{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 100.0,
  "eval_steps": 16,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2909090909090909,
      "grad_norm": 10.451888084411621,
      "learning_rate": 0.0,
      "loss": 0.7859,
      "step": 1
    },
    {
      "epoch": 0.5818181818181818,
      "grad_norm": 9.886292457580566,
      "learning_rate": 2.5e-09,
      "loss": 0.7965,
      "step": 2
    },
    {
      "epoch": 0.8727272727272727,
      "grad_norm": 10.406240463256836,
      "learning_rate": 5e-09,
      "loss": 0.7893,
      "step": 3
    },
    {
      "epoch": 1.0,
      "grad_norm": 12.746437072753906,
      "learning_rate": 7.5e-09,
      "loss": 0.8702,
      "step": 4
    },
    {
      "epoch": 1.290909090909091,
      "grad_norm": 10.291970252990723,
      "learning_rate": 1e-08,
      "loss": 0.7785,
      "step": 5
    },
    {
      "epoch": 1.5818181818181818,
      "grad_norm": 9.746875762939453,
      "learning_rate": 1.25e-08,
      "loss": 0.759,
      "step": 6
    },
    {
      "epoch": 1.8727272727272726,
      "grad_norm": 10.920265197753906,
      "learning_rate": 1.5e-08,
      "loss": 0.8308,
      "step": 7
    },
    {
      "epoch": 2.0,
      "grad_norm": 11.265154838562012,
      "learning_rate": 1.7499999999999998e-08,
      "loss": 0.8828,
      "step": 8
    },
    {
      "epoch": 2.290909090909091,
      "grad_norm": 9.750505447387695,
      "learning_rate": 2e-08,
      "loss": 0.8004,
      "step": 9
    },
    {
      "epoch": 2.581818181818182,
      "grad_norm": 11.47065544128418,
      "learning_rate": 2.25e-08,
      "loss": 0.8204,
      "step": 10
    },
    {
      "epoch": 2.8727272727272726,
      "grad_norm": 10.275605201721191,
      "learning_rate": 2.5e-08,
      "loss": 0.7771,
      "step": 11
    },
    {
      "epoch": 3.0,
      "grad_norm": 11.604477882385254,
      "learning_rate": 2.75e-08,
      "loss": 0.8295,
      "step": 12
    },
    {
      "epoch": 3.290909090909091,
      "grad_norm": 10.712018966674805,
      "learning_rate": 3e-08,
      "loss": 0.8378,
      "step": 13
    },
    {
      "epoch": 3.581818181818182,
      "grad_norm": 10.54987907409668,
      "learning_rate": 3.25e-08,
      "loss": 0.8398,
      "step": 14
    },
    {
      "epoch": 3.8727272727272726,
      "grad_norm": 9.999624252319336,
      "learning_rate": 3.4999999999999996e-08,
      "loss": 0.773,
      "step": 15
    },
    {
      "epoch": 4.0,
      "grad_norm": 10.562870979309082,
      "learning_rate": 3.75e-08,
      "loss": 0.7025,
      "step": 16
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.760595977306366,
      "eval_runtime": 0.722,
      "eval_samples_per_second": 18.005,
      "eval_steps_per_second": 18.005,
      "step": 16
    },
    {
      "epoch": 4.290909090909091,
      "grad_norm": 9.95614242553711,
      "learning_rate": 4e-08,
      "loss": 0.7785,
      "step": 17
    },
    {
      "epoch": 4.581818181818182,
      "grad_norm": 11.356291770935059,
      "learning_rate": 4.25e-08,
      "loss": 0.8645,
      "step": 18
    },
    {
      "epoch": 4.872727272727273,
      "grad_norm": 10.108142852783203,
      "learning_rate": 4.5e-08,
      "loss": 0.7834,
      "step": 19
    },
    {
      "epoch": 5.0,
      "grad_norm": 10.209877014160156,
      "learning_rate": 4.7499999999999995e-08,
      "loss": 0.7744,
      "step": 20
    },
    {
      "epoch": 5.290909090909091,
      "grad_norm": 9.586356163024902,
      "learning_rate": 5e-08,
      "loss": 0.7433,
      "step": 21
    },
    {
      "epoch": 5.581818181818182,
      "grad_norm": 10.589778900146484,
      "learning_rate": 5.25e-08,
      "loss": 0.818,
      "step": 22
    },
    {
      "epoch": 5.872727272727273,
      "grad_norm": 10.28813362121582,
      "learning_rate": 5.5e-08,
      "loss": 0.8245,
      "step": 23
    },
    {
      "epoch": 6.0,
      "grad_norm": 13.027183532714844,
      "learning_rate": 5.749999999999999e-08,
      "loss": 0.8331,
      "step": 24
    },
    {
      "epoch": 6.290909090909091,
      "grad_norm": 10.363873481750488,
      "learning_rate": 6e-08,
      "loss": 0.8331,
      "step": 25
    },
    {
      "epoch": 6.581818181818182,
      "grad_norm": 9.84264850616455,
      "learning_rate": 6.25e-08,
      "loss": 0.755,
      "step": 26
    },
    {
      "epoch": 6.872727272727273,
      "grad_norm": 10.973934173583984,
      "learning_rate": 6.5e-08,
      "loss": 0.8372,
      "step": 27
    },
    {
      "epoch": 7.0,
      "grad_norm": 10.278410911560059,
      "learning_rate": 6.75e-08,
      "loss": 0.7442,
      "step": 28
    },
    {
      "epoch": 7.290909090909091,
      "grad_norm": 10.205405235290527,
      "learning_rate": 6.999999999999999e-08,
      "loss": 0.7851,
      "step": 29
    },
    {
      "epoch": 7.581818181818182,
      "grad_norm": 10.862798690795898,
      "learning_rate": 7.25e-08,
      "loss": 0.7962,
      "step": 30
    },
    {
      "epoch": 7.872727272727273,
      "grad_norm": 9.971634864807129,
      "learning_rate": 7.5e-08,
      "loss": 0.79,
      "step": 31
    },
    {
      "epoch": 8.0,
      "grad_norm": 10.8460111618042,
      "learning_rate": 7.75e-08,
      "loss": 0.9105,
      "step": 32
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.7589532136917114,
      "eval_runtime": 0.7554,
      "eval_samples_per_second": 17.21,
      "eval_steps_per_second": 17.21,
      "step": 32
    },
    {
      "epoch": 8.290909090909091,
      "grad_norm": 10.4276704788208,
      "learning_rate": 8e-08,
      "loss": 0.7914,
      "step": 33
    },
    {
      "epoch": 8.581818181818182,
      "grad_norm": 9.807103157043457,
      "learning_rate": 8.249999999999999e-08,
      "loss": 0.8344,
      "step": 34
    },
    {
      "epoch": 8.872727272727273,
      "grad_norm": 9.850166320800781,
      "learning_rate": 8.5e-08,
      "loss": 0.7419,
      "step": 35
    },
    {
      "epoch": 9.0,
      "grad_norm": 12.547399520874023,
      "learning_rate": 8.75e-08,
      "loss": 0.8597,
      "step": 36
    },
    {
      "epoch": 9.290909090909091,
      "grad_norm": 10.39106559753418,
      "learning_rate": 9e-08,
      "loss": 0.7911,
      "step": 37
    },
    {
      "epoch": 9.581818181818182,
      "grad_norm": 10.728227615356445,
      "learning_rate": 9.25e-08,
      "loss": 0.852,
      "step": 38
    },
    {
      "epoch": 9.872727272727273,
      "grad_norm": 10.104507446289062,
      "learning_rate": 9.499999999999999e-08,
      "loss": 0.7942,
      "step": 39
    },
    {
      "epoch": 10.0,
      "grad_norm": 9.163139343261719,
      "learning_rate": 9.749999999999999e-08,
      "loss": 0.7006,
      "step": 40
    },
    {
      "epoch": 10.290909090909091,
      "grad_norm": 9.795455932617188,
      "learning_rate": 1e-07,
      "loss": 0.7496,
      "step": 41
    },
    {
      "epoch": 10.581818181818182,
      "grad_norm": 9.88698959350586,
      "learning_rate": 9.99982865378877e-08,
      "loss": 0.7978,
      "step": 42
    },
    {
      "epoch": 10.872727272727273,
      "grad_norm": 10.60831069946289,
      "learning_rate": 9.99931462820376e-08,
      "loss": 0.8437,
      "step": 43
    },
    {
      "epoch": 11.0,
      "grad_norm": 10.177803039550781,
      "learning_rate": 9.998457962390006e-08,
      "loss": 0.7926,
      "step": 44
    },
    {
      "epoch": 11.290909090909091,
      "grad_norm": 9.594599723815918,
      "learning_rate": 9.997258721585931e-08,
      "loss": 0.7521,
      "step": 45
    },
    {
      "epoch": 11.581818181818182,
      "grad_norm": 9.713711738586426,
      "learning_rate": 9.99571699711836e-08,
      "loss": 0.7497,
      "step": 46
    },
    {
      "epoch": 11.872727272727273,
      "grad_norm": 10.672869682312012,
      "learning_rate": 9.993832906395581e-08,
      "loss": 0.8709,
      "step": 47
    },
    {
      "epoch": 12.0,
      "grad_norm": 10.758075714111328,
      "learning_rate": 9.991606592898401e-08,
      "loss": 0.8193,
      "step": 48
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.7549822926521301,
      "eval_runtime": 0.7468,
      "eval_samples_per_second": 17.407,
      "eval_steps_per_second": 17.407,
      "step": 48
    },
    {
      "epoch": 12.290909090909091,
      "grad_norm": 10.45877742767334,
      "learning_rate": 9.989038226169209e-08,
      "loss": 0.8488,
      "step": 49
    },
    {
      "epoch": 12.581818181818182,
      "grad_norm": 9.969883918762207,
      "learning_rate": 9.986128001799076e-08,
      "loss": 0.7697,
      "step": 50
    },
    {
      "epoch": 12.872727272727273,
      "grad_norm": 9.769625663757324,
      "learning_rate": 9.982876141412855e-08,
      "loss": 0.769,
      "step": 51
    },
    {
      "epoch": 13.0,
      "grad_norm": 10.126288414001465,
      "learning_rate": 9.979282892652304e-08,
      "loss": 0.8111,
      "step": 52
    },
    {
      "epoch": 13.290909090909091,
      "grad_norm": 10.078240394592285,
      "learning_rate": 9.975348529157229e-08,
