{
  "best_global_step": 20000,
  "best_metric": 6.8691325187683105,
  "best_model_checkpoint": "bert_base_km_50_v2/checkpoint-20000",
  "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": 2.962919235229492,
      "learning_rate": 5e-06,
      "loss": 13.1296,
      "step": 500
    },
    {
      "epoch": 0.41981528127623846,
      "grad_norm": 5.144967555999756,
      "learning_rate": 1e-05,
      "loss": 11.2131,
      "step": 1000
    },
    {
      "epoch": 0.6297229219143576,
      "grad_norm": 15.801915168762207,
      "learning_rate": 1.5e-05,
      "loss": 9.6449,
      "step": 1500
    },
    {
      "epoch": 0.8396305625524769,
      "grad_norm": 13.248839378356934,
      "learning_rate": 2e-05,
      "loss": 8.954,
      "step": 2000
    },
    {
      "epoch": 1.0495382031905962,
      "grad_norm": 8.360651016235352,
      "learning_rate": 2.5e-05,
      "loss": 8.5006,
      "step": 2500
    },
    {
      "epoch": 1.2594458438287153,
      "grad_norm": 7.773918151855469,
      "learning_rate": 3e-05,
      "loss": 8.1113,
      "step": 3000
    },
    {
      "epoch": 1.4693534844668346,
      "grad_norm": 9.11696720123291,
      "learning_rate": 3.5e-05,
      "loss": 7.8645,
      "step": 3500
    },
    {
      "epoch": 1.6792611251049538,
      "grad_norm": 8.067415237426758,
      "learning_rate": 4e-05,
      "loss": 7.715,
      "step": 4000
    },
    {
      "epoch": 1.8891687657430731,
      "grad_norm": 4.843406677246094,
      "learning_rate": 4.5e-05,
      "loss": 7.5847,
      "step": 4500
    },
    {
      "epoch": 2.0990764063811924,
      "grad_norm": 10.34189510345459,
      "learning_rate": 5e-05,
      "loss": 7.4319,
      "step": 5000
    },
    {
      "epoch": 2.3089840470193117,
      "grad_norm": 5.566771984100342,
      "learning_rate": 5.500000000000001e-05,
      "loss": 7.3169,
      "step": 5500
    },
    {
      "epoch": 2.5188916876574305,
      "grad_norm": 4.459074020385742,
      "learning_rate": 6e-05,
      "loss": 7.2867,
      "step": 6000
    },
    {
      "epoch": 2.72879932829555,
      "grad_norm": 6.495669364929199,
      "learning_rate": 6.500000000000001e-05,
      "loss": 7.2436,
      "step": 6500
    },
    {
      "epoch": 2.938706968933669,
      "grad_norm": 4.424249649047852,
      "learning_rate": 7e-05,
      "loss": 7.205,
      "step": 7000
    },
    {
      "epoch": 3.1486146095717884,
      "grad_norm": 5.470309257507324,
      "learning_rate": 7.500000000000001e-05,
      "loss": 7.0816,
      "step": 7500
    },
    {
      "epoch": 3.3585222502099077,
      "grad_norm": 5.092069149017334,
      "learning_rate": 8e-05,
      "loss": 7.0226,
      "step": 8000
    },
    {
      "epoch": 3.568429890848027,
      "grad_norm": 4.966616630554199,
      "learning_rate": 8.5e-05,
      "loss": 7.0452,
      "step": 8500
    },
    {
      "epoch": 3.7783375314861463,
      "grad_norm": 4.715717315673828,
      "learning_rate": 9e-05,
      "loss": 7.0311,
      "step": 9000
    },
    {
      "epoch": 3.988245172124265,
      "grad_norm": 2.73116397857666,
      "learning_rate": 9.5e-05,
      "loss": 7.0215,
      "step": 9500
    },
    {
      "epoch": 4.198152812762385,
      "grad_norm": 3.9705045223236084,
      "learning_rate": 0.0001,
      "loss": 6.8758,
      "step": 10000
    },
    {
      "epoch": 4.198152812762385,
