{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.16,
  "eval_steps": 1000,
  "global_step": 8000,
  "is_hyper_param_search": false,
  "is_local_process_zero": false,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002,
      "grad_norm": 4.729869365692139,
      "learning_rate": 0.00033,
      "loss": 64.03521484375,
      "step": 100
    },
    {
      "epoch": 0.004,
      "grad_norm": 7.427474021911621,
      "learning_rate": 0.0004999988080137436,
      "loss": 50.4660400390625,
      "step": 200
    },
    {
      "epoch": 0.006,
      "grad_norm": 499.43450927734375,
      "learning_rate": 0.0004999889782951015,
      "loss": 55.620224609375,
      "step": 300
    },
    {
      "epoch": 0.008,
      "grad_norm": 5.543099880218506,
      "learning_rate": 0.0004999692199574857,
      "loss": 55.512568359375,
      "step": 400
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.913767099380493,
      "learning_rate": 0.0004999395337856224,
      "loss": 54.5164111328125,
      "step": 500
    },
    {
      "epoch": 0.012,
      "grad_norm": 3.7103848457336426,
      "learning_rate": 0.0004998999209585345,
      "loss": 48.3706103515625,
      "step": 600
    },
    {
      "epoch": 0.014,
      "grad_norm": 27.09828758239746,
      "learning_rate": 0.0004998503830494941,
      "loss": 48.7948681640625,
      "step": 700
    },
    {
      "epoch": 0.016,
      "grad_norm": 357.75579833984375,
      "learning_rate": 0.0004997909220259598,
      "loss": 47.264482421875,
      "step": 800
    },
    {
      "epoch": 0.018,
      "grad_norm": 21.32952117919922,
      "learning_rate": 0.0004997215402494993,
      "loss": 46.945810546875,
      "step": 900
    },
    {
      "epoch": 0.02,
      "grad_norm": 1339.14501953125,
      "learning_rate": 0.0004996422404756948,
      "loss": 50.6924609375,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "eval_loss": 62.66343307495117,
      "eval_runtime": 11.6524,
      "eval_samples_per_second": 1.716,
      "eval_steps_per_second": 0.086,
      "step": 1000
    },
    {
      "epoch": 0.022,
      "grad_norm": 29.14947509765625,
      "learning_rate": 0.0004995530258540343,
      "loss": 49.645908203125,
      "step": 1100
    },
    {
      "epoch": 0.024,
      "grad_norm": 199.71112060546875,
      "learning_rate": 0.0004994538999277859,
      "loss": 46.0685693359375,
      "step": 1200
    },
    {
      "epoch": 0.026,
      "grad_norm": 26.30093002319336,
      "learning_rate": 0.0004993448666338572,
      "loss": 45.418486328125,
      "step": 1300
    },
    {
      "epoch": 0.028,
      "grad_norm": 72.15017700195312,
      "learning_rate": 0.0004992259303026396,
      "loss": 46.6232666015625,
      "step": 1400
    },
    {
      "epoch": 0.03,
      "grad_norm": 16.536649703979492,
      "learning_rate": 0.0004990970956578351,
      "loss": 52.807451171875,
      "step": 1500
    },
    {
      "epoch": 0.032,
      "grad_norm": 75235.25,
      "learning_rate": 0.0004989583678162697,
      "loss": 58.359990234375,
      "step": 1600
    },
    {
      "epoch": 0.034,
      "grad_norm": 7981.73974609375,
      "learning_rate": 0.0004988097522876898,
      "loss": 58.0731103515625,
      "step": 1700
    },
    {
      "epoch": 0.036,
      "grad_norm": 10026.5595703125,
      "learning_rate": 0.0004986512549745438,
      "loss": 59.05083984375,
      "step": 1800
    },
    {
      "epoch": 0.038,
      "grad_norm": 534.1368408203125,
      "learning_rate": 0.0004984828821717466,
      "loss": 54.8443310546875,
      "step": 1900
    },
    {
      "epoch": 0.04,
      "grad_norm": 423.7434997558594,
      "learning_rate": 0.0004983046405664306,
      "loss": 54.6559326171875,
      "step": 2000
    },
    {
      "epoch": 0.04,
      "eval_loss": 55.864173889160156,
