{
  "best_metric": 6.39106559753418,
  "best_model_checkpoint": "tiny_bert_km_10_v2/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": 2.7304940223693848,
      "learning_rate": 5e-06,
      "loss": 12.0209,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 1.8483654260635376,
      "learning_rate": 1e-05,
      "loss": 10.4182,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 3.459078311920166,
      "learning_rate": 1.5e-05,
      "loss": 8.9891,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 6.021159648895264,
      "learning_rate": 2e-05,
      "loss": 8.1475,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 4.758025169372559,
      "learning_rate": 2.5e-05,
      "loss": 7.7121,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 3.8494873046875,
      "learning_rate": 3e-05,
      "loss": 7.4596,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 6.129887580871582,
      "learning_rate": 3.5e-05,
      "loss": 7.3134,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 3.0635933876037598,
      "learning_rate": 4e-05,
      "loss": 7.2024,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 4.089953422546387,
      "learning_rate": 4.5e-05,
      "loss": 7.103,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 3.628406286239624,
      "learning_rate": 5e-05,
      "loss": 6.9838,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 3.614844560623169,
      "learning_rate": 5.500000000000001e-05,
      "loss": 6.8893,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 3.9094295501708984,
      "learning_rate": 6e-05,
      "loss": 6.841,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 2.143172025680542,
      "learning_rate": 6.500000000000001e-05,
      "loss": 6.8,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 3.6978213787078857,
      "learning_rate": 7e-05,
      "loss": 6.7547,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 2.0245327949523926,
      "learning_rate": 7.500000000000001e-05,
      "loss": 6.6714,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 3.006244421005249,
      "learning_rate": 8e-05,
      "loss": 6.638,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 2.350432872772217,
      "learning_rate": 8.5e-05,
      "loss": 6.6273,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 2.483891487121582,
      "learning_rate": 9e-05,
      "loss": 6.6098,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 2.2412710189819336,
      "learning_rate": 9.5e-05,
      "loss": 6.5955,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 2.283215045928955,
      "learning_rate": 0.0001,
      "loss": 6.5128,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.14780600461893764,
      "eval_loss": 6.467541217803955,
      "eval_runtime": 0.7132,
      "eval_samples_per_second": 671.665,
      "eval_steps_per_second": 7.011,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 3.499032497406006,
      "learning_rate": 9.899091826437942e-05,
      "loss": 6.5113,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 2.1217970848083496,
      "learning_rate": 9.798183652875883e-05,
      "loss": 6.5057,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 1.8864343166351318,
      "learning_rate": 9.697275479313825e-05,
      "loss": 6.5017,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 1.210981011390686,
      "learning_rate": 9.596367305751767e-05,
      "loss": 6.4785,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 2.6956076622009277,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.4173,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 2.8102705478668213,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.4129,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 1.9733837842941284,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.4126,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 2.384289026260376,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.405,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 1.6935827732086182,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.3775,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 2.105038642883301,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.3423,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 3.012019395828247,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.3457,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 1.689050555229187,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.338,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 1.2937899827957153,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.3375,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 1.861311912536621,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.2957,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 1.989940881729126,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.2763,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 2.316828727722168,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.2825,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 2.945849657058716,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.2839,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 3.09977650642395,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.2841,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 1.4335596561431885,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.2407,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 1.524065375328064,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.2388,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.15208058587502454,
      "eval_loss": 6.415696620941162,
      "eval_runtime": 0.6961,
      "eval_samples_per_second": 688.158,
      "eval_steps_per_second": 7.183,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 1.8658263683319092,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.2288,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 1.3634886741638184,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.2311,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 2.1126773357391357,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.2262,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 1.409554123878479,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.1935,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 1.6136277914047241,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.1951,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 2.8996996879577637,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.1954,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 0.967384397983551,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.1928,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 1.1264610290527344,
      "learning_rate": 7.174571140262362e-05,
      "loss": 6.1768,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 1.8727537393569946,
      "learning_rate": 7.073662966700304e-05,
      "loss": 6.1626,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 1.518284797668457,
      "learning_rate": 6.972754793138244e-05,
      "loss": 6.1583,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 1.6655353307724,
      "learning_rate": 6.871846619576186e-05,
      "loss": 6.1576,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 1.6358898878097534,
      "learning_rate": 6.770938446014128e-05,
      "loss": 6.1615,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 2.3945817947387695,
      "learning_rate": 6.670030272452068e-05,
      "loss": 6.141,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 2.2548344135284424,
      "learning_rate": 6.56912209889001e-05,
      "loss": 6.1337,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 0.9111269116401672,
      "learning_rate": 6.468213925327952e-05,
      "loss": 6.1302,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 2.538729429244995,
      "learning_rate": 6.367305751765893e-05,
      "loss": 6.1264,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 1.252650260925293,
      "learning_rate": 6.266397578203835e-05,
      "loss": 6.1262,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 3.069003105163574,
      "learning_rate": 6.165489404641777e-05,
      "loss": 6.1105,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 0.7584158182144165,
      "learning_rate": 6.064581231079718e-05,
      "loss": 6.1021,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 1.7243436574935913,
      "learning_rate": 5.963673057517659e-05,
      "loss": 6.1049,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15426588689593418,
      "eval_loss": 6.4011969566345215,
      "eval_runtime": 0.7173,
      "eval_samples_per_second": 667.803,
      "eval_steps_per_second": 6.971,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 1.5291446447372437,
      "learning_rate": 5.8627648839556e-05,