      "loss": 0.798,
      "step": 53
    },
    {
      "epoch": 13.581818181818182,
      "grad_norm": 10.470067024230957,
      "learning_rate": 9.971073350544643e-08,
      "loss": 0.8011,
      "step": 54
    },
    {
      "epoch": 13.872727272727273,
      "grad_norm": 9.311027526855469,
      "learning_rate": 9.966457682385949e-08,
      "loss": 0.7109,
      "step": 55
    },
    {
      "epoch": 14.0,
      "grad_norm": 10.48747730255127,
      "learning_rate": 9.961501876182147e-08,
      "loss": 0.969,
      "step": 56
    },
    {
      "epoch": 14.290909090909091,
      "grad_norm": 9.95235538482666,
      "learning_rate": 9.956206309337066e-08,
      "loss": 0.7757,
      "step": 57
    },
    {
      "epoch": 14.581818181818182,
      "grad_norm": 9.8017578125,
      "learning_rate": 9.950571385128625e-08,
      "loss": 0.7974,
      "step": 58
    },
    {
      "epoch": 14.872727272727273,
      "grad_norm": 10.287720680236816,
      "learning_rate": 9.94459753267812e-08,
      "loss": 0.842,
      "step": 59
    },
    {
      "epoch": 15.0,
      "grad_norm": 9.790772438049316,
      "learning_rate": 9.938285206917541e-08,
      "loss": 0.7127,
      "step": 60
    },
    {
      "epoch": 15.290909090909091,
      "grad_norm": 9.698260307312012,
      "learning_rate": 9.931634888554937e-08,
      "loss": 0.7662,
      "step": 61
    },
    {
      "epoch": 15.581818181818182,
      "grad_norm": 8.966428756713867,
      "learning_rate": 9.924647084037797e-08,
      "loss": 0.7652,
      "step": 62
    },
    {
      "epoch": 15.872727272727273,
      "grad_norm": 9.97366714477539,
      "learning_rate": 9.917322325514488e-08,
      "loss": 0.854,
      "step": 63
    },
    {
      "epoch": 16.0,
      "grad_norm": 8.8572359085083,
      "learning_rate": 9.909661170793732e-08,
      "loss": 0.6939,
      "step": 64
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.7460314631462097,
      "eval_runtime": 0.7685,
      "eval_samples_per_second": 16.917,
      "eval_steps_per_second": 16.917,
      "step": 64
    },
    {
      "epoch": 16.29090909090909,
      "grad_norm": 9.27105712890625,
      "learning_rate": 9.901664203302125e-08,
      "loss": 0.7894,
      "step": 65
    },
    {
      "epoch": 16.581818181818182,
      "grad_norm": 9.316947937011719,
      "learning_rate": 9.8933320320397e-08,
      "loss": 0.7476,
      "step": 66
    },
    {
      "epoch": 16.87272727272727,
      "grad_norm": 9.482162475585938,
      "learning_rate": 9.884665291533559e-08,
      "loss": 0.8294,
      "step": 67
    },
    {
      "epoch": 17.0,
      "grad_norm": 8.661737442016602,
      "learning_rate": 9.875664641789545e-08,
      "loss": 0.741,
      "step": 68
    },
    {
      "epoch": 17.29090909090909,
      "grad_norm": 8.976079940795898,
      "learning_rate": 9.866330768241983e-08,
      "loss": 0.7999,
      "step": 69
    },
    {
      "epoch": 17.581818181818182,
      "grad_norm": 9.500139236450195,
      "learning_rate": 9.856664381701484e-08,
      "loss": 0.8324,
      "step": 70
    },
    {
      "epoch": 17.87272727272727,
      "grad_norm": 8.916980743408203,
      "learning_rate": 9.846666218300807e-08,
      "loss": 0.7308,
      "step": 71
    },
    {
      "epoch": 18.0,
      "grad_norm": 9.658391952514648,
      "learning_rate": 9.836337039438803e-08,
      "loss": 0.7691,
      "step": 72
    },
    {
      "epoch": 18.29090909090909,
      "grad_norm": 9.849284172058105,
      "learning_rate": 9.825677631722435e-08,
      "loss": 0.7804,
      "step": 73
    },
    {
      "epoch": 18.581818181818182,
      "grad_norm": 9.135872840881348,
      "learning_rate": 9.814688806906868e-08,
      "loss": 0.808,
      "step": 74
    },
    {
      "epoch": 18.87272727272727,
      "grad_norm": 8.595244407653809,
      "learning_rate": 9.80337140183366e-08,
      "loss": 0.7593,
      "step": 75
    },
    {
      "epoch": 19.0,
      "grad_norm": 9.557754516601562,
      "learning_rate": 9.791726278367021e-08,
      "loss": 0.795,
      "step": 76
    },
    {
      "epoch": 19.29090909090909,
      "grad_norm": 9.38132095336914,
      "learning_rate": 9.779754323328191e-08,
      "loss": 0.8303,
      "step": 77
    },
    {
      "epoch": 19.581818181818182,
      "grad_norm": 9.702839851379395,
      "learning_rate": 9.767456448427896e-08,
      "loss": 0.7404,
      "step": 78
    },
    {
      "epoch": 19.87272727272727,
      "grad_norm": 8.890022277832031,
      "learning_rate": 9.754833590196926e-08,
      "loss": 0.8152,
      "step": 79
    },
    {
      "epoch": 20.0,
      "grad_norm": 8.762577056884766,
      "learning_rate": 9.741886709914803e-08,
      "loss": 0.6623,
      "step": 80
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.7418057918548584,
      "eval_runtime": 0.7512,
      "eval_samples_per_second": 17.305,
      "eval_steps_per_second": 17.305,
      "step": 80
    },
    {
      "epoch": 20.29090909090909,
      "grad_norm": 9.43545913696289,
      "learning_rate": 9.728616793536587e-08,
      "loss": 0.726,
      "step": 81
    },
    {
      "epoch": 20.581818181818182,
      "grad_norm": 8.36042308807373,
      "learning_rate": 9.715024851617789e-08,
      "loss": 0.7908,
      "step": 82
    },
    {
      "epoch": 20.87272727272727,
      "grad_norm": 9.46149730682373,
      "learning_rate": 9.701111919237408e-08,
      "loss": 0.8219,
      "step": 83
    },
    {
      "epoch": 21.0,
      "grad_norm": 9.277331352233887,
      "learning_rate": 9.68687905591911e-08,
      "loss": 0.7955,
      "step": 84
    },
    {
      "epoch": 21.29090909090909,
      "grad_norm": 9.980899810791016,
      "learning_rate": 9.672327345550542e-08,
      "loss": 0.8459,
      "step": 85
    },
    {
      "epoch": 21.581818181818182,
      "grad_norm": 8.734892845153809,
      "learning_rate": 9.65745789630079e-08,
      "loss": 0.7952,
      "step": 86
    },
    {
      "epoch": 21.87272727272727,
      "grad_norm": 7.979213714599609,
      "learning_rate": 9.642271840535982e-08,
      "loss": 0.6928,
      "step": 87
    },
    {
      "epoch": 22.0,
      "grad_norm": 9.570889472961426,
      "learning_rate": 9.626770334733058e-08,
      "loss": 0.7813,
      "step": 88
    },
    {
      "epoch": 22.29090909090909,
      "grad_norm": 9.478497505187988,
      "learning_rate": 9.610954559391703e-08,
      "loss": 0.783,
      "step": 89
    },
    {
      "epoch": 22.581818181818182,
      "grad_norm": 8.57199478149414,
      "learning_rate": 9.594825718944444e-08,
      "loss": 0.7859,
      "step": 90
    },
    {
      "epoch": 22.87272727272727,
      "grad_norm": 8.782203674316406,
      "learning_rate": 9.578385041664925e-08,
      "loss": 0.7784,
      "step": 91
    },
    {
      "epoch": 23.0,
      "grad_norm": 9.160470008850098,
      "learning_rate": 9.561633779574373e-08,
      "loss": 0.7613,
      "step": 92
    },
    {
      "epoch": 23.29090909090909,
      "grad_norm": 8.80034065246582,
      "learning_rate": 9.544573208346251e-08,
      "loss": 0.7708,
      "step": 93
    },
    {
      "epoch": 23.581818181818182,
      "grad_norm": 9.001204490661621,
      "learning_rate": 9.527204627209113e-08,
      "loss": 0.7975,
      "step": 94
    },
    {
      "epoch": 23.87272727272727,
      "grad_norm": 8.64294147491455,
      "learning_rate": 9.509529358847655e-08,
      "loss": 0.7533,
      "step": 95
    },
    {
      "epoch": 24.0,
      "grad_norm": 9.539164543151855,
      "learning_rate": 9.491548749301997e-08,
      "loss": 0.8112,
      "step": 96
    },
    {
      "epoch": 24.0,
      "eval_loss": 0.7388671040534973,
      "eval_runtime": 0.7379,
      "eval_samples_per_second": 17.617,
      "eval_steps_per_second": 17.617,
      "step": 96
    },
    {
      "epoch": 24.29090909090909,
      "grad_norm": 8.564647674560547,
      "learning_rate": 9.473264167865172e-08,
      "loss": 0.779,
      "step": 97
    },
    {
      "epoch": 24.581818181818182,
      "grad_norm": 8.466269493103027,
      "learning_rate": 9.454677006978843e-08,
      "loss": 0.7427,
      "step": 98
    },
    {
      "epoch": 24.87272727272727,
      "grad_norm": 9.549156188964844,
      "learning_rate": 9.435788682127281e-08,
      "loss": 0.7749,
      "step": 99
    },
    {
      "epoch": 25.0,
      "grad_norm": 8.791007041931152,
      "learning_rate": 9.416600631729548e-08,
      "loss": 0.8413,
      "step": 100
    },
    {
      "epoch": 25.29090909090909,
      "grad_norm": 8.481273651123047,
      "learning_rate": 9.397114317029974e-08,
      "loss": 0.7987,
      "step": 101
    },
    {
      "epoch": 25.581818181818182,
      "grad_norm": 7.957334518432617,
      "learning_rate": 9.377331221986867e-08,
      "loss": 0.7579,
      "step": 102
    },
    {
      "epoch": 25.87272727272727,
      "grad_norm": 7.695952415466309,
      "learning_rate": 9.357252853159505e-08,
      "loss": 0.7138,
      "step": 103
    },
    {
      "epoch": 26.0,
