{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.13063357282821686,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0013063357282821686,
      "grad_norm": 2.7966933250427246,
      "learning_rate": 1e-05,
      "loss": 1.3025,
      "step": 1
    },
    {
      "epoch": 0.0013063357282821686,
      "eval_loss": 1.3974988460540771,
      "eval_runtime": 9.4606,
      "eval_samples_per_second": 272.498,
      "eval_steps_per_second": 8.562,
      "step": 1
    },
    {
      "epoch": 0.002612671456564337,
      "grad_norm": 3.432299852371216,
      "learning_rate": 2e-05,
      "loss": 1.4064,
      "step": 2
    },
    {
      "epoch": 0.003919007184846506,
      "grad_norm": 3.0004944801330566,
      "learning_rate": 3e-05,
      "loss": 1.392,
      "step": 3
    },
    {
      "epoch": 0.005225342913128674,
      "grad_norm": 2.326035737991333,
      "learning_rate": 4e-05,
      "loss": 1.3005,
      "step": 4
    },
    {
      "epoch": 0.006531678641410843,
      "grad_norm": 0.9703150987625122,
      "learning_rate": 5e-05,
      "loss": 1.3609,
      "step": 5
    },
    {
      "epoch": 0.007838014369693011,
      "grad_norm": 0.906290590763092,
      "learning_rate": 6e-05,
      "loss": 1.3509,
      "step": 6
    },
    {
      "epoch": 0.009144350097975179,
      "grad_norm": 0.7477251887321472,
      "learning_rate": 7e-05,
      "loss": 1.3931,
      "step": 7
    },
    {
      "epoch": 0.010450685826257348,
      "grad_norm": 0.7451111078262329,
      "learning_rate": 8e-05,
      "loss": 1.4216,
      "step": 8
    },
    {
      "epoch": 0.011757021554539516,
      "grad_norm": 0.6804422736167908,
      "learning_rate": 9e-05,
      "loss": 1.3584,
      "step": 9
    },
    {
      "epoch": 0.011757021554539516,
      "eval_loss": 1.30021333694458,
      "eval_runtime": 9.4409,
      "eval_samples_per_second": 273.066,
      "eval_steps_per_second": 8.58,
      "step": 9
    },
    {
      "epoch": 0.013063357282821686,
      "grad_norm": 0.533569872379303,
      "learning_rate": 0.0001,
      "loss": 1.2749,
      "step": 10
    },
    {
      "epoch": 0.014369693011103853,
      "grad_norm": 0.5280372500419617,
      "learning_rate": 9.99695413509548e-05,
      "loss": 1.2876,
      "step": 11
    },
    {
      "epoch": 0.015676028739386023,
      "grad_norm": 0.5439796447753906,
      "learning_rate": 9.987820251299122e-05,
      "loss": 1.2296,
      "step": 12
    },
    {
      "epoch": 0.016982364467668192,
      "grad_norm": 0.8776497840881348,
      "learning_rate": 9.972609476841367e-05,
      "loss": 1.2337,
      "step": 13
    },
    {
      "epoch": 0.018288700195950358,
      "grad_norm": 0.5302695631980896,
      "learning_rate": 9.951340343707852e-05,
      "loss": 1.1875,
      "step": 14
    },
    {
      "epoch": 0.019595035924232528,
      "grad_norm": 0.4204191565513611,
      "learning_rate": 9.924038765061042e-05,
      "loss": 1.1547,
      "step": 15
    },
    {
      "epoch": 0.020901371652514697,
      "grad_norm": 0.3968534469604492,
      "learning_rate": 9.890738003669029e-05,
      "loss": 1.185,
      "step": 16
    },
    {
      "epoch": 0.022207707380796866,
      "grad_norm": 0.41285037994384766,
      "learning_rate": 9.851478631379982e-05,
      "loss": 1.1338,
      "step": 17
    },
    {
      "epoch": 0.023514043109079032,
      "grad_norm": 0.4262624680995941,
      "learning_rate": 9.806308479691595e-05,
      "loss": 1.1731,
      "step": 18
    },
    {
      "epoch": 0.023514043109079032,
      "eval_loss": 1.133678674697876,
      "eval_runtime": 9.442,
      "eval_samples_per_second": 273.035,
      "eval_steps_per_second": 8.579,
      "step": 18
    },
    {
      "epoch": 0.024820378837361202,
      "grad_norm": 0.3659999966621399,
      "learning_rate": 9.755282581475769e-05,
      "loss": 1.1046,
      "step": 19
    },
    {
      "epoch": 0.02612671456564337,
