{
  "best_metric": 6.274591445922852,
  "best_model_checkpoint": "tiny_bert_km_5_v1/checkpoint-50000",
  "epoch": 25.0,
  "eval_steps": 10000,
  "global_step": 59550,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.20990764063811923,
      "grad_norm": 3.1972155570983887,
      "learning_rate": 5e-06,
      "loss": 11.3677,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 2.3281173706054688,
      "learning_rate": 1e-05,
      "loss": 9.6632,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 5.539621829986572,
      "learning_rate": 1.5e-05,
      "loss": 8.1526,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 5.410781383514404,
      "learning_rate": 2e-05,
      "loss": 7.5642,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 4.83973503112793,
      "learning_rate": 2.5e-05,
      "loss": 7.3548,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 3.246814250946045,
      "learning_rate": 3e-05,
      "loss": 7.2205,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 2.9470717906951904,
      "learning_rate": 3.5e-05,
      "loss": 7.1258,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 8.336593627929688,
      "learning_rate": 4e-05,
      "loss": 7.0365,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 4.815243721008301,
      "learning_rate": 4.5e-05,
      "loss": 6.9546,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 7.398708343505859,
      "learning_rate": 5e-05,
      "loss": 6.8642,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 2.613419532775879,
      "learning_rate": 5.500000000000001e-05,
      "loss": 6.7883,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 2.7809555530548096,
      "learning_rate": 6e-05,
      "loss": 6.7455,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 3.2186288833618164,
      "learning_rate": 6.500000000000001e-05,
      "loss": 6.701,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 1.9320868253707886,
      "learning_rate": 7e-05,
      "loss": 6.6635,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 3.2172491550445557,
      "learning_rate": 7.500000000000001e-05,
      "loss": 6.6028,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 2.1641852855682373,
      "learning_rate": 8e-05,
      "loss": 6.578,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 3.1677591800689697,
      "learning_rate": 8.5e-05,
      "loss": 6.5705,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 3.2325103282928467,
      "learning_rate": 9e-05,
      "loss": 6.5428,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 1.4941664934158325,
      "learning_rate": 9.5e-05,
      "loss": 6.5247,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 2.004518508911133,
      "learning_rate": 0.0001,
      "loss": 6.4666,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.15031024792008682,
      "eval_loss": 6.48938512802124,
      "eval_runtime": 0.7338,
      "eval_samples_per_second": 652.771,
      "eval_steps_per_second": 6.814,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 3.867565155029297,
      "learning_rate": 9.899091826437942e-05,
      "loss": 6.4551,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 2.2647457122802734,
      "learning_rate": 9.798183652875883e-05,
      "loss": 6.4407,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 1.393721580505371,
      "learning_rate": 9.697275479313825e-05,
      "loss": 6.4316,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 1.392156958580017,
      "learning_rate": 9.596367305751767e-05,
      "loss": 6.4073,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 2.2742393016815186,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.3642,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 2.0838069915771484,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.3525,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 1.76252019405365,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.3452,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 3.2777695655822754,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.3363,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 2.185377359390259,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.3095,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 2.041048526763916,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.2853,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 1.8073434829711914,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.28,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 2.208595037460327,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.2725,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 0.8567367792129517,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.2672,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 2.97397518157959,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.2397,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 1.0208189487457275,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.2212,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 2.4018354415893555,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.2235,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 1.77510404586792,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.2169,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 2.013493061065674,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.214,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 2.9306461811065674,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.1796,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 1.1682988405227661,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.1797,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.15247341395662056,
      "eval_loss": 6.309634685516357,
      "eval_runtime": 0.6966,
      "eval_samples_per_second": 687.613,
      "eval_steps_per_second": 7.178,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 1.8521651029586792,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.1745,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 1.1917282342910767,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.1767,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 1.105756163597107,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.1639,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 1.9092799425125122,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.1432,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 1.4869287014007568,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.1397,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 2.701622724533081,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.1424,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 1.5534542798995972,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.1365,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 1.4863390922546387,
      "learning_rate": 7.174571140262362e-05,
      "loss": 6.1205,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 1.7302894592285156,
      "learning_rate": 7.073662966700304e-05,
      "loss": 6.1171,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 2.6879420280456543,
      "learning_rate": 6.972754793138244e-05,
      "loss": 6.1065,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 1.3013033866882324,
      "learning_rate": 6.871846619576186e-05,
      "loss": 6.1066,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 1.0252957344055176,
      "learning_rate": 6.770938446014128e-05,
      "loss": 6.1062,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 1.5186256170272827,
      "learning_rate": 6.670030272452068e-05,
      "loss": 6.092,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 1.4508055448532104,
      "learning_rate": 6.56912209889001e-05,
      "loss": 6.0853,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 1.5264109373092651,
      "learning_rate": 6.468213925327952e-05,
      "loss": 6.0769,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 2.863663911819458,
      "learning_rate": 6.367305751765893e-05,
      "loss": 6.0752,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 1.5135879516601562,
      "learning_rate": 6.266397578203835e-05,
      "loss": 6.0727,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 2.314114570617676,
      "learning_rate": 6.165489404641777e-05,
      "loss": 6.062,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 2.374826192855835,
      "learning_rate": 6.064581231079718e-05,
      "loss": 6.0542,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 1.6091474294662476,
      "learning_rate": 5.963673057517659e-05,
      "loss": 6.059,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15359431401628565,
      "eval_loss": 6.306853771209717,
      "eval_runtime": 0.7037,
      "eval_samples_per_second": 680.645,
      "eval_steps_per_second": 7.105,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 1.9479657411575317,
