{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.013735320376347778,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0001373532037634778,
      "grad_norm": 3.2098865509033203,
      "learning_rate": 1e-05,
      "loss": 4.1873,
      "step": 1
    },
    {
      "epoch": 0.0001373532037634778,
      "eval_loss": 5.127926349639893,
      "eval_runtime": 1493.3416,
      "eval_samples_per_second": 4.106,
      "eval_steps_per_second": 0.514,
      "step": 1
    },
    {
      "epoch": 0.0002747064075269556,
      "grad_norm": 6.169426441192627,
      "learning_rate": 2e-05,
      "loss": 5.9119,
      "step": 2
    },
    {
      "epoch": 0.00041205961129043336,
      "grad_norm": 4.486110687255859,
      "learning_rate": 3e-05,
      "loss": 5.4337,
      "step": 3
    },
    {
      "epoch": 0.0005494128150539112,
      "grad_norm": 3.3929038047790527,
      "learning_rate": 4e-05,
      "loss": 4.3407,
      "step": 4
    },
    {
      "epoch": 0.000686766018817389,
      "grad_norm": 4.019403457641602,
      "learning_rate": 5e-05,
      "loss": 4.6766,
      "step": 5
    },
    {
      "epoch": 0.0008241192225808667,
      "grad_norm": 4.6286540031433105,
      "learning_rate": 6e-05,
      "loss": 4.734,
      "step": 6
    },
    {
      "epoch": 0.0009614724263443445,
      "grad_norm": 4.8122735023498535,
      "learning_rate": 7e-05,
      "loss": 4.4077,
      "step": 7
    },
    {
      "epoch": 0.0010988256301078224,
      "grad_norm": 5.264736652374268,
      "learning_rate": 8e-05,
      "loss": 4.4438,
      "step": 8
    },
    {
      "epoch": 0.0012361788338713,
      "grad_norm": 6.661530017852783,
      "learning_rate": 9e-05,
      "loss": 4.7209,
      "step": 9
    },
    {
      "epoch": 0.0012361788338713,
      "eval_loss": 3.330467700958252,
      "eval_runtime": 1499.4096,
      "eval_samples_per_second": 4.089,
      "eval_steps_per_second": 0.512,
      "step": 9
    },
    {
      "epoch": 0.001373532037634778,
      "grad_norm": 5.09337043762207,
      "learning_rate": 0.0001,
      "loss": 3.358,
      "step": 10
    },
    {
      "epoch": 0.0015108852413982556,
      "grad_norm": 3.219647169113159,
      "learning_rate": 9.99695413509548e-05,
      "loss": 2.5805,
      "step": 11
    },
    {
      "epoch": 0.0016482384451617334,
      "grad_norm": 4.629909992218018,
      "learning_rate": 9.987820251299122e-05,
      "loss": 2.6967,
      "step": 12
    },
    {
      "epoch": 0.001785591648925211,
      "grad_norm": 4.305159568786621,
      "learning_rate": 9.972609476841367e-05,
      "loss": 2.4067,
      "step": 13
    },
    {
      "epoch": 0.001922944852688689,
      "grad_norm": 4.787576198577881,
      "learning_rate": 9.951340343707852e-05,
      "loss": 2.1925,
      "step": 14
    },
    {
      "epoch": 0.0020602980564521666,
      "grad_norm": 3.349574327468872,
      "learning_rate": 9.924038765061042e-05,
      "loss": 1.8936,
      "step": 15
    },
    {
      "epoch": 0.0021976512602156447,
      "grad_norm": 4.581472873687744,
      "learning_rate": 9.890738003669029e-05,
      "loss": 1.9205,
      "step": 16
    },
    {
      "epoch": 0.0023350044639791224,
      "grad_norm": 2.8699538707733154,
      "learning_rate": 9.851478631379982e-05,
      "loss": 1.5474,
      "step": 17
    },
    {
      "epoch": 0.0024723576677426,
      "grad_norm": 3.6284587383270264,
      "learning_rate": 9.806308479691595e-05,
      "loss": 1.4992,
      "step": 18
    },
    {
      "epoch": 0.0024723576677426,
      "eval_loss": 1.570014476776123,
      "eval_runtime": 1498.6079,
      "eval_samples_per_second": 4.091,
      "eval_steps_per_second": 0.512,
      "step": 18
    },
    {
      "epoch": 0.0026097108715060777,
      "grad_norm": 2.9308242797851562,
      "learning_rate": 9.755282581475769e-05,
      "loss": 1.4584,
      "step": 19
    },
    {
      "epoch": 0.002747064075269556,