      "grad_norm": 0.36007174849510193,
      "learning_rate": 9.698463103929542e-05,
      "loss": 1.0745,
      "step": 20
    },
    {
      "epoch": 0.027433050293925537,
      "grad_norm": 0.5450918674468994,
      "learning_rate": 9.635919272833938e-05,
      "loss": 1.1434,
      "step": 21
    },
    {
      "epoch": 0.028739386022207707,
      "grad_norm": 0.4918366074562073,
      "learning_rate": 9.567727288213005e-05,
      "loss": 1.1579,
      "step": 22
    },
    {
      "epoch": 0.030045721750489876,
      "grad_norm": 0.5449302196502686,
      "learning_rate": 9.493970231495835e-05,
      "loss": 1.0585,
      "step": 23
    },
    {
      "epoch": 0.031352057478772045,
      "grad_norm": 0.3714740574359894,
      "learning_rate": 9.414737964294636e-05,
      "loss": 1.0611,
      "step": 24
    },
    {
      "epoch": 0.032658393207054215,
      "grad_norm": 0.40191322565078735,
      "learning_rate": 9.330127018922194e-05,
      "loss": 1.1165,
      "step": 25
    },
    {
      "epoch": 0.033964728935336384,
      "grad_norm": 0.38914942741394043,
      "learning_rate": 9.24024048078213e-05,
      "loss": 1.052,
      "step": 26
    },
    {
      "epoch": 0.03527106466361855,
      "grad_norm": 0.31749430298805237,
      "learning_rate": 9.145187862775209e-05,
      "loss": 1.0732,
      "step": 27
    },
    {
      "epoch": 0.03527106466361855,
      "eval_loss": 1.076000452041626,
      "eval_runtime": 9.4827,
      "eval_samples_per_second": 271.862,
      "eval_steps_per_second": 8.542,
      "step": 27
    },
    {
      "epoch": 0.036577400391900716,
      "grad_norm": 0.32451197504997253,
      "learning_rate": 9.045084971874738e-05,
      "loss": 1.0701,
      "step": 28
    },
    {
      "epoch": 0.037883736120182886,
      "grad_norm": 0.32556551694869995,
      "learning_rate": 8.940053768033609e-05,
      "loss": 1.1153,
      "step": 29
    },
    {
      "epoch": 0.039190071848465055,
      "grad_norm": 0.34510084986686707,
      "learning_rate": 8.83022221559489e-05,
      "loss": 1.0123,
      "step": 30
    },
    {
      "epoch": 0.040496407576747225,
      "grad_norm": 0.39450201392173767,
      "learning_rate": 8.715724127386972e-05,
      "loss": 1.1092,
      "step": 31
    },
    {
      "epoch": 0.041802743305029394,
      "grad_norm": 0.31846481561660767,
      "learning_rate": 8.596699001693255e-05,
      "loss": 1.0717,
      "step": 32
    },
    {
      "epoch": 0.04310907903331156,
      "grad_norm": 0.27703917026519775,
      "learning_rate": 8.473291852294987e-05,
      "loss": 1.0751,
      "step": 33
    },
    {
      "epoch": 0.04441541476159373,
      "grad_norm": 0.3374369144439697,
      "learning_rate": 8.345653031794292e-05,
      "loss": 1.0965,
      "step": 34
    },
    {
      "epoch": 0.045721750489875895,
      "grad_norm": 0.3880363404750824,
      "learning_rate": 8.213938048432697e-05,
      "loss": 1.0428,
      "step": 35
    },
    {
      "epoch": 0.047028086218158065,
      "grad_norm": 0.3148273229598999,
      "learning_rate": 8.07830737662829e-05,
      "loss": 1.0309,
      "step": 36
    },
    {
      "epoch": 0.047028086218158065,
      "eval_loss": 1.0441398620605469,
      "eval_runtime": 9.4448,
      "eval_samples_per_second": 272.954,
      "eval_steps_per_second": 8.576,
      "step": 36
    },
    {
      "epoch": 0.048334421946440234,
      "grad_norm": 0.2607744336128235,
      "learning_rate": 7.938926261462366e-05,
      "loss": 1.0598,
      "step": 37
    },
    {
      "epoch": 0.049640757674722404,
      "grad_norm": 0.5195264220237732,
      "learning_rate": 7.795964517353735e-05,
      "loss": 1.0636,
      "step": 38
    },
    {
      "epoch": 0.05094709340300457,
      "grad_norm": 0.31607645750045776,
      "learning_rate": 7.649596321166024e-05,
      "loss": 1.0021,
      "step": 39
    },
    {
      "epoch": 0.05225342913128674,
