "grad_norm": 0.16922174394130707,
      "learning_rate": 4.007713246515695e-05,
      "loss": 0.9171142578125,
      "step": 54
    },
    {
      "epoch": 2.619047619047619,
      "grad_norm": 0.18479417264461517,
      "learning_rate": 3.8401123396419656e-05,
      "loss": 0.896240234375,
      "step": 55
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.15129074454307556,
      "learning_rate": 3.675136056341528e-05,
      "loss": 0.880126953125,
      "step": 56
    },
    {
      "epoch": 2.7142857142857144,
      "grad_norm": 0.23219619691371918,
      "learning_rate": 3.513045258028399e-05,
      "loss": 0.900390625,
      "step": 57
    },
    {
      "epoch": 2.761904761904762,
      "grad_norm": 0.11648663133382797,
      "learning_rate": 3.354096243571724e-05,
      "loss": 0.6685791015625,
      "step": 58
    },
    {
      "epoch": 2.8095238095238093,
      "grad_norm": 0.20129463076591492,
      "learning_rate": 3.1985403440346004e-05,
      "loss": 0.97607421875,
      "step": 59
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 0.17487110197544098,
      "learning_rate": 3.0466235252679976e-05,
      "loss": 0.8857421875,
      "step": 60
    },
    {
      "epoch": 2.9047619047619047,
      "grad_norm": 0.18634696304798126,
      "learning_rate": 2.898585998988182e-05,
      "loss": 0.904296875,
      "step": 61
    },
    {
      "epoch": 2.9523809523809526,
      "grad_norm": 0.21948011219501495,
      "learning_rate": 2.7546618429526404e-05,
      "loss": 0.935302734375,
      "step": 62
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.17362090945243835,
      "learning_rate": 2.6150786308350112e-05,
      "loss": 0.8018798828125,
      "step": 63
    },
    {
      "epoch": 3.0476190476190474,
      "grad_norm": 0.16069509088993073,
      "learning_rate": 2.480057072384334e-05,
      "loss": 0.8336181640625,
      "step": 64
    },
    {
      "epoch": 3.0952380952380953,
      "grad_norm": 0.1560245156288147,
      "learning_rate": 2.3498106644375867e-05,
      "loss": 0.8359375,
      "step": 65
    },
    {
      "epoch": 3.142857142857143,
      "grad_norm": 0.21654845774173737,
      "learning_rate": 2.2245453533372768e-05,
      "loss": 0.9223861694335938,
      "step": 66
    },
    {
      "epoch": 3.1904761904761907,
      "grad_norm": 0.20104853808879852,
      "learning_rate": 2.1044592092879643e-05,
      "loss": 0.82958984375,
      "step": 67
    },
    {
      "epoch": 3.238095238095238,
      "grad_norm": 0.19536712765693665,
      "learning_rate": 1.9897421131665462e-05,
      "loss": 0.943115234375,
      "step": 68
    },
    {
      "epoch": 3.238095238095238,
      "eval_loss": 0.875,
      "eval_runtime": 5.8891,
      "eval_samples_per_second": 0.679,
      "eval_steps_per_second": 0.17,
      "step": 68
    },
    {
      "epoch": 3.2857142857142856,
      "grad_norm": 0.21967998147010803,
      "learning_rate": 1.8805754562815585e-05,
      "loss": 0.899658203125,
      "step": 69
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.19596749544143677,
      "learning_rate": 1.7771318535562435e-05,
      "loss": 0.834228515625,
      "step": 70
    },
    {
      "epoch": 3.380952380952381,
      "grad_norm": 0.16757430136203766,
      "learning_rate": 1.679574870588862e-05,
      "loss": 0.82275390625,
      "step": 71
    },
    {
      "epoch": 3.4285714285714284,
      "grad_norm": 0.2443944215774536,
      "learning_rate": 1.588058765021879e-05,
      "loss": 0.936767578125,
      "step": 72
    },
    {
      "epoch": 3.4761904761904763,
      "grad_norm": 0.18178075551986694,
      "learning_rate": 1.5027282426289077e-05,
      "loss": 0.85498046875,
      "step": 73
    },
    {
      "epoch": 3.5238095238095237,
      "grad_norm": 0.1797463297843933,
      "learning_rate": 1.4237182285051453e-05,
      "loss": 0.8779296875,
      "step": 74
    },
    {
      "epoch": 3.571428571428571,
      "grad_norm": 0.19107820093631744,
      "learning_rate": 1.3511536537230707e-05,
      "loss": 0.8829345703125,
      "step": 75
    },
    {
      "epoch": 3.619047619047619,
      "grad_norm": 0.1874479353427887,
      "learning_rate": 1.2851492577907251e-05,
      "loss": 0.8621826171875,
      "step": 76
    },
    {
      "epoch": 3.6666666666666665,
      "grad_norm": 0.16661852598190308,
      "learning_rate": 1.225809407224978e-05,
      "loss": 0.855712890625,
      "step": 77
    },
    {
      "epoch": 3.7142857142857144,
      "grad_norm": 0.2627502381801605,
      "learning_rate": 1.1732279305266082e-05,
      "loss": 0.8544921875,
      "step": 78
    },
    {
      "epoch": 3.761904761904762,
      "grad_norm": 0.12383028864860535,
      "learning_rate": 1.1274879698181547e-05,
      "loss": 0.6553955078125,
      "step": 79
    },
    {
      "epoch": 3.8095238095238093,
      "grad_norm": 0.21695609390735626,
      "learning_rate": 1.0886618493791376e-05,
      "loss": 0.940673828125,
      "step": 80
    },
    {
      "epoch": 3.857142857142857,
      "grad_norm": 0.18359997868537903,
      "learning_rate": 1.0568109612865e-05,
      "loss": 0.85888671875,
      "step": 81
    },
    {
      "epoch": 3.9047619047619047,
      "grad_norm": 0.19658944010734558,
      "learning_rate": 1.0319856683411197e-05,
      "loss": 0.8824462890625,
      "step": 82
    },
    {
      "epoch": 3.9523809523809526,
      "grad_norm": 0.24310623109340668,
      "learning_rate": 1.0142252244338688e-05,
      "loss": 0.905029296875,
      "step": 83
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.2020317167043686,
      "learning_rate": 1.0035577124771419e-05,
      "loss": 0.7667236328125,
      "step": 84
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.8740234375,
      "eval_runtime": 5.901,
      "eval_samples_per_second": 0.678,
      "eval_steps_per_second": 0.169,
      "step": 84
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 84,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3441704712428585e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}