      "grad_norm": 8.535294532775879,
      "learning_rate": 9.336880739593415e-08,
      "loss": 0.8143,
      "step": 104
    },
    {
      "epoch": 26.29090909090909,
      "grad_norm": 7.785234451293945,
      "learning_rate": 9.316216432703917e-08,
      "loss": 0.7595,
      "step": 105
    },
    {
      "epoch": 26.581818181818182,
      "grad_norm": 7.210692882537842,
      "learning_rate": 9.295261506157986e-08,
      "loss": 0.6892,
      "step": 106
    },
    {
      "epoch": 26.87272727272727,
      "grad_norm": 7.439105033874512,
      "learning_rate": 9.274017555754408e-08,
      "loss": 0.7828,
      "step": 107
    },
    {
      "epoch": 27.0,
      "grad_norm": 8.5601167678833,
      "learning_rate": 9.252486199302256e-08,
      "loss": 0.8267,
      "step": 108
    },
    {
      "epoch": 27.29090909090909,
      "grad_norm": 7.751751899719238,
      "learning_rate": 9.230669076497686e-08,
      "loss": 0.7837,
      "step": 109
    },
    {
      "epoch": 27.581818181818182,
      "grad_norm": 7.58750057220459,
      "learning_rate": 9.20856784879907e-08,
      "loss": 0.7629,
      "step": 110
    },
    {
      "epoch": 27.87272727272727,
      "grad_norm": 7.078155040740967,
      "learning_rate": 9.186184199300463e-08,
      "loss": 0.732,
      "step": 111
    },
    {
      "epoch": 28.0,
      "grad_norm": 7.075254440307617,
      "learning_rate": 9.163519832603437e-08,
      "loss": 0.708,
      "step": 112
    },
    {
      "epoch": 28.0,
      "eval_loss": 0.7153984904289246,
      "eval_runtime": 0.7384,
      "eval_samples_per_second": 17.605,
      "eval_steps_per_second": 17.605,
      "step": 112
    },
    {
      "epoch": 28.29090909090909,
      "grad_norm": 7.660149097442627,
      "learning_rate": 9.140576474687262e-08,
      "loss": 0.7923,
      "step": 113
    },
    {
      "epoch": 28.581818181818182,
      "grad_norm": 6.516578674316406,
      "learning_rate": 9.117355872777476e-08,
      "loss": 0.6965,
      "step": 114
    },
    {
      "epoch": 28.87272727272727,
      "grad_norm": 6.818985462188721,
      "learning_rate": 9.093859795212817e-08,
      "loss": 0.7564,
      "step": 115
    },
    {
      "epoch": 29.0,
      "grad_norm": 9.244476318359375,
      "learning_rate": 9.070090031310558e-08,
      "loss": 0.7325,
      "step": 116
    },
    {
      "epoch": 29.29090909090909,
      "grad_norm": 7.331173419952393,
      "learning_rate": 9.046048391230248e-08,
      "loss": 0.6957,
      "step": 117
    },
    {
      "epoch": 29.581818181818182,
      "grad_norm": 7.235352993011475,
      "learning_rate": 9.021736705835861e-08,
      "loss": 0.8011,
      "step": 118
    },
    {
      "epoch": 29.87272727272727,
      "grad_norm": 7.370168209075928,
      "learning_rate": 8.997156826556369e-08,
      "loss": 0.7767,
      "step": 119
    },
    {
      "epoch": 30.0,
      "grad_norm": 6.547177314758301,
      "learning_rate": 8.97231062524474e-08,
      "loss": 0.6864,
      "step": 120
    },
    {
      "epoch": 30.29090909090909,
      "grad_norm": 6.999849796295166,
      "learning_rate": 8.9471999940354e-08,
      "loss": 0.757,
      "step": 121
    },
    {
      "epoch": 30.581818181818182,
      "grad_norm": 7.369142055511475,
      "learning_rate": 8.921826845200139e-08,
      "loss": 0.7184,
      "step": 122
    },
    {
      "epoch": 30.87272727272727,
      "grad_norm": 7.147704601287842,
      "learning_rate": 8.896193111002475e-08,
      "loss": 0.8074,
      "step": 123
    },
    {
      "epoch": 31.0,
      "grad_norm": 6.590007305145264,
      "learning_rate": 8.87030074355051e-08,
      "loss": 0.666,
      "step": 124
    },
    {
      "epoch": 31.29090909090909,
      "grad_norm": 6.5022711753845215,
      "learning_rate": 8.844151714648274e-08,
      "loss": 0.7109,
      "step": 125
    },
    {
      "epoch": 31.581818181818182,
      "grad_norm": 7.46487283706665,
      "learning_rate": 8.817748015645558e-08,
      "loss": 0.7848,
      "step": 126
    },
    {
      "epoch": 31.87272727272727,
      "grad_norm": 7.2371721267700195,
      "learning_rate": 8.791091657286267e-08,
      "loss": 0.7756,
      "step": 127
    },
    {
      "epoch": 32.0,
      "grad_norm": 6.450557708740234,
      "learning_rate": 8.764184669555293e-08,
      "loss": 0.6471,
      "step": 128
    },
    {
      "epoch": 32.0,
      "eval_loss": 0.7097088694572449,
      "eval_runtime": 0.7853,
      "eval_samples_per_second": 16.554,
      "eval_steps_per_second": 16.554,
      "step": 128
    },
    {
      "epoch": 32.29090909090909,
      "grad_norm": 7.1595611572265625,
      "learning_rate": 8.737029101523929e-08,
      "loss": 0.7418,
      "step": 129
    },
    {
      "epoch": 32.58181818181818,
      "grad_norm": 7.2520294189453125,
      "learning_rate": 8.709627021193817e-08,
      "loss": 0.7407,
      "step": 130
    },
    {
      "epoch": 32.872727272727275,
      "grad_norm": 6.757298469543457,
      "learning_rate": 8.681980515339464e-08,
      "loss": 0.7486,
      "step": 131
    },
    {
      "epoch": 33.0,
      "grad_norm": 6.70634651184082,
      "learning_rate": 8.65409168934933e-08,
      "loss": 0.7381,
      "step": 132
    },
    {
      "epoch": 33.29090909090909,
      "grad_norm": 8.267258644104004,
      "learning_rate": 8.625962667065488e-08,
      "loss": 0.8277,
      "step": 133
    },
    {
      "epoch": 33.58181818181818,
      "grad_norm": 6.568601608276367,
      "learning_rate": 8.597595590621892e-08,
      "loss": 0.7345,
      "step": 134
    },
    {
      "epoch": 33.872727272727275,
      "grad_norm": 6.368529796600342,
      "learning_rate": 8.568992620281244e-08,
      "loss": 0.6949,
      "step": 135
    },
    {
      "epoch": 34.0,
      "grad_norm": 6.077971458435059,
      "learning_rate": 8.540155934270471e-08,
      "loss": 0.6427,
      "step": 136
    },
    {
      "epoch": 34.29090909090909,
      "grad_norm": 6.2005743980407715,
      "learning_rate": 8.511087728614862e-08,
      "loss": 0.7113,
      "step": 137
    },
    {
      "epoch": 34.58181818181818,
      "grad_norm": 6.390923023223877,
      "learning_rate": 8.481790216970819e-08,
      "loss": 0.7422,
      "step": 138
    },
    {
      "epoch": 34.872727272727275,
      "grad_norm": 7.773628234863281,
      "learning_rate": 8.452265630457283e-08,
      "loss": 0.7829,
      "step": 139
    },
    {
      "epoch": 35.0,
      "grad_norm": 7.676466941833496,
      "learning_rate": 8.422516217485826e-08,
      "loss": 0.718,
      "step": 140
    },
    {
      "epoch": 35.29090909090909,
      "grad_norm": 6.630233287811279,
      "learning_rate": 8.392544243589427e-08,
      "loss": 0.7046,
      "step": 141
    },
    {
      "epoch": 35.58181818181818,
      "grad_norm": 6.816230297088623,
      "learning_rate": 8.362351991249938e-08,
      "loss": 0.7685,
      "step": 142
    },
    {
      "epoch": 35.872727272727275,
      "grad_norm": 6.341788291931152,
      "learning_rate": 8.331941759724268e-08,
      "loss": 0.6774,
      "step": 143
    },
    {
      "epoch": 36.0,
      "grad_norm": 8.306670188903809,
      "learning_rate": 8.301315864869288e-08,
      "loss": 0.9019,
      "step": 144
    },
    {
      "epoch": 36.0,
      "eval_loss": 0.7050113677978516,
      "eval_runtime": 0.7541,
      "eval_samples_per_second": 17.239,
      "eval_steps_per_second": 17.239,
      "step": 144
    },
    {
      "epoch": 36.29090909090909,
      "grad_norm": 6.883708953857422,
      "learning_rate": 8.270476638965461e-08,
      "loss": 0.7921,
      "step": 145
    },
    {
      "epoch": 36.58181818181818,
      "grad_norm": 6.389072418212891,
      "learning_rate": 8.239426430539243e-08,
      "loss": 0.6827,
      "step": 146
    },
    {
      "epoch": 36.872727272727275,
      "grad_norm": 6.924624443054199,
      "learning_rate": 8.208167604184218e-08,
      "loss": 0.7774,
      "step": 147
    },
    {
      "epoch": 37.0,
      "grad_norm": 6.5363450050354,
      "learning_rate": 8.176702540381035e-08,
      "loss": 0.6709,
      "step": 148
    },
    {
      "epoch": 37.29090909090909,
      "grad_norm": 6.154909610748291,
      "learning_rate": 8.145033635316129e-08,
      "loss": 0.7302,
      "step": 149
    },
    {
      "epoch": 37.58181818181818,
      "grad_norm": 6.3788676261901855,
      "learning_rate": 8.113163300699229e-08,
      "loss": 0.7301,
      "step": 150
    },
    {
      "epoch": 37.872727272727275,
      "grad_norm": 7.089733600616455,
      "learning_rate": 8.081093963579708e-08,
      "loss": 0.7223,
      "step": 151
    },
    {
      "epoch": 38.0,
      "grad_norm": 7.056278228759766,
      "learning_rate": 8.048828066161747e-08,
      "loss": 0.7989,
      "step": 152
    },
    {
      "epoch": 38.29090909090909,
      "grad_norm": 5.888208389282227,
      "learning_rate": 8.016368065618359e-08,
      "loss": 0.6768,
      "step": 153
    },
    {
      "epoch": 38.58181818181818,
      "grad_norm": 7.345203399658203,