      "grad_norm": 3.2358460426330566,
      "learning_rate": 9.698463103929542e-05,
      "loss": 1.6333,
      "step": 20
    },
    {
      "epoch": 0.0028844172790330335,
      "grad_norm": 2.4403843879699707,
      "learning_rate": 9.635919272833938e-05,
      "loss": 1.5618,
      "step": 21
    },
    {
      "epoch": 0.003021770482796511,
      "grad_norm": 2.0725436210632324,
      "learning_rate": 9.567727288213005e-05,
      "loss": 1.1391,
      "step": 22
    },
    {
      "epoch": 0.003159123686559989,
      "grad_norm": 1.9925494194030762,
      "learning_rate": 9.493970231495835e-05,
      "loss": 1.173,
      "step": 23
    },
    {
      "epoch": 0.003296476890323467,
      "grad_norm": 1.7557494640350342,
      "learning_rate": 9.414737964294636e-05,
      "loss": 1.0151,
      "step": 24
    },
    {
      "epoch": 0.0034338300940869445,
      "grad_norm": 2.060401439666748,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.9741,
      "step": 25
    },
    {
      "epoch": 0.003571183297850422,
      "grad_norm": 1.5280568599700928,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.7625,
      "step": 26
    },
    {
      "epoch": 0.0037085365016139003,
      "grad_norm": 2.04048490524292,
      "learning_rate": 9.145187862775209e-05,
      "loss": 1.0372,
      "step": 27
    },
    {
      "epoch": 0.0037085365016139003,
      "eval_loss": 1.0555644035339355,
      "eval_runtime": 1498.6144,
      "eval_samples_per_second": 4.091,
      "eval_steps_per_second": 0.512,
      "step": 27
    },
    {
      "epoch": 0.003845889705377378,
      "grad_norm": 2.096092462539673,
      "learning_rate": 9.045084971874738e-05,
      "loss": 1.0439,
      "step": 28
    },
    {
      "epoch": 0.003983242909140856,
      "grad_norm": 4.1492486000061035,
      "learning_rate": 8.940053768033609e-05,
      "loss": 1.1595,
      "step": 29
    },
    {
      "epoch": 0.004120596112904333,
      "grad_norm": 2.5651235580444336,
      "learning_rate": 8.83022221559489e-05,
      "loss": 1.0073,
      "step": 30
    },
    {
      "epoch": 0.004257949316667811,
      "grad_norm": 2.9436209201812744,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.7517,
      "step": 31
    },
    {
      "epoch": 0.0043953025204312894,
      "grad_norm": 2.7678189277648926,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.8658,
      "step": 32
    },
    {
      "epoch": 0.004532655724194767,
      "grad_norm": 2.158916711807251,
      "learning_rate": 8.473291852294987e-05,
      "loss": 0.9779,
      "step": 33
    },
    {
      "epoch": 0.004670008927958245,
      "grad_norm": 1.9354743957519531,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.619,
      "step": 34
    },
    {
      "epoch": 0.004807362131721722,
      "grad_norm": 2.492863893508911,
      "learning_rate": 8.213938048432697e-05,
      "loss": 0.709,
      "step": 35
    },
    {
      "epoch": 0.0049447153354852,
      "grad_norm": 1.5627326965332031,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.5263,
      "step": 36
    },
    {
      "epoch": 0.0049447153354852,
      "eval_loss": 0.6903370022773743,
      "eval_runtime": 1497.3825,
      "eval_samples_per_second": 4.094,
      "eval_steps_per_second": 0.512,
      "step": 36
    },
    {
      "epoch": 0.005082068539248678,
      "grad_norm": 2.2282795906066895,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.4985,
      "step": 37
    },
    {
      "epoch": 0.005219421743012155,
      "grad_norm": 4.293792247772217,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.7172,
      "step": 38
    },
    {
      "epoch": 0.0053567749467756335,
      "grad_norm": 3.52012038230896,
      "learning_rate": 7.649596321166024e-05,
      "loss": 0.5202,
      "step": 39
    },
    {
      "epoch": 0.005494128150539112,
      "grad_norm": 2.2252063751220703,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.4629,