      "eval_runtime": 1.8383,
      "eval_samples_per_second": 10.88,
      "eval_steps_per_second": 0.544,
      "step": 2000
    },
    {
      "epoch": 0.042,
      "grad_norm": 7411.40234375,
      "learning_rate": 0.0004981165372376802,
      "loss": 58.2853564453125,
      "step": 2100
    },
    {
      "epoch": 0.044,
      "grad_norm": 33.24531173706055,
      "learning_rate": 0.0004979185796562494,
      "loss": 57.3351123046875,
      "step": 2200
    },
    {
      "epoch": 0.046,
      "grad_norm": 1044.70458984375,
      "learning_rate": 0.0004977107756842668,
      "loss": 55.583466796875,
      "step": 2300
    },
    {
      "epoch": 0.048,
      "grad_norm": 488.1427001953125,
      "learning_rate": 0.0004974931335749219,
      "loss": 53.566806640625,
      "step": 2400
    },
    {
      "epoch": 0.05,
      "grad_norm": 949.36181640625,
      "learning_rate": 0.000497265661972138,
      "loss": 52.38724609375,
      "step": 2500
    },
    {
      "epoch": 0.052,
      "grad_norm": 1916.361328125,
      "learning_rate": 0.0004970283699102291,
      "loss": 55.593505859375,
      "step": 2600
    },
    {
      "epoch": 0.054,
      "grad_norm": 40184.4609375,
      "learning_rate": 0.0004967812668135405,
      "loss": 59.1807861328125,
      "step": 2700
    },
    {
      "epoch": 0.056,
      "grad_norm": 1653.9197998046875,
      "learning_rate": 0.0004965243624960747,
      "loss": 60.0856396484375,
      "step": 2800
    },
    {
      "epoch": 0.058,
      "grad_norm": 570.0091552734375,
      "learning_rate": 0.0004962576671611021,
      "loss": 55.3298486328125,
      "step": 2900
    },
    {
      "epoch": 0.06,
      "grad_norm": 767297.5625,
      "learning_rate": 0.000495981191400755,
      "loss": 59.248984375,
      "step": 3000
    },
    {
      "epoch": 0.06,
      "eval_loss": 65.7476806640625,
      "eval_runtime": 1.8469,
      "eval_samples_per_second": 10.829,
      "eval_steps_per_second": 0.541,
      "step": 3000
    },
    {
      "epoch": 0.062,
      "grad_norm": 1796894.75,
      "learning_rate": 0.0004956949461956074,
      "loss": 61.4278955078125,
      "step": 3100
    },
    {
      "epoch": 0.064,
      "grad_norm": 372332.40625,
      "learning_rate": 0.0004953989429142387,
      "loss": 58.2212158203125,
      "step": 3200
    },
    {
      "epoch": 0.066,
      "grad_norm": 839582.25,
      "learning_rate": 0.0004950931933127826,
      "loss": 60.9759326171875,
      "step": 3300
    },
    {
      "epoch": 0.068,
      "grad_norm": 456507.9375,
      "learning_rate": 0.0004947777095344596,
      "loss": 60.5664697265625,
      "step": 3400
    },
    {
      "epoch": 0.07,
      "grad_norm": 42298.0859375,
      "learning_rate": 0.0004944525041090948,
      "loss": 59.9312353515625,
      "step": 3500
    },
    {
      "epoch": 0.072,
      "grad_norm": 44746.13671875,
      "learning_rate": 0.0004941175899526208,
      "loss": 60.0662353515625,
      "step": 3600
    },
    {
      "epoch": 0.074,
      "grad_norm": 6649.8818359375,
      "learning_rate": 0.0004937729803665643,
      "loss": 61.2509716796875,
      "step": 3700
    },
    {
      "epoch": 0.076,
      "grad_norm": 15284.4833984375,
      "learning_rate": 0.0004934186890375175,
      "loss": 59.5713232421875,
      "step": 3800
    },
    {
      "epoch": 0.078,
      "grad_norm": 13539.5712890625,
      "learning_rate": 0.0004930547300365956,
      "loss": 58.5672265625,
      "step": 3900
    },
    {
      "epoch": 0.08,
      "grad_norm": 53576.8984375,
      "learning_rate": 0.0004926811178188765,
      "loss": 58.3544189453125,
      "step": 4000
    },
    {
      "epoch": 0.08,
      "eval_loss": 58.579750061035156,
      "eval_runtime": 1.8442,
      "eval_samples_per_second": 10.845,
      "eval_steps_per_second": 0.542,