      "learning_rate": 7.983716433904262e-08,
      "loss": 0.7454,
      "step": 154
    },
    {
      "epoch": 38.872727272727275,
      "grad_norm": 6.323718070983887,
      "learning_rate": 7.950875657567622e-08,
      "loss": 0.7511,
      "step": 155
    },
    {
      "epoch": 39.0,
      "grad_norm": 7.311026096343994,
      "learning_rate": 7.917848237560708e-08,
      "loss": 0.8113,
      "step": 156
    },
    {
      "epoch": 39.29090909090909,
      "grad_norm": 6.139308452606201,
      "learning_rate": 7.884636689049422e-08,
      "loss": 0.7076,
      "step": 157
    },
    {
      "epoch": 39.58181818181818,
      "grad_norm": 6.485006809234619,
      "learning_rate": 7.851243541221769e-08,
      "loss": 0.7437,
      "step": 158
    },
    {
      "epoch": 39.872727272727275,
      "grad_norm": 6.589916706085205,
      "learning_rate": 7.817671337095244e-08,
      "loss": 0.7404,
      "step": 159
    },
    {
      "epoch": 40.0,
      "grad_norm": 6.963124752044678,
      "learning_rate": 7.78392263332317e-08,
      "loss": 0.7328,
      "step": 160
    },
    {
      "epoch": 40.0,
      "eval_loss": 0.7006868124008179,
      "eval_runtime": 0.7566,
      "eval_samples_per_second": 17.183,
      "eval_steps_per_second": 17.183,
      "step": 160
    },
    {
      "epoch": 40.29090909090909,
      "grad_norm": 6.547840118408203,
      "learning_rate": 7.75e-08,
      "loss": 0.7431,
      "step": 161
    },
    {
      "epoch": 40.58181818181818,
      "grad_norm": 6.299688816070557,
      "learning_rate": 7.715906020465603e-08,
      "loss": 0.7585,
      "step": 162
    },
    {
      "epoch": 40.872727272727275,
      "grad_norm": 6.586760997772217,
      "learning_rate": 7.681643291108518e-08,
      "loss": 0.7324,
      "step": 163
    },
    {
      "epoch": 41.0,
      "grad_norm": 6.389430999755859,
      "learning_rate": 7.647214421168238e-08,
      "loss": 0.6533,
      "step": 164
    },
    {
      "epoch": 41.29090909090909,
      "grad_norm": 6.027109146118164,
      "learning_rate": 7.612622032536508e-08,
      "loss": 0.7135,
      "step": 165
    },
    {
      "epoch": 41.58181818181818,
      "grad_norm": 6.719674110412598,
      "learning_rate": 7.577868759557654e-08,
      "loss": 0.7597,
      "step": 166
    },
    {
      "epoch": 41.872727272727275,
      "grad_norm": 6.458725929260254,
      "learning_rate": 7.54295724882796e-08,
      "loss": 0.7109,
      "step": 167
    },
    {
      "epoch": 42.0,
      "grad_norm": 6.904190540313721,
      "learning_rate": 7.507890158994139e-08,
      "loss": 0.7504,
      "step": 168
    },
    {
      "epoch": 42.29090909090909,
      "grad_norm": 6.580723285675049,
      "learning_rate": 7.472670160550848e-08,
      "loss": 0.7096,
      "step": 169
    },
    {
      "epoch": 42.58181818181818,
      "grad_norm": 6.410011291503906,
      "learning_rate": 7.437299935637328e-08,
      "loss": 0.7692,
      "step": 170
    },
    {
      "epoch": 42.872727272727275,
      "grad_norm": 6.3067827224731445,
      "learning_rate": 7.401782177833146e-08,
      "loss": 0.7346,
      "step": 171
    },
    {
      "epoch": 43.0,
      "grad_norm": 5.755003929138184,
      "learning_rate": 7.366119591953075e-08,
      "loss": 0.6633,
      "step": 172
    },
    {
      "epoch": 43.29090909090909,
      "grad_norm": 6.46678352355957,
      "learning_rate": 7.3303148938411e-08,
      "loss": 0.7365,
      "step": 173
    },
    {
      "epoch": 43.58181818181818,
      "grad_norm": 6.622053623199463,
      "learning_rate": 7.294370810163607e-08,
      "loss": 0.7511,
      "step": 174
    },
    {
      "epoch": 43.872727272727275,
      "grad_norm": 5.266422748565674,
      "learning_rate": 7.258290078201731e-08,
      "loss": 0.6481,
      "step": 175
    },
    {
      "epoch": 44.0,
      "grad_norm": 7.341455936431885,
      "learning_rate": 7.222075445642904e-08,
      "loss": 0.8191,
      "step": 176
    },
    {
      "epoch": 44.0,
      "eval_loss": 0.6937930583953857,
      "eval_runtime": 0.7725,
      "eval_samples_per_second": 16.828,
      "eval_steps_per_second": 16.828,
      "step": 176
    },
    {
      "epoch": 44.29090909090909,
      "grad_norm": 6.314858436584473,
      "learning_rate": 7.185729670371604e-08,
      "loss": 0.7001,
      "step": 177
    },
    {
      "epoch": 44.58181818181818,
      "grad_norm": 6.364148139953613,
      "learning_rate": 7.149255520259337e-08,
      "loss": 0.786,
      "step": 178
    },
    {
      "epoch": 44.872727272727275,
      "grad_norm": 5.679451942443848,
      "learning_rate": 7.11265577295385e-08,
      "loss": 0.6767,
      "step": 179
    },
    {
      "epoch": 45.0,
      "grad_norm": 6.4454216957092285,
      "learning_rate": 7.075933215667603e-08,
      "loss": 0.7351,
      "step": 180
    },
    {
      "epoch": 45.29090909090909,
      "grad_norm": 5.991427421569824,
      "learning_rate": 7.039090644965509e-08,
      "loss": 0.7047,
      "step": 181
    },
    {
      "epoch": 45.58181818181818,
      "grad_norm": 5.386115550994873,
      "learning_rate": 7.002130866551968e-08,
      "loss": 0.7113,
      "step": 182
    },
    {
      "epoch": 45.872727272727275,
      "grad_norm": 6.815364360809326,
      "learning_rate": 6.965056695057204e-08,
      "loss": 0.7255,
      "step": 183
    },
    {
      "epoch": 46.0,
      "grad_norm": 6.38714599609375,
      "learning_rate": 6.927870953822915e-08,
      "loss": 0.7503,
      "step": 184
    },
    {
      "epoch": 46.29090909090909,
      "grad_norm": 5.759856224060059,
      "learning_rate": 6.890576474687262e-08,
      "loss": 0.7008,
      "step": 185
    },
    {
      "epoch": 46.58181818181818,
      "grad_norm": 5.1396918296813965,
      "learning_rate": 6.853176097769228e-08,
      "loss": 0.6925,
      "step": 186
    },
    {
      "epoch": 46.872727272727275,
      "grad_norm": 5.9070539474487305,
      "learning_rate": 6.815672671252315e-08,
      "loss": 0.7409,
      "step": 187
    },
    {
      "epoch": 47.0,
      "grad_norm": 5.90541410446167,
      "learning_rate": 6.778069051167653e-08,
      "loss": 0.702,
      "step": 188
    },
    {
      "epoch": 47.29090909090909,
      "grad_norm": 5.474076747894287,
      "learning_rate": 6.740368101176495e-08,
      "loss": 0.7085,
      "step": 189
    },
    {
      "epoch": 47.58181818181818,
      "grad_norm": 5.111520767211914,
      "learning_rate": 6.702572692352155e-08,
      "loss": 0.685,
      "step": 190
    },
    {
      "epoch": 47.872727272727275,
      "grad_norm": 5.618140697479248,
      "learning_rate": 6.664685702961344e-08,
      "loss": 0.7551,
      "step": 191
    },
    {
      "epoch": 48.0,
      "grad_norm": 4.961245059967041,
      "learning_rate": 6.626710018244986e-08,
      "loss": 0.6327,
      "step": 192
    },
    {
      "epoch": 48.0,
      "eval_loss": 0.6752312183380127,
      "eval_runtime": 0.7832,
      "eval_samples_per_second": 16.599,
      "eval_steps_per_second": 16.599,
      "step": 192
    },
    {
      "epoch": 48.29090909090909,
      "grad_norm": 5.36975622177124,
      "learning_rate": 6.588648530198504e-08,
      "loss": 0.7312,
      "step": 193
    },
    {
      "epoch": 48.58181818181818,
      "grad_norm": 5.021007061004639,
      "learning_rate": 6.550504137351574e-08,
      "loss": 0.7467,
      "step": 194
    },
    {
      "epoch": 48.872727272727275,
      "grad_norm": 4.721583843231201,
      "learning_rate": 6.512279744547392e-08,
      "loss": 0.6271,
      "step": 195
    },
    {
      "epoch": 49.0,
      "grad_norm": 5.531439304351807,
      "learning_rate": 6.473978262721462e-08,
      "loss": 0.7127,
      "step": 196
    },
    {
      "epoch": 49.29090909090909,
      "grad_norm": 5.3525309562683105,
      "learning_rate": 6.435602608679917e-08,
      "loss": 0.7255,
      "step": 197
    },
    {
      "epoch": 49.58181818181818,
      "grad_norm": 4.411137104034424,
      "learning_rate": 6.397155704877387e-08,
      "loss": 0.6177,
      "step": 198
    },
    {
      "epoch": 49.872727272727275,
      "grad_norm": 4.907252788543701,
      "learning_rate": 6.358640479194451e-08,
      "loss": 0.7295,
      "step": 199
    },
    {
      "epoch": 50.0,
      "grad_norm": 4.626101493835449,
      "learning_rate": 6.320059864714664e-08,
      "loss": 0.7091,
      "step": 200
    },
    {
      "epoch": 50.29090909090909,
      "grad_norm": 4.853626728057861,
      "learning_rate": 6.281416799501187e-08,
      "loss": 0.7432,
      "step": 201
    },
    {
      "epoch": 50.58181818181818,
      "grad_norm": 4.439899921417236,
      "learning_rate": 6.242714226373049e-08,
      "loss": 0.676,
      "step": 202
    },
    {
      "epoch": 50.872727272727275,
      "grad_norm": 4.5280985832214355,
      "learning_rate": 6.203955092681039e-08,
      "loss": 0.7086,
      "step": 203
    },
    {
      "epoch": 51.0,
      "grad_norm": 4.414018154144287,
      "learning_rate": 6.165142350083249e-08,