      "step": 40
    },
    {
      "epoch": 0.005631481354302589,
      "grad_norm": 4.003269195556641,
      "learning_rate": 7.347357813929454e-05,
      "loss": 0.7875,
      "step": 41
    },
    {
      "epoch": 0.005768834558066067,
      "grad_norm": 3.8624725341796875,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.7379,
      "step": 42
    },
    {
      "epoch": 0.005906187761829545,
      "grad_norm": 4.79985237121582,
      "learning_rate": 7.033683215379002e-05,
      "loss": 0.5905,
      "step": 43
    },
    {
      "epoch": 0.006043540965593022,
      "grad_norm": 2.62689471244812,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.3897,
      "step": 44
    },
    {
      "epoch": 0.0061808941693565,
      "grad_norm": 2.633169412612915,
      "learning_rate": 6.710100716628344e-05,
      "loss": 0.4792,
      "step": 45
    },
    {
      "epoch": 0.0061808941693565,
      "eval_loss": 0.4047314524650574,
      "eval_runtime": 1497.4649,
      "eval_samples_per_second": 4.094,
      "eval_steps_per_second": 0.512,
      "step": 45
    },
    {
      "epoch": 0.006318247373119978,
      "grad_norm": 2.664032459259033,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.4602,
      "step": 46
    },
    {
      "epoch": 0.006455600576883456,
      "grad_norm": 2.5823395252227783,
      "learning_rate": 6.378186779084995e-05,
      "loss": 0.6398,
      "step": 47
    },
    {
      "epoch": 0.006592953780646934,
      "grad_norm": 2.515225648880005,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.3104,
      "step": 48
    },
    {
      "epoch": 0.006730306984410411,
      "grad_norm": 2.0998854637145996,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 0.3107,
      "step": 49
    },
    {
      "epoch": 0.006867660188173889,
      "grad_norm": 2.1281323432922363,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.2989,
      "step": 50
    },
    {
      "epoch": 0.007005013391937367,
      "grad_norm": 2.5533089637756348,
      "learning_rate": 5.695865504800327e-05,
      "loss": 0.2884,
      "step": 51
    },
    {
      "epoch": 0.007142366595700844,
      "grad_norm": 1.415504813194275,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.1548,
      "step": 52
    },
    {
      "epoch": 0.0072797197994643225,
      "grad_norm": 2.212552309036255,
      "learning_rate": 5.348782368720626e-05,
      "loss": 0.1524,
      "step": 53
    },
    {
      "epoch": 0.0074170730032278006,
      "grad_norm": 2.2554523944854736,
      "learning_rate": 5.174497483512506e-05,
      "loss": 0.1038,
      "step": 54
    },
    {
      "epoch": 0.0074170730032278006,
      "eval_loss": 0.24693803489208221,
      "eval_runtime": 1498.651,
      "eval_samples_per_second": 4.091,
      "eval_steps_per_second": 0.512,
      "step": 54
    },
    {
      "epoch": 0.007554426206991278,
      "grad_norm": 4.748584270477295,
      "learning_rate": 5e-05,
      "loss": 0.2923,
      "step": 55
    },
    {
      "epoch": 0.007691779410754756,
      "grad_norm": 3.5152838230133057,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.1623,
      "step": 56
    },
    {
      "epoch": 0.007829132614518234,
      "grad_norm": 3.639317274093628,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 0.1558,
      "step": 57
    },
    {
      "epoch": 0.007966485818281712,
      "grad_norm": 1.4831420183181763,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.0574,
      "step": 58
    },
    {
      "epoch": 0.008103839022045188,
      "grad_norm": 3.9552788734436035,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 0.1435,
      "step": 59
    },
    {
      "epoch": 0.008241192225808667,
      "grad_norm": 5.947100639343262,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.2129,
      "step": 60
    },
    {
      "epoch": 0.008378545429572145,