      "step": 4000
    },
    {
      "epoch": 0.082,
      "grad_norm": 160393.71875,
      "learning_rate": 0.000492297867222828,
      "loss": 59.3321435546875,
      "step": 4100
    },
    {
      "epoch": 0.084,
      "grad_norm": 12930.431640625,
      "learning_rate": 0.0004919049934697177,
      "loss": 60.8889404296875,
      "step": 4200
    },
    {
      "epoch": 0.086,
      "grad_norm": 4836.6787109375,
      "learning_rate": 0.0004915025121630086,
      "loss": 61.493935546875,
      "step": 4300
    },
    {
      "epoch": 0.088,
      "grad_norm": 595.083984375,
      "learning_rate": 0.0004910904392877396,
      "loss": 58.8386474609375,
      "step": 4400
    },
    {
      "epoch": 0.09,
      "grad_norm": 12180.2373046875,
      "learning_rate": 0.0004906687912098906,
      "loss": 63.4604150390625,
      "step": 4500
    },
    {
      "epoch": 0.092,
      "grad_norm": 414.4292297363281,
      "learning_rate": 0.0004902375846757322,
      "loss": 59.018310546875,
      "step": 4600
    },
    {
      "epoch": 0.094,
      "grad_norm": 2297.899169921875,
      "learning_rate": 0.0004897968368111611,
      "loss": 57.6424853515625,
      "step": 4700
    },
    {
      "epoch": 0.096,
      "grad_norm": 2821.689697265625,
      "learning_rate": 0.0004893465651210193,
      "loss": 58.3983154296875,
      "step": 4800
    },
    {
      "epoch": 0.098,
      "grad_norm": 2559.970947265625,
      "learning_rate": 0.0004888867874883995,
      "loss": 57.3033935546875,
      "step": 4900
    },
    {
      "epoch": 0.1,
      "grad_norm": 8444.46484375,
      "learning_rate": 0.0004884175221739343,
      "loss": 56.6698193359375,
      "step": 5000
    },
    {
      "epoch": 0.1,
      "eval_loss": 57.39847946166992,
      "eval_runtime": 1.8416,
      "eval_samples_per_second": 10.86,
      "eval_steps_per_second": 0.543,
      "step": 5000
    },
    {
      "epoch": 0.102,
      "grad_norm": 6454.49658203125,
      "learning_rate": 0.0004879387878150716,
      "loss": 56.9357958984375,
      "step": 5100
    },
    {
      "epoch": 0.104,
      "grad_norm": 337538.0,
      "learning_rate": 0.00048745060342533363,
      "loss": 56.982119140625,
      "step": 5200
    },
    {
      "epoch": 0.106,
      "grad_norm": 68615.4609375,
      "learning_rate": 0.0004869529883935625,
      "loss": 56.92587890625,
      "step": 5300
    },
    {
      "epoch": 0.108,
      "grad_norm": 116072.59375,
      "learning_rate": 0.00048644596248314967,
      "loss": 57.190751953125,
      "step": 5400
    },
    {
      "epoch": 0.11,
      "grad_norm": 291359.625,
      "learning_rate": 0.0004859295458312511,
      "loss": 58.6766162109375,
      "step": 5500
    },
    {
      "epoch": 0.112,
      "grad_norm": 89412.578125,
      "learning_rate": 0.0004854037589479878,
      "loss": 58.688193359375,
      "step": 5600
    },
    {
      "epoch": 0.114,
      "grad_norm": 293593.4375,
      "learning_rate": 0.0004848686227156309,
      "loss": 59.063056640625,
      "step": 5700
    },
    {
      "epoch": 0.116,
      "grad_norm": 30479.330078125,
      "learning_rate": 0.0004843241583877724,
      "loss": 58.46580078125,
      "step": 5800
    },
    {
      "epoch": 0.118,
      "grad_norm": 89906.7734375,
      "learning_rate": 0.000483770387588481,
      "loss": 58.2409228515625,
      "step": 5900
    },
    {
      "epoch": 0.12,
      "grad_norm": 630418.5,
      "learning_rate": 0.00048320733231144354,
      "loss": 58.333916015625,
      "step": 6000
    },
    {
      "epoch": 0.12,
      "eval_loss": 59.336669921875,
      "eval_runtime": 1.8507,
      "eval_samples_per_second": 10.807,
      "eval_steps_per_second": 0.54,
      "step": 6000
    },
    {
      "epoch": 0.122,
      "grad_norm": 462631.5,