      "loss": 0.5264,
      "step": 204
    },
    {
      "epoch": 51.29090909090909,
      "grad_norm": 4.17572021484375,
      "learning_rate": 6.126278954320294e-08,
      "loss": 0.7346,
      "step": 205
    },
    {
      "epoch": 51.58181818181818,
      "grad_norm": 4.015255928039551,
      "learning_rate": 6.087367864990232e-08,
      "loss": 0.6239,
      "step": 206
    },
    {
      "epoch": 51.872727272727275,
      "grad_norm": 4.698182582855225,
      "learning_rate": 6.048412045323163e-08,
      "loss": 0.688,
      "step": 207
    },
    {
      "epoch": 52.0,
      "grad_norm": 5.5075297355651855,
      "learning_rate": 6.00941446195558e-08,
      "loss": 0.6903,
      "step": 208
    },
    {
      "epoch": 52.0,
      "eval_loss": 0.6604220271110535,
      "eval_runtime": 0.6915,
      "eval_samples_per_second": 18.8,
      "eval_steps_per_second": 18.8,
      "step": 208
    },
    {
      "epoch": 52.29090909090909,
      "grad_norm": 3.8842809200286865,
      "learning_rate": 5.970378084704441e-08,
      "loss": 0.6428,
      "step": 209
    },
    {
      "epoch": 52.58181818181818,
      "grad_norm": 4.9067301750183105,
      "learning_rate": 5.931305886341008e-08,
      "loss": 0.7572,
      "step": 210
    },
    {
      "epoch": 52.872727272727275,
      "grad_norm": 4.025907516479492,
      "learning_rate": 5.892200842364462e-08,
      "loss": 0.6545,
      "step": 211
    },
    {
      "epoch": 53.0,
      "grad_norm": 4.105547904968262,
      "learning_rate": 5.853065930775303e-08,
      "loss": 0.6439,
      "step": 212
    },
    {
      "epoch": 53.29090909090909,
      "grad_norm": 3.7520296573638916,
      "learning_rate": 5.813904131848564e-08,
      "loss": 0.677,
      "step": 213
    },
    {
      "epoch": 53.58181818181818,
      "grad_norm": 3.975045680999756,
      "learning_rate": 5.7747184279068564e-08,
      "loss": 0.6321,
      "step": 214
    },
    {
      "epoch": 53.872727272727275,
      "grad_norm": 4.536473274230957,
      "learning_rate": 5.735511803093248e-08,
      "loss": 0.7326,
      "step": 215
    },
    {
      "epoch": 54.0,
      "grad_norm": 5.148712158203125,
      "learning_rate": 5.696287243144012e-08,
      "loss": 0.6819,
      "step": 216
    },
    {
      "epoch": 54.29090909090909,
      "grad_norm": 3.6721999645233154,
      "learning_rate": 5.6570477351612554e-08,
      "loss": 0.6655,
      "step": 217
    },
    {
      "epoch": 54.58181818181818,
      "grad_norm": 4.29323148727417,
      "learning_rate": 5.61779626738543e-08,
      "loss": 0.6743,
      "step": 218
    },
    {
      "epoch": 54.872727272727275,
      "grad_norm": 4.018572807312012,
      "learning_rate": 5.5785358289677765e-08,
      "loss": 0.711,
      "step": 219
    },
    {
      "epoch": 55.0,
      "grad_norm": 4.6550445556640625,
      "learning_rate": 5.539269409742683e-08,
      "loss": 0.6398,
      "step": 220
    },
    {
      "epoch": 55.29090909090909,
      "grad_norm": 4.599621295928955,
      "learning_rate": 5.5e-08,
      "loss": 0.6885,
      "step": 221
    },
    {
      "epoch": 55.58181818181818,
      "grad_norm": 3.6876866817474365,
      "learning_rate": 5.460730590257318e-08,
      "loss": 0.6391,
      "step": 222
    },
    {
      "epoch": 55.872727272727275,
      "grad_norm": 3.641345262527466,
      "learning_rate": 5.421464171032224e-08,
      "loss": 0.6684,
      "step": 223
    },
    {
      "epoch": 56.0,
      "grad_norm": 4.325244903564453,
      "learning_rate": 5.382203732614572e-08,
      "loss": 0.7467,
      "step": 224
    },
    {
      "epoch": 56.0,
      "eval_loss": 0.6532977819442749,
      "eval_runtime": 0.746,
      "eval_samples_per_second": 17.427,
      "eval_steps_per_second": 17.427,
      "step": 224
    },
    {
      "epoch": 56.29090909090909,
      "grad_norm": 4.434227466583252,
      "learning_rate": 5.342952264838747e-08,
      "loss": 0.7395,
      "step": 225
    },
    {
      "epoch": 56.58181818181818,
      "grad_norm": 4.03561544418335,
      "learning_rate": 5.303712756855988e-08,
      "loss": 0.7176,
      "step": 226
    },
    {
      "epoch": 56.872727272727275,
      "grad_norm": 3.4329726696014404,
      "learning_rate": 5.264488196906752e-08,
      "loss": 0.5565,
      "step": 227
    },
    {
      "epoch": 57.0,
      "grad_norm": 3.6157584190368652,
      "learning_rate": 5.225281572093143e-08,
      "loss": 0.7052,
      "step": 228
    },
    {
      "epoch": 57.29090909090909,
      "grad_norm": 3.654561996459961,
      "learning_rate": 5.1860958681514355e-08,
      "loss": 0.6931,
      "step": 229
    },
    {
      "epoch": 57.58181818181818,
      "grad_norm": 3.4616754055023193,
      "learning_rate": 5.1469340692246985e-08,
      "loss": 0.6126,
      "step": 230
    },
    {
      "epoch": 57.872727272727275,
      "grad_norm": 4.538090229034424,
      "learning_rate": 5.107799157635537e-08,
      "loss": 0.7149,
      "step": 231
    },
    {
      "epoch": 58.0,
      "grad_norm": 3.8424854278564453,
      "learning_rate": 5.068694113658992e-08,
      "loss": 0.6564,
      "step": 232
    },
    {
      "epoch": 58.29090909090909,
      "grad_norm": 3.360053777694702,
      "learning_rate": 5.02962191529556e-08,
      "loss": 0.6657,
      "step": 233
    },
    {
      "epoch": 58.58181818181818,
      "grad_norm": 4.166203022003174,
      "learning_rate": 4.9905855380444194e-08,
      "loss": 0.7461,
      "step": 234
    },
    {
      "epoch": 58.872727272727275,
      "grad_norm": 3.4333815574645996,
      "learning_rate": 4.9515879546768366e-08,
      "loss": 0.5924,
      "step": 235
    },
    {
      "epoch": 59.0,
      "grad_norm": 4.719890594482422,
      "learning_rate": 4.912632135009769e-08,
      "loss": 0.6793,
      "step": 236
    },
    {
      "epoch": 59.29090909090909,
      "grad_norm": 3.6366472244262695,
      "learning_rate": 4.873721045679706e-08,
      "loss": 0.6648,
      "step": 237
    },
    {
      "epoch": 59.58181818181818,
      "grad_norm": 4.29836893081665,
      "learning_rate": 4.8348576499167516e-08,
      "loss": 0.6871,
      "step": 238
    },
    {
      "epoch": 59.872727272727275,
      "grad_norm": 3.3436715602874756,
      "learning_rate": 4.7960449073189604e-08,
      "loss": 0.6136,
      "step": 239
    },
    {
      "epoch": 60.0,
      "grad_norm": 3.974397897720337,
      "learning_rate": 4.75728577362695e-08,
      "loss": 0.7364,
      "step": 240
    },
    {
      "epoch": 60.0,
      "eval_loss": 0.6488688588142395,
      "eval_runtime": 0.7429,
      "eval_samples_per_second": 17.5,
      "eval_steps_per_second": 17.5,
      "step": 240
    },
    {
      "epoch": 60.29090909090909,
      "grad_norm": 4.133732318878174,
      "learning_rate": 4.718583200498813e-08,
      "loss": 0.7386,
      "step": 241
    },
    {
      "epoch": 60.58181818181818,
      "grad_norm": 3.358363151550293,
      "learning_rate": 4.6799401352853365e-08,
      "loss": 0.6255,
      "step": 242
    },
    {
      "epoch": 60.872727272727275,
      "grad_norm": 3.73943829536438,
      "learning_rate": 4.641359520805548e-08,
      "loss": 0.6834,
      "step": 243
    },
    {
      "epoch": 61.0,
      "grad_norm": 3.680448532104492,
      "learning_rate": 4.6028442951226135e-08,
      "loss": 0.5903,
      "step": 244
    },
    {
      "epoch": 61.29090909090909,
      "grad_norm": 3.3045241832733154,
      "learning_rate": 4.564397391320084e-08,
      "loss": 0.5871,
      "step": 245
    },
    {
      "epoch": 61.58181818181818,
      "grad_norm": 3.690742015838623,
      "learning_rate": 4.526021737278537e-08,
      "loss": 0.6913,
      "step": 246
    },
    {
      "epoch": 61.872727272727275,
      "grad_norm": 4.233401775360107,
      "learning_rate": 4.4877202554526084e-08,
      "loss": 0.7115,
      "step": 247
    },
    {
      "epoch": 62.0,
      "grad_norm": 3.5080771446228027,
      "learning_rate": 4.449495862648427e-08,
      "loss": 0.687,
      "step": 248
    },
    {
      "epoch": 62.29090909090909,
      "grad_norm": 3.3871119022369385,
      "learning_rate": 4.4113514698014955e-08,
      "loss": 0.6901,
      "step": 249
    },
    {
      "epoch": 62.58181818181818,
      "grad_norm": 3.6088693141937256,
      "learning_rate": 4.373289981755013e-08,
      "loss": 0.631,
      "step": 250
    },
    {
      "epoch": 62.872727272727275,
      "grad_norm": 3.743149518966675,
      "learning_rate": 4.335314297038656e-08,
      "loss": 0.6351,
      "step": 251
    },
    {
      "epoch": 63.0,
      "grad_norm": 4.030084133148193,
      "learning_rate": 4.297427307647844e-08,
      "loss": 0.7212,
      "step": 252
    },
    {
      "epoch": 63.29090909090909,
      "grad_norm": 3.458228349685669,
      "learning_rate": 4.2596318988235035e-08,
      "loss": 0.629,
      "step": 253
    },
    {
      "epoch": 63.58181818181818,
      "grad_norm": 4.063506126403809,
      "learning_rate": 4.2219309488323486e-08,
      "loss": 0.6565,
      "step": 254