      "loss": 6.1001,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 1.6922167539596558,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 6.1052,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 0.828056275844574,
      "learning_rate": 5.660948536831484e-05,
      "loss": 6.083,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 2.126030683517456,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 6.082,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 3.7158570289611816,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 6.0837,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 2.393825054168701,
      "learning_rate": 5.358224016145308e-05,
      "loss": 6.0813,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 1.366176962852478,
      "learning_rate": 5.257315842583249e-05,
      "loss": 6.0716,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 2.310323715209961,
      "learning_rate": 5.15640766902119e-05,
      "loss": 6.0614,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 1.5324995517730713,
      "learning_rate": 5.055499495459133e-05,
      "loss": 6.061,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 1.696831226348877,
      "learning_rate": 4.954591321897074e-05,
      "loss": 6.0589,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 2.3485970497131348,
      "learning_rate": 4.853683148335015e-05,
      "loss": 6.0569,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 0.9480715990066528,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 6.0498,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 2.2884645462036133,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 6.0452,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 1.8395435810089111,
      "learning_rate": 4.55095862764884e-05,
      "loss": 6.0418,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 1.9194903373718262,
      "learning_rate": 4.450050454086781e-05,
      "loss": 6.0424,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 1.959952473640442,
      "learning_rate": 4.349142280524723e-05,
      "loss": 6.0396,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 2.3246662616729736,
      "learning_rate": 4.248234106962664e-05,
      "loss": 6.0328,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 1.5601322650909424,
      "learning_rate": 4.147325933400605e-05,
      "loss": 6.0265,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 2.609283924102783,
      "learning_rate": 4.046417759838547e-05,
      "loss": 6.0287,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 2.2822744846343994,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 6.018,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.15232140854547493,
      "eval_loss": 6.487921237945557,
      "eval_runtime": 0.7129,
      "eval_samples_per_second": 671.86,
      "eval_steps_per_second": 7.013,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 0.7430895566940308,
      "learning_rate": 3.84460141271443e-05,
      "loss": 6.0234,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 2.7523844242095947,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 6.0127,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 2.402616500854492,
      "learning_rate": 3.642785065590313e-05,
      "loss": 6.0084,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 2.4253292083740234,
      "learning_rate": 3.541876892028255e-05,
      "loss": 6.0083,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 1.3113913536071777,
      "learning_rate": 3.440968718466196e-05,
      "loss": 6.0072,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 1.3224964141845703,
      "learning_rate": 3.340060544904137e-05,
      "loss": 6.0032,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 1.4089961051940918,
      "learning_rate": 3.239152371342079e-05,
      "loss": 5.9962,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 1.5170563459396362,
      "learning_rate": 3.13824419778002e-05,
      "loss": 5.9962,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 0.9209235906600952,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 5.997,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 1.42068350315094,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 5.9968,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 1.3573191165924072,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 5.9923,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 0.8555002212524414,
      "learning_rate": 2.734611503531786e-05,
      "loss": 5.9812,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 1.19484281539917,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 5.9837,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 0.8434991240501404,
      "learning_rate": 2.532795156407669e-05,
      "loss": 5.9837,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 2.876748561859131,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 5.9784,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 3.0139734745025635,
      "learning_rate": 2.330978809283552e-05,
      "loss": 5.9774,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 1.8150054216384888,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 5.9724,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 0.7705669403076172,
      "learning_rate": 2.129162462159435e-05,
      "loss": 5.9684,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 0.7841638922691345,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 5.9622,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 0.7931294441223145,
      "learning_rate": 1.927346115035318e-05,
      "loss": 5.9573,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.15294249586452463,
      "eval_loss": 6.39106559753418,
      "eval_runtime": 0.7037,
      "eval_samples_per_second": 680.715,
      "eval_steps_per_second": 7.106,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 2.107837677001953,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 5.9413,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 0.8583787083625793,
      "learning_rate": 1.725529767911201e-05,
      "loss": 5.9342,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 2.443636178970337,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 5.9255,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 2.174328327178955,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 5.913,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 0.9343647360801697,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 5.9066,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 0.8494040369987488,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 5.8894,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 2.2647204399108887,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 5.8848,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 1.6729694604873657,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 5.8819,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 0.9109405279159546,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 5.8721,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 1.0412042140960693,
      "learning_rate": 9.182643794147326e-06,
      "loss": 5.8647,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 1.7123284339904785,
      "learning_rate": 8.173562058526741e-06,
      "loss": 5.8577,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 1.6566359996795654,
      "learning_rate": 7.164480322906156e-06,
      "loss": 5.8539,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 3.271916627883911,
      "learning_rate": 6.155398587285571e-06,
      "loss": 5.846,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 0.9478620290756226,
      "learning_rate": 5.146316851664986e-06,
      "loss": 5.8452,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 2.892528533935547,
      "learning_rate": 4.1372351160444e-06,
      "loss": 5.8403,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 2.1352267265319824,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 5.8402,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 3.0447394847869873,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 5.8347,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 1.9377501010894775,
      "learning_rate": 1.109989909182644e-06,
      "loss": 5.8315,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 1.139212965965271,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 5.8333,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 3.050247306064896e+17,
      "train_loss": 6.330714561276632,
      "train_runtime": 11800.5284,
      "train_samples_per_second": 484.383,
      "train_steps_per_second": 5.046
    }
  ],
  "logging_steps": 500,
  "max_steps": 59550,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 10000,
  "total_flos": 3.050247306064896e+17,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}