      "grad_norm": 0.3106956481933594,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.0262,
      "step": 40
    },
    {
      "epoch": 0.05355976485956891,
      "grad_norm": 0.29179874062538147,
      "learning_rate": 7.347357813929454e-05,
      "loss": 1.0262,
      "step": 41
    },
    {
      "epoch": 0.054866100587851074,
      "grad_norm": 0.3367427885532379,
      "learning_rate": 7.191855733945387e-05,
      "loss": 1.0069,
      "step": 42
    },
    {
      "epoch": 0.056172436316133244,
      "grad_norm": 0.2948434352874756,
      "learning_rate": 7.033683215379002e-05,
      "loss": 1.025,
      "step": 43
    },
    {
      "epoch": 0.05747877204441541,
      "grad_norm": 0.2905190885066986,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.996,
      "step": 44
    },
    {
      "epoch": 0.05878510777269758,
      "grad_norm": 0.3415267765522003,
      "learning_rate": 6.710100716628344e-05,
      "loss": 0.9931,
      "step": 45
    },
    {
      "epoch": 0.05878510777269758,
      "eval_loss": 1.0246894359588623,
      "eval_runtime": 9.4671,
      "eval_samples_per_second": 272.311,
      "eval_steps_per_second": 8.556,
      "step": 45
    },
    {
      "epoch": 0.06009144350097975,
      "grad_norm": 0.28528183698654175,
      "learning_rate": 6.545084971874738e-05,
      "loss": 1.0149,
      "step": 46
    },
    {
      "epoch": 0.06139777922926192,
      "grad_norm": 0.28322818875312805,
      "learning_rate": 6.378186779084995e-05,
      "loss": 1.0145,
      "step": 47
    },
    {
      "epoch": 0.06270411495754409,
      "grad_norm": 0.31846946477890015,
      "learning_rate": 6.209609477998338e-05,
      "loss": 1.0618,
      "step": 48
    },
    {
      "epoch": 0.06401045068582625,
      "grad_norm": 0.3214743137359619,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 1.0372,
      "step": 49
    },
    {
      "epoch": 0.06531678641410843,
      "grad_norm": 0.30648520588874817,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.9778,
      "step": 50
    },
    {
      "epoch": 0.06662312214239059,
      "grad_norm": 0.28608712553977966,
      "learning_rate": 5.695865504800327e-05,
      "loss": 0.9961,
      "step": 51
    },
    {
      "epoch": 0.06792945787067277,
      "grad_norm": 0.2914606034755707,
      "learning_rate": 5.522642316338268e-05,
      "loss": 1.0403,
      "step": 52
    },
    {
      "epoch": 0.06923579359895493,
      "grad_norm": 0.2740266025066376,
      "learning_rate": 5.348782368720626e-05,
      "loss": 1.0018,
      "step": 53
    },
    {
      "epoch": 0.0705421293272371,
      "grad_norm": 0.2767113745212555,
      "learning_rate": 5.174497483512506e-05,
      "loss": 1.006,
      "step": 54
    },
    {
      "epoch": 0.0705421293272371,
      "eval_loss": 1.0122777223587036,
      "eval_runtime": 9.487,
      "eval_samples_per_second": 271.739,
      "eval_steps_per_second": 8.538,
      "step": 54
    },
    {
      "epoch": 0.07184846505551927,
      "grad_norm": 0.30016326904296875,
      "learning_rate": 5e-05,
      "loss": 1.1104,
      "step": 55
    },
    {
      "epoch": 0.07315480078380143,
      "grad_norm": 0.3731127083301544,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.999,
      "step": 56
    },
    {
      "epoch": 0.07446113651208361,
      "grad_norm": 0.29544034600257874,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 1.0414,
      "step": 57
    },
    {
      "epoch": 0.07576747224036577,
      "grad_norm": 0.287106990814209,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.9942,
      "step": 58
    },
    {
      "epoch": 0.07707380796864795,
      "grad_norm": 0.3369489908218384,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 0.9757,
      "step": 59
    },
    {
      "epoch": 0.07838014369693011,
      "grad_norm": 0.2900789678096771,
      "learning_rate": 4.131759111665349e-05,
      "loss": 1.047,
      "step": 60
    },
    {
      "epoch": 0.07968647942521227,