    },
    {
      "epoch": 63.872727272727275,
      "grad_norm": 3.257892370223999,
      "learning_rate": 4.184327328747685e-08,
      "loss": 0.6644,
      "step": 255
    },
    {
      "epoch": 64.0,
      "grad_norm": 3.964184284210205,
      "learning_rate": 4.1468239022307716e-08,
      "loss": 0.7706,
      "step": 256
    },
    {
      "epoch": 64.0,
      "eval_loss": 0.6460027694702148,
      "eval_runtime": 0.7572,
      "eval_samples_per_second": 17.168,
      "eval_steps_per_second": 17.168,
      "step": 256
    },
    {
      "epoch": 64.2909090909091,
      "grad_norm": 3.444884777069092,
      "learning_rate": 4.1094235253127375e-08,
      "loss": 0.5848,
      "step": 257
    },
    {
      "epoch": 64.58181818181818,
      "grad_norm": 3.34226131439209,
      "learning_rate": 4.072129046177086e-08,
      "loss": 0.6438,
      "step": 258
    },
    {
      "epoch": 64.87272727272727,
      "grad_norm": 4.081578254699707,
      "learning_rate": 4.034943304942796e-08,
      "loss": 0.7825,
      "step": 259
    },
    {
      "epoch": 65.0,
      "grad_norm": 3.9306929111480713,
      "learning_rate": 3.997869133448031e-08,
      "loss": 0.7003,
      "step": 260
    },
    {
      "epoch": 65.2909090909091,
      "grad_norm": 3.3377864360809326,
      "learning_rate": 3.960909355034491e-08,
      "loss": 0.6723,
      "step": 261
    },
    {
      "epoch": 65.58181818181818,
      "grad_norm": 4.126795291900635,
      "learning_rate": 3.924066784332396e-08,
      "loss": 0.6778,
      "step": 262
    },
    {
      "epoch": 65.87272727272727,
      "grad_norm": 3.283628225326538,
      "learning_rate": 3.8873442270461487e-08,
      "loss": 0.6196,
      "step": 263
    },
    {
      "epoch": 66.0,
      "grad_norm": 3.599966526031494,
      "learning_rate": 3.850744479740663e-08,
      "loss": 0.7125,
      "step": 264
    },
    {
      "epoch": 66.2909090909091,
      "grad_norm": 3.398857831954956,
      "learning_rate": 3.814270329628395e-08,
      "loss": 0.6958,
      "step": 265
    },
    {
      "epoch": 66.58181818181818,
      "grad_norm": 3.538728952407837,
      "learning_rate": 3.777924554357096e-08,
      "loss": 0.6089,
      "step": 266
    },
    {
      "epoch": 66.87272727272727,
      "grad_norm": 3.549941062927246,
      "learning_rate": 3.7417099217982684e-08,
      "loss": 0.6794,
      "step": 267
    },
    {
      "epoch": 67.0,
      "grad_norm": 3.7104790210723877,
      "learning_rate": 3.7056291898363926e-08,
      "loss": 0.6845,
      "step": 268
    },
    {
      "epoch": 67.2909090909091,
      "grad_norm": 3.7661495208740234,
      "learning_rate": 3.669685106158899e-08,
      "loss": 0.6726,
      "step": 269
    },
    {
      "epoch": 67.58181818181818,
      "grad_norm": 3.2654290199279785,
      "learning_rate": 3.633880408046926e-08,
      "loss": 0.6597,
      "step": 270
    },
    {
      "epoch": 67.87272727272727,
      "grad_norm": 3.3389344215393066,
      "learning_rate": 3.598217822166854e-08,
      "loss": 0.5967,
      "step": 271
    },
    {
      "epoch": 68.0,
      "grad_norm": 4.037345886230469,
      "learning_rate": 3.5627000643626705e-08,
      "loss": 0.7777,
      "step": 272
    },
    {
      "epoch": 68.0,
      "eval_loss": 0.6440867185592651,
      "eval_runtime": 0.7501,
      "eval_samples_per_second": 17.331,
      "eval_steps_per_second": 17.331,
      "step": 272
    },
    {
      "epoch": 68.2909090909091,
      "grad_norm": 3.359079360961914,
      "learning_rate": 3.527329839449151e-08,
      "loss": 0.6824,
      "step": 273
    },
    {
      "epoch": 68.58181818181818,
      "grad_norm": 3.1845285892486572,
      "learning_rate": 3.49210984100586e-08,
      "loss": 0.5956,
      "step": 274
    },
    {
      "epoch": 68.87272727272727,
      "grad_norm": 3.564899206161499,
      "learning_rate": 3.4570427511720395e-08,
      "loss": 0.679,
      "step": 275
    },
    {
      "epoch": 69.0,
      "grad_norm": 5.116268157958984,
      "learning_rate": 3.4221312404423483e-08,
      "loss": 0.7194,
      "step": 276
    },
    {
      "epoch": 69.2909090909091,
      "grad_norm": 3.2692816257476807,
      "learning_rate": 3.387377967463493e-08,
      "loss": 0.5994,
      "step": 277
    },
    {
      "epoch": 69.58181818181818,
      "grad_norm": 3.1827392578125,
      "learning_rate": 3.3527855788317614e-08,
      "loss": 0.6582,
      "step": 278
    },
    {
      "epoch": 69.87272727272727,
      "grad_norm": 3.776779890060425,
      "learning_rate": 3.3183567088914834e-08,
      "loss": 0.7129,
      "step": 279
    },
    {
      "epoch": 70.0,
      "grad_norm": 4.038694381713867,
      "learning_rate": 3.2840939795343986e-08,
      "loss": 0.7197,
      "step": 280
    },
    {
      "epoch": 70.2909090909091,
      "grad_norm": 4.152867317199707,
      "learning_rate": 3.250000000000001e-08,
      "loss": 0.6844,
      "step": 281
    },
    {
      "epoch": 70.58181818181818,
      "grad_norm": 3.327399492263794,
      "learning_rate": 3.2160773666768325e-08,
      "loss": 0.7062,
      "step": 282
    },
    {
      "epoch": 70.87272727272727,
      "grad_norm": 3.2052102088928223,
      "learning_rate": 3.182328662904756e-08,
      "loss": 0.597,
      "step": 283
    },
    {
      "epoch": 71.0,
      "grad_norm": 3.528104305267334,
      "learning_rate": 3.14875645877823e-08,
      "loss": 0.6592,
      "step": 284
    },
    {
      "epoch": 71.2909090909091,
      "grad_norm": 3.380600929260254,
      "learning_rate": 3.1153633109505784e-08,
      "loss": 0.6398,
      "step": 285
    },
    {
      "epoch": 71.58181818181818,
      "grad_norm": 3.3184642791748047,
      "learning_rate": 3.082151762439292e-08,
      "loss": 0.6979,
      "step": 286
    },
    {
      "epoch": 71.87272727272727,
      "grad_norm": 3.4902994632720947,
      "learning_rate": 3.049124342432378e-08,
      "loss": 0.6352,
      "step": 287
    },
    {
      "epoch": 72.0,
      "grad_norm": 3.682870388031006,
      "learning_rate": 3.0162835660957385e-08,
      "loss": 0.6391,
      "step": 288
    },
    {
      "epoch": 72.0,
      "eval_loss": 0.641921877861023,
      "eval_runtime": 0.7653,
      "eval_samples_per_second": 16.986,
      "eval_steps_per_second": 16.986,
      "step": 288
    },
    {
      "epoch": 72.2909090909091,
      "grad_norm": 3.4279274940490723,
      "learning_rate": 2.983631934381639e-08,
      "loss": 0.6219,
      "step": 289
    },
    {
      "epoch": 72.58181818181818,
      "grad_norm": 3.627363681793213,
      "learning_rate": 2.9511719338382535e-08,
      "loss": 0.6635,
      "step": 290
    },
    {
      "epoch": 72.87272727272727,
      "grad_norm": 3.1634864807128906,
      "learning_rate": 2.918906036420294e-08,
      "loss": 0.6377,
      "step": 291
    },
    {
      "epoch": 73.0,
      "grad_norm": 3.896449327468872,
      "learning_rate": 2.886836699300771e-08,
      "loss": 0.7822,
      "step": 292
    },
    {
      "epoch": 73.2909090909091,
      "grad_norm": 3.168968677520752,
      "learning_rate": 2.8549663646838718e-08,
      "loss": 0.609,
      "step": 293
    },
    {
      "epoch": 73.58181818181818,
      "grad_norm": 3.3781349658966064,
      "learning_rate": 2.8232974596189653e-08,
      "loss": 0.6728,
      "step": 294
    },
    {
      "epoch": 73.87272727272727,
      "grad_norm": 3.41473650932312,
      "learning_rate": 2.791832395815782e-08,
      "loss": 0.7013,
      "step": 295
    },
    {
      "epoch": 74.0,
      "grad_norm": 3.771911859512329,
      "learning_rate": 2.760573569460757e-08,
      "loss": 0.6343,
      "step": 296
    },
    {
      "epoch": 74.2909090909091,
      "grad_norm": 3.527878761291504,
      "learning_rate": 2.729523361034538e-08,
      "loss": 0.6528,
      "step": 297
    },
    {
      "epoch": 74.58181818181818,
      "grad_norm": 3.105755090713501,
      "learning_rate": 2.6986841351307128e-08,
      "loss": 0.6243,
      "step": 298
    },
    {
      "epoch": 74.87272727272727,
      "grad_norm": 3.3217263221740723,
      "learning_rate": 2.6680582402757322e-08,
      "loss": 0.6658,
      "step": 299
    },
    {
      "epoch": 75.0,
      "grad_norm": 4.193359375,
      "learning_rate": 2.637648008750062e-08,
      "loss": 0.7016,
      "step": 300
    },
    {
      "epoch": 75.2909090909091,
      "grad_norm": 3.2874765396118164,
      "learning_rate": 2.6074557564105726e-08,
      "loss": 0.6661,
      "step": 301
    },
    {
      "epoch": 75.58181818181818,
      "grad_norm": 3.4806275367736816,
      "learning_rate": 2.5774837825141737e-08,
      "loss": 0.6277,
      "step": 302
    },
    {
      "epoch": 75.87272727272727,
      "grad_norm": 3.398120880126953,
      "learning_rate": 2.547734369542718e-08,
      "loss": 0.6863,
      "step": 303
    },
    {
      "epoch": 76.0,
      "grad_norm": 3.1762161254882812,
      "learning_rate": 2.5182097830291825e-08,
      "loss": 0.648,
      "step": 304
    },
    {
      "epoch": 76.0,