      "learning_rate": 0.000482635014919091,
      "loss": 59.005234375,
      "step": 6100
    },
    {
      "epoch": 0.124,
      "grad_norm": 273326.59375,
      "learning_rate": 0.00048205345814171074,
      "loss": 58.9213232421875,
      "step": 6200
    },
    {
      "epoch": 0.126,
      "grad_norm": 36951.60546875,
      "learning_rate": 0.00048146268507654377,
      "loss": 58.82025390625,
      "step": 6300
    },
    {
      "epoch": 0.128,
      "grad_norm": 103255.4765625,
      "learning_rate": 0.00048086271918686716,
      "loss": 58.84619140625,
      "step": 6400
    },
    {
      "epoch": 0.13,
      "grad_norm": 102904.96875,
      "learning_rate": 0.00048025358430106227,
      "loss": 58.514482421875,
      "step": 6500
    },
    {
      "epoch": 0.132,
      "grad_norm": 368401.625,
      "learning_rate": 0.00047963530461166826,
      "loss": 58.3533984375,
      "step": 6600
    },
    {
      "epoch": 0.134,
      "grad_norm": 307524.84375,
      "learning_rate": 0.0004790079046744218,
      "loss": 58.44912109375,
      "step": 6700
    },
    {
      "epoch": 0.136,
      "grad_norm": 218012.390625,
      "learning_rate": 0.000478371409407281,
      "loss": 58.45431640625,
      "step": 6800
    },
    {
      "epoch": 0.138,
      "grad_norm": 1243047.5,
      "learning_rate": 0.0004777258440894362,
      "loss": 58.24828125,
      "step": 6900
    },
    {
      "epoch": 0.14,
      "grad_norm": 158677.6875,
      "learning_rate": 0.0004770712343603062,
      "loss": 58.133662109375,
      "step": 7000
    },
    {
      "epoch": 0.14,
      "eval_loss": 58.2567138671875,
      "eval_runtime": 1.8382,
      "eval_samples_per_second": 10.88,
      "eval_steps_per_second": 0.544,
      "step": 7000
    },
    {
      "epoch": 0.142,
      "grad_norm": 237161.1875,
      "learning_rate": 0.0004764076062185194,
      "loss": 58.3987744140625,
      "step": 7100
    },
    {
      "epoch": 0.144,
      "grad_norm": 60140.4921875,
      "learning_rate": 0.00047573498602088154,
      "loss": 58.801220703125,
      "step": 7200
    },
    {
      "epoch": 0.146,
      "grad_norm": 494529.25,
      "learning_rate": 0.00047505340048132916,
      "loss": 58.503544921875,
      "step": 7300
    },
    {
      "epoch": 0.148,
      "grad_norm": 561465.625,
      "learning_rate": 0.00047436287666986803,
      "loss": 58.614169921875,
      "step": 7400
    },
    {
      "epoch": 0.15,
      "grad_norm": 572518.4375,
      "learning_rate": 0.00047366344201149856,
      "loss": 58.28416015625,
      "step": 7500
    },
    {
      "epoch": 0.152,
      "grad_norm": 1078630.125,
      "learning_rate": 0.0004729551242851264,
      "loss": 58.4676513671875,
      "step": 7600
    },
    {
      "epoch": 0.154,
      "grad_norm": 70299.53125,
      "learning_rate": 0.00047223795162245886,
      "loss": 58.200947265625,
      "step": 7700
    },
    {
      "epoch": 0.156,
      "grad_norm": 68182.265625,
      "learning_rate": 0.0004715119525068883,
      "loss": 58.55353515625,
      "step": 7800
    },
    {
      "epoch": 0.158,
      "grad_norm": 82698.640625,
      "learning_rate": 0.00047077715577236015,
      "loss": 58.704990234375,
      "step": 7900
    },
    {
      "epoch": 0.16,
      "grad_norm": 375052.03125,
      "learning_rate": 0.0004700335906022283,
      "loss": 58.521640625,
      "step": 8000
    },
    {
      "epoch": 0.16,
      "eval_loss": 57.7498779296875,
      "eval_runtime": 1.8363,
      "eval_samples_per_second": 10.892,
      "eval_steps_per_second": 0.545,
      "step": 8000
    }
  ],
  "logging_steps": 100,
  "max_steps": 50000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 2000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.699606151168e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}