      "grad_norm": 4.984751224517822,
      "learning_rate": 3.960441545911204e-05,
      "loss": 0.1273,
      "step": 61
    },
    {
      "epoch": 0.008515898633335623,
      "grad_norm": 3.6993088722229004,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.1048,
      "step": 62
    },
    {
      "epoch": 0.0086532518370991,
      "grad_norm": 4.781378746032715,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 0.1657,
      "step": 63
    },
    {
      "epoch": 0.0086532518370991,
      "eval_loss": 0.18375518918037415,
      "eval_runtime": 1499.3899,
      "eval_samples_per_second": 4.089,
      "eval_steps_per_second": 0.512,
      "step": 63
    },
    {
      "epoch": 0.008790605040862579,
      "grad_norm": 3.943474054336548,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.1574,
      "step": 64
    },
    {
      "epoch": 0.008927958244626055,
      "grad_norm": 6.085975170135498,
      "learning_rate": 3.289899283371657e-05,
      "loss": 0.2722,
      "step": 65
    },
    {
      "epoch": 0.009065311448389533,
      "grad_norm": 6.4064717292785645,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.2512,
      "step": 66
    },
    {
      "epoch": 0.009202664652153011,
      "grad_norm": 2.87799072265625,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 0.108,
      "step": 67
    },
    {
      "epoch": 0.00934001785591649,
      "grad_norm": 8.213472366333008,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.472,
      "step": 68
    },
    {
      "epoch": 0.009477371059679968,
      "grad_norm": 3.442542791366577,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 0.0854,
      "step": 69
    },
    {
      "epoch": 0.009614724263443444,
      "grad_norm": 6.024069309234619,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.3747,
      "step": 70
    },
    {
      "epoch": 0.009752077467206922,
      "grad_norm": 2.477121114730835,
      "learning_rate": 2.350403678833976e-05,
      "loss": 0.0355,
      "step": 71
    },
    {
      "epoch": 0.0098894306709704,
      "grad_norm": 5.309243679046631,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 0.0862,
      "step": 72
    },
    {
      "epoch": 0.0098894306709704,
      "eval_loss": 0.1321064531803131,
      "eval_runtime": 1498.0742,
      "eval_samples_per_second": 4.093,
      "eval_steps_per_second": 0.512,
      "step": 72
    },
    {
      "epoch": 0.010026783874733878,
      "grad_norm": 6.38518762588501,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.0985,
      "step": 73
    },
    {
      "epoch": 0.010164137078497356,
      "grad_norm": 0.5211701989173889,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 0.0106,
      "step": 74
    },
    {
      "epoch": 0.010301490282260834,
      "grad_norm": 3.0065691471099854,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 0.1057,
      "step": 75
    },
    {
      "epoch": 0.01043884348602431,
      "grad_norm": 2.810657262802124,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.1014,
      "step": 76
    },
    {
      "epoch": 0.010576196689787789,
      "grad_norm": 2.7257590293884277,
      "learning_rate": 1.526708147705013e-05,
      "loss": 0.1751,
      "step": 77
    },
    {
      "epoch": 0.010713549893551267,
      "grad_norm": 2.120305061340332,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 0.1735,
      "step": 78
    },
    {
      "epoch": 0.010850903097314745,
      "grad_norm": 1.8682059049606323,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 0.0433,
      "step": 79
    },
    {
      "epoch": 0.010988256301078223,
      "grad_norm": 1.0465031862258911,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.1019,
      "step": 80
    },
    {
      "epoch": 0.011125609504841701,
      "grad_norm": 1.4673629999160767,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 0.0376,
      "step": 81
    },
    {
      "epoch": 0.011125609504841701,