      "learning_rate": 5.8627648839556e-05,
      "loss": 6.0496,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 1.1945366859436035,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 6.0518,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 1.310284972190857,
      "learning_rate": 5.660948536831484e-05,
      "loss": 6.0369,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 1.7654529809951782,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 6.033,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 1.9125797748565674,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 6.0375,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 0.7785618305206299,
      "learning_rate": 5.358224016145308e-05,
      "loss": 6.0314,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 2.054070234298706,
      "learning_rate": 5.257315842583249e-05,
      "loss": 6.0225,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 2.3837015628814697,
      "learning_rate": 5.15640766902119e-05,
      "loss": 6.0177,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 3.238194227218628,
      "learning_rate": 5.055499495459133e-05,
      "loss": 6.0151,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 1.9961788654327393,
      "learning_rate": 4.954591321897074e-05,
      "loss": 6.0103,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 1.2372167110443115,
      "learning_rate": 4.853683148335015e-05,
      "loss": 6.0116,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 0.8311449885368347,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 6.0021,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 1.5514942407608032,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 5.9994,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 0.7380921244621277,
      "learning_rate": 4.55095862764884e-05,
      "loss": 5.9957,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 2.2512073516845703,
      "learning_rate": 4.450050454086781e-05,
      "loss": 5.9953,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 1.4760878086090088,
      "learning_rate": 4.349142280524723e-05,
      "loss": 5.9932,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 2.796647787094116,
      "learning_rate": 4.248234106962664e-05,
      "loss": 5.9851,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 0.7970018982887268,
      "learning_rate": 4.147325933400605e-05,
      "loss": 5.98,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 2.277040719985962,
      "learning_rate": 4.046417759838547e-05,
      "loss": 5.9847,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 0.7612630128860474,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 5.9706,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.1523494448805652,
      "eval_loss": 6.286978721618652,
      "eval_runtime": 0.6883,
      "eval_samples_per_second": 695.895,
      "eval_steps_per_second": 7.264,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 0.7531949281692505,
      "learning_rate": 3.84460141271443e-05,
      "loss": 5.975,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 0.7882994413375854,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 5.9666,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 0.7500940561294556,
      "learning_rate": 3.642785065590313e-05,
      "loss": 5.9605,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 0.7568057179450989,
      "learning_rate": 3.541876892028255e-05,
      "loss": 5.9614,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 0.7655161619186401,
      "learning_rate": 3.440968718466196e-05,
      "loss": 5.9624,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 1.392951488494873,
      "learning_rate": 3.340060544904137e-05,
      "loss": 5.9584,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 2.2189531326293945,
      "learning_rate": 3.239152371342079e-05,
      "loss": 5.9493,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 2.1000075340270996,
      "learning_rate": 3.13824419778002e-05,
      "loss": 5.9491,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 0.7946440577507019,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 5.9491,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 1.4779889583587646,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 5.9523,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 0.8644980192184448,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 5.9477,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 4.648371696472168,
      "learning_rate": 2.734611503531786e-05,
      "loss": 5.935,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 2.3201775550842285,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 5.9369,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 0.7412110567092896,
      "learning_rate": 2.532795156407669e-05,
      "loss": 5.9398,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 1.6979233026504517,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 5.9331,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 0.7549757361412048,
      "learning_rate": 2.330978809283552e-05,
      "loss": 5.9318,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 1.6753371953964233,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 5.9288,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 1.4623737335205078,
      "learning_rate": 2.129162462159435e-05,
      "loss": 5.9252,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 2.390078544616699,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 5.9249,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 1.5144931077957153,
      "learning_rate": 1.927346115035318e-05,
      "loss": 5.9245,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.15221353071466623,
      "eval_loss": 6.274591445922852,
      "eval_runtime": 0.703,
      "eval_samples_per_second": 681.355,
      "eval_steps_per_second": 7.112,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 2.33760929107666,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 5.9166,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 2.081544876098633,
      "learning_rate": 1.725529767911201e-05,
      "loss": 5.9202,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 1.2940326929092407,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 5.9181,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 0.8403103947639465,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 5.9165,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 0.8664686679840088,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 5.9147,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 0.7717369794845581,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 5.9069,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 1.4351696968078613,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 5.9092,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 0.8040247559547424,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 5.9101,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 0.8442730903625488,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 5.9016,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 3.029385805130005,
      "learning_rate": 9.182643794147326e-06,
      "loss": 5.8927,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 0.8424391150474548,
      "learning_rate": 8.173562058526741e-06,
      "loss": 5.8863,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 1.016583800315857,
      "learning_rate": 7.164480322906156e-06,
      "loss": 5.883,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 0.8057863116264343,
      "learning_rate": 6.155398587285571e-06,
      "loss": 5.8751,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 0.9153059124946594,
      "learning_rate": 5.146316851664986e-06,
      "loss": 5.8748,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 2.148061990737915,
      "learning_rate": 4.1372351160444e-06,
      "loss": 5.8698,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 0.9412943124771118,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 5.8698,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 0.9936394691467285,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 5.8654,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 0.8264398574829102,
      "learning_rate": 1.109989909182644e-06,
      "loss": 5.864,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 1.1113249063491821,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 5.862,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 3.049796905526016e+17,
      "train_loss": 6.258596241115823,
      "train_runtime": 11648.8655,
      "train_samples_per_second": 490.689,
      "train_steps_per_second": 5.112
    }
  ],
  "logging_steps": 500,
  "max_steps": 59550,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 10000,
  "total_flos": 3.049796905526016e+17,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}