      "grad_norm": 0.3095402419567108,
      "learning_rate": 3.960441545911204e-05,
      "loss": 1.0334,
      "step": 61
    },
    {
      "epoch": 0.08099281515349445,
      "grad_norm": 0.3259451389312744,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.9733,
      "step": 62
    },
    {
      "epoch": 0.08229915088177661,
      "grad_norm": 0.3178972601890564,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 1.0362,
      "step": 63
    },
    {
      "epoch": 0.08229915088177661,
      "eval_loss": 1.0038740634918213,
      "eval_runtime": 9.4702,
      "eval_samples_per_second": 272.221,
      "eval_steps_per_second": 8.553,
      "step": 63
    },
    {
      "epoch": 0.08360548661005879,
      "grad_norm": 0.3139575719833374,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.9568,
      "step": 64
    },
    {
      "epoch": 0.08491182233834095,
      "grad_norm": 0.2903759777545929,
      "learning_rate": 3.289899283371657e-05,
      "loss": 0.993,
      "step": 65
    },
    {
      "epoch": 0.08621815806662313,
      "grad_norm": 0.3004719018936157,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.9933,
      "step": 66
    },
    {
      "epoch": 0.08752449379490529,
      "grad_norm": 0.3223623037338257,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 1.0011,
      "step": 67
    },
    {
      "epoch": 0.08883082952318747,
      "grad_norm": 0.29526814818382263,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.895,
      "step": 68
    },
    {
      "epoch": 0.09013716525146963,
      "grad_norm": 0.30608242750167847,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 1.0278,
      "step": 69
    },
    {
      "epoch": 0.09144350097975179,
      "grad_norm": 0.2746555209159851,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.9357,
      "step": 70
    },
    {
      "epoch": 0.09274983670803397,
      "grad_norm": 0.3204227089881897,
      "learning_rate": 2.350403678833976e-05,
      "loss": 0.9548,
      "step": 71
    },
    {
      "epoch": 0.09405617243631613,
      "grad_norm": 0.3303455114364624,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 0.9782,
      "step": 72
    },
    {
      "epoch": 0.09405617243631613,
      "eval_loss": 0.9981375932693481,
      "eval_runtime": 9.4327,
      "eval_samples_per_second": 273.306,
      "eval_steps_per_second": 8.587,
      "step": 72
    },
    {
      "epoch": 0.0953625081645983,
      "grad_norm": 0.3481445610523224,
      "learning_rate": 2.061073738537635e-05,
      "loss": 1.0681,
      "step": 73
    },
    {
      "epoch": 0.09666884389288047,
      "grad_norm": 0.32196563482284546,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 0.97,
      "step": 74
    },
    {
      "epoch": 0.09797517962116264,
      "grad_norm": 0.26915478706359863,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 0.9716,
      "step": 75
    },
    {
      "epoch": 0.09928151534944481,
      "grad_norm": 0.31112876534461975,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.9679,
      "step": 76
    },
    {
      "epoch": 0.10058785107772697,
      "grad_norm": 0.2868760824203491,
      "learning_rate": 1.526708147705013e-05,
      "loss": 0.9714,
      "step": 77
    },
    {
      "epoch": 0.10189418680600915,
      "grad_norm": 0.2732284963130951,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 1.0294,
      "step": 78
    },
    {
      "epoch": 0.10320052253429131,
      "grad_norm": 0.3132527768611908,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 0.9823,
      "step": 79
    },
    {
      "epoch": 0.10450685826257348,
      "grad_norm": 0.34114107489585876,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 1.0015,
      "step": 80
    },
    {
      "epoch": 0.10581319399085565,
      "grad_norm": 0.3598445653915405,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 1.0128,
      "step": 81
    },
    {