      "eval_loss": 0.6407743096351624,
      "eval_runtime": 0.7838,
      "eval_samples_per_second": 16.585,
      "eval_steps_per_second": 16.585,
      "step": 304
    },
    {
      "epoch": 76.2909090909091,
      "grad_norm": 3.250011444091797,
      "learning_rate": 2.4889122713851394e-08,
      "loss": 0.6552,
      "step": 305
    },
    {
      "epoch": 76.58181818181818,
      "grad_norm": 3.1045658588409424,
      "learning_rate": 2.4598440657295288e-08,
      "loss": 0.6147,
      "step": 306
    },
    {
      "epoch": 76.87272727272727,
      "grad_norm": 4.007096290588379,
      "learning_rate": 2.4310073797187574e-08,
      "loss": 0.7181,
      "step": 307
    },
    {
      "epoch": 77.0,
      "grad_norm": 3.300295829772949,
      "learning_rate": 2.4024044093781064e-08,
      "loss": 0.6115,
      "step": 308
    },
    {
      "epoch": 77.2909090909091,
      "grad_norm": 3.376610517501831,
      "learning_rate": 2.3740373329345117e-08,
      "loss": 0.7065,
      "step": 309
    },
    {
      "epoch": 77.58181818181818,
      "grad_norm": 3.1987497806549072,
      "learning_rate": 2.3459083106506712e-08,
      "loss": 0.6265,
      "step": 310
    },
    {
      "epoch": 77.87272727272727,
      "grad_norm": 3.428140878677368,
      "learning_rate": 2.3180194846605363e-08,
      "loss": 0.629,
      "step": 311
    },
    {
      "epoch": 78.0,
      "grad_norm": 3.489027261734009,
      "learning_rate": 2.2903729788061836e-08,
      "loss": 0.6626,
      "step": 312
    },
    {
      "epoch": 78.2909090909091,
      "grad_norm": 3.7477946281433105,
      "learning_rate": 2.2629708984760707e-08,
      "loss": 0.7006,
      "step": 313
    },
    {
      "epoch": 78.58181818181818,
      "grad_norm": 3.2413809299468994,
      "learning_rate": 2.2358153304447067e-08,
      "loss": 0.6363,
      "step": 314
    },
    {
      "epoch": 78.87272727272727,
      "grad_norm": 3.0365958213806152,
      "learning_rate": 2.2089083427137328e-08,
      "loss": 0.6307,
      "step": 315
    },
    {
      "epoch": 79.0,
      "grad_norm": 3.5392417907714844,
      "learning_rate": 2.182251984354442e-08,
      "loss": 0.6594,
      "step": 316
    },
    {
      "epoch": 79.2909090909091,
      "grad_norm": 3.2169861793518066,
      "learning_rate": 2.1558482853517254e-08,
      "loss": 0.6261,
      "step": 317
    },
    {
      "epoch": 79.58181818181818,
      "grad_norm": 3.1975908279418945,
      "learning_rate": 2.1296992564494903e-08,
      "loss": 0.6303,
      "step": 318
    },
    {
      "epoch": 79.87272727272727,
      "grad_norm": 3.5037009716033936,
      "learning_rate": 2.103806888997526e-08,
      "loss": 0.6847,
      "step": 319
    },
    {
      "epoch": 80.0,
      "grad_norm": 3.49397611618042,
      "learning_rate": 2.078173154799861e-08,
      "loss": 0.704,
      "step": 320
    },
    {
      "epoch": 80.0,
      "eval_loss": 0.6397803425788879,
      "eval_runtime": 0.7407,
      "eval_samples_per_second": 17.552,
      "eval_steps_per_second": 17.552,
      "step": 320
    },
    {
      "epoch": 80.2909090909091,
      "grad_norm": 3.312922954559326,
      "learning_rate": 2.0528000059645996e-08,
      "loss": 0.6742,
      "step": 321
    },
    {
      "epoch": 80.58181818181818,
      "grad_norm": 3.8569176197052,
      "learning_rate": 2.027689374755261e-08,
      "loss": 0.6589,
      "step": 322
    },
    {
      "epoch": 80.87272727272727,
      "grad_norm": 3.100782871246338,
      "learning_rate": 2.0028431734436306e-08,
      "loss": 0.6384,
      "step": 323
    },
    {
      "epoch": 81.0,
      "grad_norm": 3.488448143005371,
      "learning_rate": 1.9782632941641373e-08,
      "loss": 0.6523,
      "step": 324
    },
    {
      "epoch": 81.2909090909091,
      "grad_norm": 3.713844060897827,
      "learning_rate": 1.9539516087697516e-08,
      "loss": 0.7186,
      "step": 325
    },
    {
      "epoch": 81.58181818181818,
      "grad_norm": 3.2771830558776855,
      "learning_rate": 1.9299099686894422e-08,
      "loss": 0.7155,
      "step": 326
    },
    {
      "epoch": 81.87272727272727,
      "grad_norm": 3.013705015182495,
      "learning_rate": 1.9061402047871834e-08,
      "loss": 0.5796,
      "step": 327
    },
    {
      "epoch": 82.0,
      "grad_norm": 3.9129855632781982,
      "learning_rate": 1.8826441272225223e-08,
      "loss": 0.5744,
      "step": 328
    },
    {
      "epoch": 82.2909090909091,
      "grad_norm": 3.432311773300171,
      "learning_rate": 1.8594235253127372e-08,
      "loss": 0.6681,
      "step": 329
    },
    {
      "epoch": 82.58181818181818,
      "grad_norm": 3.2865946292877197,
      "learning_rate": 1.8364801673965642e-08,
      "loss": 0.7103,
      "step": 330
    },
    {
      "epoch": 82.87272727272727,
      "grad_norm": 3.1664698123931885,
      "learning_rate": 1.8138158006995365e-08,
      "loss": 0.5704,
      "step": 331
    },
    {
      "epoch": 83.0,
      "grad_norm": 3.5827629566192627,
      "learning_rate": 1.7914321512009295e-08,
      "loss": 0.7128,
      "step": 332
    },
    {
      "epoch": 83.2909090909091,
      "grad_norm": 3.228314161300659,
      "learning_rate": 1.7693309235023126e-08,
      "loss": 0.6072,
      "step": 333
    },
    {
      "epoch": 83.58181818181818,
      "grad_norm": 3.2263855934143066,
      "learning_rate": 1.7475138006977434e-08,
      "loss": 0.6525,
      "step": 334
    },
    {
      "epoch": 83.87272727272727,
      "grad_norm": 3.2911698818206787,
      "learning_rate": 1.7259824442455922e-08,
      "loss": 0.7323,
      "step": 335
    },
    {
      "epoch": 84.0,
      "grad_norm": 3.629072666168213,
      "learning_rate": 1.704738493842015e-08,
      "loss": 0.6316,
      "step": 336
    },
    {
      "epoch": 84.0,
      "eval_loss": 0.6387331485748291,
      "eval_runtime": 0.761,
      "eval_samples_per_second": 17.082,
      "eval_steps_per_second": 17.082,
      "step": 336
    },
    {
      "epoch": 84.2909090909091,
      "grad_norm": 3.0507638454437256,
      "learning_rate": 1.6837835672960833e-08,
      "loss": 0.6576,
      "step": 337
    },
    {
      "epoch": 84.58181818181818,
      "grad_norm": 3.1356823444366455,
      "learning_rate": 1.663119260406585e-08,
      "loss": 0.6216,
      "step": 338
    },
    {
      "epoch": 84.87272727272727,
      "grad_norm": 3.6667861938476562,
      "learning_rate": 1.642747146840495e-08,
      "loss": 0.6975,
      "step": 339
    },
    {
      "epoch": 85.0,
      "grad_norm": 3.397160768508911,
      "learning_rate": 1.6226687780131337e-08,
      "loss": 0.6022,
      "step": 340
    },
    {
      "epoch": 85.2909090909091,
      "grad_norm": 3.163560152053833,
      "learning_rate": 1.602885682970026e-08,
      "loss": 0.6316,
      "step": 341
    },
    {
      "epoch": 85.58181818181818,
      "grad_norm": 3.1787586212158203,
      "learning_rate": 1.5833993682704515e-08,
      "loss": 0.6725,
      "step": 342
    },
    {
      "epoch": 85.87272727272727,
      "grad_norm": 3.379927158355713,
      "learning_rate": 1.5642113178727193e-08,
      "loss": 0.6319,
      "step": 343
    },
    {
      "epoch": 86.0,
      "grad_norm": 3.783219337463379,
      "learning_rate": 1.5453229930211566e-08,
      "loss": 0.72,
      "step": 344
    },
    {
      "epoch": 86.2909090909091,
      "grad_norm": 3.678173065185547,
      "learning_rate": 1.5267358321348288e-08,
      "loss": 0.7343,
      "step": 345
    },
    {
      "epoch": 86.58181818181818,
      "grad_norm": 3.0277297496795654,
      "learning_rate": 1.5084512506980025e-08,
      "loss": 0.6112,
      "step": 346
    },
    {
      "epoch": 86.87272727272727,
      "grad_norm": 2.9994826316833496,
      "learning_rate": 1.490470641152345e-08,
      "loss": 0.6377,
      "step": 347
    },
    {
      "epoch": 87.0,
      "grad_norm": 3.4868507385253906,
      "learning_rate": 1.4727953727908877e-08,
      "loss": 0.603,
      "step": 348
    },
    {
      "epoch": 87.2909090909091,
      "grad_norm": 3.3681607246398926,
      "learning_rate": 1.4554267916537493e-08,
      "loss": 0.6832,
      "step": 349
    },
    {
      "epoch": 87.58181818181818,
      "grad_norm": 3.123229742050171,
      "learning_rate": 1.438366220425628e-08,
      "loss": 0.6086,
      "step": 350
    },
    {
      "epoch": 87.87272727272727,
      "grad_norm": 3.2728774547576904,
      "learning_rate": 1.4216149583350753e-08,
      "loss": 0.6735,
      "step": 351
    },
    {
      "epoch": 88.0,
      "grad_norm": 3.185558319091797,
      "learning_rate": 1.405174281055556e-08,
      "loss": 0.6232,
      "step": 352
    },
    {
      "epoch": 88.0,
      "eval_loss": 0.6380465030670166,
      "eval_runtime": 0.6934,
      "eval_samples_per_second": 18.748,
      "eval_steps_per_second": 18.748,
      "step": 352
    },
    {
      "epoch": 88.2909090909091,
      "grad_norm": 3.0658743381500244,
      "learning_rate": 1.3890454406082957e-08,
      "loss": 0.6388,