      "eval_loss": 0.09728392958641052,
      "eval_runtime": 1498.9976,
      "eval_samples_per_second": 4.09,
      "eval_steps_per_second": 0.512,
      "step": 81
    },
    {
      "epoch": 0.011262962708605178,
      "grad_norm": 1.0142756700515747,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0356,
      "step": 82
    },
    {
      "epoch": 0.011400315912368656,
      "grad_norm": 1.2843859195709229,
      "learning_rate": 8.548121372247918e-06,
      "loss": 0.0466,
      "step": 83
    },
    {
      "epoch": 0.011537669116132134,
      "grad_norm": 4.476285457611084,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.1879,
      "step": 84
    },
    {
      "epoch": 0.011675022319895612,
      "grad_norm": 4.4546332359313965,
      "learning_rate": 6.698729810778065e-06,
      "loss": 0.1602,
      "step": 85
    },
    {
      "epoch": 0.01181237552365909,
      "grad_norm": 3.740358352661133,
      "learning_rate": 5.852620357053651e-06,
      "loss": 0.0667,
      "step": 86
    },
    {
      "epoch": 0.011949728727422566,
      "grad_norm": 1.2667216062545776,
      "learning_rate": 5.060297685041659e-06,
      "loss": 0.0298,
      "step": 87
    },
    {
      "epoch": 0.012087081931186044,
      "grad_norm": 1.5209143161773682,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.051,
      "step": 88
    },
    {
      "epoch": 0.012224435134949523,
      "grad_norm": 0.6315830945968628,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 0.0155,
      "step": 89
    },
    {
      "epoch": 0.012361788338713,
      "grad_norm": 2.6171979904174805,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.1119,
      "step": 90
    },
    {
      "epoch": 0.012361788338713,
      "eval_loss": 0.08409945666790009,
      "eval_runtime": 1498.4857,
      "eval_samples_per_second": 4.091,
      "eval_steps_per_second": 0.512,
      "step": 90
    },
    {
      "epoch": 0.012499141542476479,
      "grad_norm": 0.7829715013504028,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.0263,
      "step": 91
    },
    {
      "epoch": 0.012636494746239957,
      "grad_norm": 1.01038658618927,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 0.0265,
      "step": 92
    },
    {
      "epoch": 0.012773847950003433,
      "grad_norm": 0.4501360356807709,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 0.0112,
      "step": 93
    },
    {
      "epoch": 0.012911201153766911,
      "grad_norm": 0.3697107434272766,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.0121,
      "step": 94
    },
    {
      "epoch": 0.01304855435753039,
      "grad_norm": 1.2892919778823853,
      "learning_rate": 7.596123493895991e-07,
      "loss": 0.0215,
      "step": 95
    },
    {
      "epoch": 0.013185907561293867,
      "grad_norm": 1.3150074481964111,
      "learning_rate": 4.865965629214819e-07,
      "loss": 0.0209,
      "step": 96
    },
    {
      "epoch": 0.013323260765057346,
      "grad_norm": 0.9793353080749512,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 0.0262,
      "step": 97
    },
    {
      "epoch": 0.013460613968820822,
      "grad_norm": 2.4705395698547363,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 0.143,
      "step": 98
    },
    {
      "epoch": 0.0135979671725843,
      "grad_norm": 2.2415056228637695,
      "learning_rate": 3.04586490452119e-08,
      "loss": 0.1183,
      "step": 99
    },
    {
      "epoch": 0.0135979671725843,
      "eval_loss": 0.08113499730825424,
      "eval_runtime": 1498.1084,
      "eval_samples_per_second": 4.092,
      "eval_steps_per_second": 0.512,
      "step": 99
    },
    {
      "epoch": 0.013735320376347778,
      "grad_norm": 0.6419519186019897,
      "learning_rate": 0.0,
      "loss": 0.0252,
      "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": 5.235380292943872e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}