      "epoch": 0.10581319399085565,
      "eval_loss": 0.9954747557640076,
      "eval_runtime": 9.4558,
      "eval_samples_per_second": 272.636,
      "eval_steps_per_second": 8.566,
      "step": 81
    },
    {
      "epoch": 0.10711952971913782,
      "grad_norm": 0.286565899848938,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.9809,
      "step": 82
    },
    {
      "epoch": 0.10842586544741999,
      "grad_norm": 0.3290863037109375,
      "learning_rate": 8.548121372247918e-06,
      "loss": 1.0103,
      "step": 83
    },
    {
      "epoch": 0.10973220117570215,
      "grad_norm": 0.29899007081985474,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.9652,
      "step": 84
    },
    {
      "epoch": 0.11103853690398433,
      "grad_norm": 0.3017849922180176,
      "learning_rate": 6.698729810778065e-06,
      "loss": 1.0355,
      "step": 85
    },
    {
      "epoch": 0.11234487263226649,
      "grad_norm": 0.30627602338790894,
      "learning_rate": 5.852620357053651e-06,
      "loss": 1.0068,
      "step": 86
    },
    {
      "epoch": 0.11365120836054866,
      "grad_norm": 0.29219281673431396,
      "learning_rate": 5.060297685041659e-06,
      "loss": 1.0003,
      "step": 87
    },
    {
      "epoch": 0.11495754408883083,
      "grad_norm": 0.29773980379104614,
      "learning_rate": 4.322727117869951e-06,
      "loss": 1.0151,
      "step": 88
    },
    {
      "epoch": 0.116263879817113,
      "grad_norm": 0.276870995759964,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 0.9312,
      "step": 89
    },
    {
      "epoch": 0.11757021554539517,
      "grad_norm": 0.2851667106151581,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 1.0124,
      "step": 90
    },
    {
      "epoch": 0.11757021554539517,
      "eval_loss": 0.9941932559013367,
      "eval_runtime": 9.4226,
      "eval_samples_per_second": 273.599,
      "eval_steps_per_second": 8.596,
      "step": 90
    },
    {
      "epoch": 0.11887655127367734,
      "grad_norm": 0.2754754424095154,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.9848,
      "step": 91
    },
    {
      "epoch": 0.1201828870019595,
      "grad_norm": 0.30135542154312134,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 0.9696,
      "step": 92
    },
    {
      "epoch": 0.12148922273024167,
      "grad_norm": 0.36859849095344543,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 1.0371,
      "step": 93
    },
    {
      "epoch": 0.12279555845852384,
      "grad_norm": 0.28066200017929077,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.9595,
      "step": 94
    },
    {
      "epoch": 0.124101894186806,
      "grad_norm": 0.28233689069747925,
      "learning_rate": 7.596123493895991e-07,
      "loss": 1.0157,
      "step": 95
    },
    {
      "epoch": 0.12540822991508818,
      "grad_norm": 0.27524614334106445,
      "learning_rate": 4.865965629214819e-07,
      "loss": 1.0362,
      "step": 96
    },
    {
      "epoch": 0.12671456564337036,
      "grad_norm": 0.3097120523452759,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 0.995,
      "step": 97
    },
    {
      "epoch": 0.1280209013716525,
      "grad_norm": 0.2737129330635071,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 0.9566,
      "step": 98
    },
    {
      "epoch": 0.12932723709993468,
      "grad_norm": 0.3093501925468445,
      "learning_rate": 3.04586490452119e-08,
      "loss": 1.0087,
      "step": 99
    },
    {
      "epoch": 0.12932723709993468,
      "eval_loss": 0.9940328598022461,
      "eval_runtime": 9.4334,
      "eval_samples_per_second": 273.283,
      "eval_steps_per_second": 8.586,
      "step": 99
    },
    {
      "epoch": 0.13063357282821686,
      "grad_norm": 0.3667088449001312,
      "learning_rate": 0.0,
      "loss": 1.0538,
      "step": 100
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.11747508813824e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}