      "step": 353
    },
    {
      "epoch": 88.58181818181818,
      "grad_norm": 3.2591443061828613,
      "learning_rate": 1.3732296652669417e-08,
      "loss": 0.6658,
      "step": 354
    },
    {
      "epoch": 88.87272727272727,
      "grad_norm": 3.252021551132202,
      "learning_rate": 1.3577281594640182e-08,
      "loss": 0.6671,
      "step": 355
    },
    {
      "epoch": 89.0,
      "grad_norm": 3.7164716720581055,
      "learning_rate": 1.3425421036992096e-08,
      "loss": 0.6185,
      "step": 356
    },
    {
      "epoch": 89.2909090909091,
      "grad_norm": 3.1124889850616455,
      "learning_rate": 1.327672654449457e-08,
      "loss": 0.6648,
      "step": 357
    },
    {
      "epoch": 89.58181818181818,
      "grad_norm": 3.4343039989471436,
      "learning_rate": 1.3131209440808898e-08,
      "loss": 0.6731,
      "step": 358
    },
    {
      "epoch": 89.87272727272727,
      "grad_norm": 2.8693790435791016,
      "learning_rate": 1.2988880807625927e-08,
      "loss": 0.5938,
      "step": 359
    },
    {
      "epoch": 90.0,
      "grad_norm": 4.409383773803711,
      "learning_rate": 1.284975148382211e-08,
      "loss": 0.7107,
      "step": 360
    },
    {
      "epoch": 90.2909090909091,
      "grad_norm": 3.080493450164795,
      "learning_rate": 1.2713832064634124e-08,
      "loss": 0.6798,
      "step": 361
    },
    {
      "epoch": 90.58181818181818,
      "grad_norm": 2.9749112129211426,
      "learning_rate": 1.2581132900851971e-08,
      "loss": 0.6249,
      "step": 362
    },
    {
      "epoch": 90.87272727272727,
      "grad_norm": 3.626858949661255,
      "learning_rate": 1.2451664098030743e-08,
      "loss": 0.6616,
      "step": 363
    },
    {
      "epoch": 91.0,
      "grad_norm": 3.859955310821533,
      "learning_rate": 1.232543551572103e-08,
      "loss": 0.6418,
      "step": 364
    },
    {
      "epoch": 91.2909090909091,
      "grad_norm": 3.4120571613311768,
      "learning_rate": 1.2202456766718091e-08,
      "loss": 0.707,
      "step": 365
    },
    {
      "epoch": 91.58181818181818,
      "grad_norm": 2.8337764739990234,
      "learning_rate": 1.2082737216329793e-08,
      "loss": 0.6083,
      "step": 366
    },
    {
      "epoch": 91.87272727272727,
      "grad_norm": 3.1251420974731445,
      "learning_rate": 1.1966285981663406e-08,
      "loss": 0.6327,
      "step": 367
    },
    {
      "epoch": 92.0,
      "grad_norm": 4.236498832702637,
      "learning_rate": 1.1853111930931313e-08,
      "loss": 0.6545,
      "step": 368
    },
    {
      "epoch": 92.0,
      "eval_loss": 0.6371535062789917,
      "eval_runtime": 0.7492,
      "eval_samples_per_second": 17.353,
      "eval_steps_per_second": 17.353,
      "step": 368
    },
    {
      "epoch": 92.2909090909091,
      "grad_norm": 3.1066205501556396,
      "learning_rate": 1.174322368277565e-08,
      "loss": 0.5913,
      "step": 369
    },
    {
      "epoch": 92.58181818181818,
      "grad_norm": 3.334169626235962,
      "learning_rate": 1.1636629605611967e-08,
      "loss": 0.6869,
      "step": 370
    },
    {
      "epoch": 92.87272727272727,
      "grad_norm": 3.105184316635132,
      "learning_rate": 1.1533337816991931e-08,
      "loss": 0.6699,
      "step": 371
    },
    {
      "epoch": 93.0,
      "grad_norm": 3.6188406944274902,
      "learning_rate": 1.1433356182985158e-08,
      "loss": 0.658,
      "step": 372
    },
    {
      "epoch": 93.2909090909091,
      "grad_norm": 3.376845598220825,
      "learning_rate": 1.133669231758016e-08,
      "loss": 0.7064,
      "step": 373
    },
    {
      "epoch": 93.58181818181818,
      "grad_norm": 3.327584981918335,
      "learning_rate": 1.1243353582104555e-08,
      "loss": 0.6268,
      "step": 374
    },
    {
      "epoch": 93.87272727272727,
      "grad_norm": 2.9586260318756104,
      "learning_rate": 1.115334708466442e-08,
      "loss": 0.6366,
      "step": 375
    },
    {
      "epoch": 94.0,
      "grad_norm": 3.442078113555908,
      "learning_rate": 1.1066679679602998e-08,
      "loss": 0.5847,
      "step": 376
    },
    {
      "epoch": 94.2909090909091,
      "grad_norm": 2.9368817806243896,
      "learning_rate": 1.0983357966978745e-08,
      "loss": 0.6578,
      "step": 377
    },
    {
      "epoch": 94.58181818181818,
      "grad_norm": 3.0193896293640137,
      "learning_rate": 1.0903388292062667e-08,
      "loss": 0.6523,
      "step": 378
    },
    {
      "epoch": 94.87272727272727,
      "grad_norm": 3.727072238922119,
      "learning_rate": 1.0826776744855121e-08,
      "loss": 0.6582,
      "step": 379
    },
    {
      "epoch": 95.0,
      "grad_norm": 3.880216598510742,
      "learning_rate": 1.0753529159622047e-08,
      "loss": 0.6047,
      "step": 380
    },
    {
      "epoch": 95.2909090909091,
      "grad_norm": 3.0474436283111572,
      "learning_rate": 1.068365111445064e-08,
      "loss": 0.6078,
      "step": 381
    },
    {
      "epoch": 95.58181818181818,
      "grad_norm": 3.09653639793396,
      "learning_rate": 1.0617147930824585e-08,
      "loss": 0.6347,
      "step": 382
    },
    {
      "epoch": 95.87272727272727,
      "grad_norm": 3.2464277744293213,
      "learning_rate": 1.0554024673218806e-08,
      "loss": 0.6709,
      "step": 383
    },
    {
      "epoch": 96.0,
      "grad_norm": 3.866807699203491,
      "learning_rate": 1.0494286148713743e-08,
      "loss": 0.7126,
      "step": 384
    },
    {
      "epoch": 96.0,
      "eval_loss": 0.6363654732704163,
      "eval_runtime": 0.7605,
      "eval_samples_per_second": 17.094,
      "eval_steps_per_second": 17.094,
      "step": 384
    },
    {
      "epoch": 96.2909090909091,
      "grad_norm": 3.1493375301361084,
      "learning_rate": 1.0437936906629333e-08,
      "loss": 0.5571,
      "step": 385
    },
    {
      "epoch": 96.58181818181818,
      "grad_norm": 3.1544456481933594,
      "learning_rate": 1.0384981238178533e-08,
      "loss": 0.7043,
      "step": 386
    },
    {
      "epoch": 96.87272727272727,
      "grad_norm": 3.5196638107299805,
      "learning_rate": 1.033542317614051e-08,
      "loss": 0.6956,
      "step": 387
    },
    {
      "epoch": 97.0,
      "grad_norm": 2.829664707183838,
      "learning_rate": 1.0289266494553564e-08,
      "loss": 0.5839,
      "step": 388
    },
    {
      "epoch": 97.2909090909091,
      "grad_norm": 3.240220308303833,
      "learning_rate": 1.0246514708427701e-08,
      "loss": 0.629,
      "step": 389
    },
    {
      "epoch": 97.58181818181818,
      "grad_norm": 3.419234275817871,
      "learning_rate": 1.0207171073476952e-08,
      "loss": 0.7125,
      "step": 390
    },
    {
      "epoch": 97.87272727272727,
      "grad_norm": 3.266242742538452,
      "learning_rate": 1.017123858587145e-08,
      "loss": 0.7004,
      "step": 391
    },
    {
      "epoch": 98.0,
      "grad_norm": 2.8885867595672607,
      "learning_rate": 1.0138719982009241e-08,
      "loss": 0.4986,
      "step": 392
    },
    {
      "epoch": 98.2909090909091,
      "grad_norm": 3.4574053287506104,
      "learning_rate": 1.0109617738307912e-08,
      "loss": 0.7095,
      "step": 393
    },
    {
      "epoch": 98.58181818181818,
      "grad_norm": 3.2674267292022705,
      "learning_rate": 1.0083934071015988e-08,
      "loss": 0.5806,
      "step": 394
    },
    {
      "epoch": 98.87272727272727,
      "grad_norm": 2.897749423980713,
      "learning_rate": 1.0061670936044179e-08,
      "loss": 0.6434,
      "step": 395
    },
    {
      "epoch": 99.0,
      "grad_norm": 3.9228150844573975,
      "learning_rate": 1.0042830028816398e-08,
      "loss": 0.7094,
      "step": 396
    },
    {
      "epoch": 99.2909090909091,
      "grad_norm": 2.946876287460327,
      "learning_rate": 1.002741278414069e-08,
      "loss": 0.5678,
      "step": 397
    },
    {
      "epoch": 99.58181818181818,
      "grad_norm": 2.9825222492218018,
      "learning_rate": 1.0015420376099922e-08,
      "loss": 0.6347,
      "step": 398
    },
    {
      "epoch": 99.87272727272727,
      "grad_norm": 3.46803879737854,
      "learning_rate": 1.0006853717962394e-08,
      "loss": 0.7428,
      "step": 399
    },
    {
      "epoch": 100.0,
      "grad_norm": 4.149415969848633,
      "learning_rate": 1.0001713462112291e-08,
      "loss": 0.6465,
      "step": 400
    },
    {
      "epoch": 100.0,
      "eval_loss": 0.6363555192947388,
      "eval_runtime": 0.7511,
      "eval_samples_per_second": 17.308,
      "eval_steps_per_second": 17.308,
      "step": 400
    },
    {
      "epoch": 100.0,
      "step": 400,
      "total_flos": 1.34153286008832e+17,
      "train_loss": 0.7119150696694851,
      "train_runtime": 2950.0217,
      "train_samples_per_second": 3.729,
      "train_steps_per_second": 0.136
    }
  ],
  "logging_steps": 1,
  "max_steps": 400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 16,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.34153286008832e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}