      "eval_accuracy": 0.14967027463201535,
      "eval_loss": 6.927347183227539,
      "eval_runtime": 1.0905,
      "eval_samples_per_second": 439.234,
      "eval_steps_per_second": 4.585,
      "step": 10000
    },
    {
      "epoch": 4.408060453400504,
      "grad_norm": 3.8936572074890137,
      "learning_rate": 9.899091826437942e-05,
      "loss": 6.8931,
      "step": 10500
    },
    {
      "epoch": 4.617968094038623,
      "grad_norm": 5.23145055770874,
      "learning_rate": 9.798183652875883e-05,
      "loss": 6.8933,
      "step": 11000
    },
    {
      "epoch": 4.827875734676742,
      "grad_norm": 3.151768684387207,
      "learning_rate": 9.697275479313825e-05,
      "loss": 6.8693,
      "step": 11500
    },
    {
      "epoch": 5.037783375314861,
      "grad_norm": 3.051431655883789,
      "learning_rate": 9.596367305751767e-05,
      "loss": 6.8359,
      "step": 12000
    },
    {
      "epoch": 5.247691015952981,
      "grad_norm": 5.356419086456299,
      "learning_rate": 9.495459132189707e-05,
      "loss": 6.7252,
      "step": 12500
    },
    {
      "epoch": 5.4575986565911,
      "grad_norm": 3.3402647972106934,
      "learning_rate": 9.394550958627649e-05,
      "loss": 6.7238,
      "step": 13000
    },
    {
      "epoch": 5.667506297229219,
      "grad_norm": 4.156595230102539,
      "learning_rate": 9.293642785065591e-05,
      "loss": 6.7356,
      "step": 13500
    },
    {
      "epoch": 5.877413937867338,
      "grad_norm": 3.473693609237671,
      "learning_rate": 9.192734611503532e-05,
      "loss": 6.7393,
      "step": 14000
    },
    {
      "epoch": 6.087321578505458,
      "grad_norm": 2.4502835273742676,
      "learning_rate": 9.091826437941474e-05,
      "loss": 6.6668,
      "step": 14500
    },
    {
      "epoch": 6.297229219143577,
      "grad_norm": 3.11409068107605,
      "learning_rate": 8.990918264379415e-05,
      "loss": 6.5997,
      "step": 15000
    },
    {
      "epoch": 6.507136859781696,
      "grad_norm": 2.629638910293579,
      "learning_rate": 8.890010090817356e-05,
      "loss": 6.6165,
      "step": 15500
    },
    {
      "epoch": 6.717044500419815,
      "grad_norm": 4.085334300994873,
      "learning_rate": 8.789101917255298e-05,
      "loss": 6.6231,
      "step": 16000
    },
    {
      "epoch": 6.926952141057934,
      "grad_norm": 2.955427408218384,
      "learning_rate": 8.68819374369324e-05,
      "loss": 6.6264,
      "step": 16500
    },
    {
      "epoch": 7.136859781696054,
      "grad_norm": 3.5441067218780518,
      "learning_rate": 8.58728557013118e-05,
      "loss": 6.5343,
      "step": 17000
    },
    {
      "epoch": 7.346767422334173,
      "grad_norm": 4.077515602111816,
      "learning_rate": 8.486377396569122e-05,
      "loss": 6.5035,
      "step": 17500
    },
    {
      "epoch": 7.5566750629722925,
      "grad_norm": 3.0771689414978027,
      "learning_rate": 8.385469223007064e-05,
      "loss": 6.5172,
      "step": 18000
    },
    {
      "epoch": 7.766582703610411,
      "grad_norm": 2.8012757301330566,
      "learning_rate": 8.284561049445006e-05,
      "loss": 6.5141,
      "step": 18500
    },
    {
      "epoch": 7.97649034424853,
      "grad_norm": 3.968254566192627,
      "learning_rate": 8.183652875882947e-05,
      "loss": 6.528,
      "step": 19000
    },
    {
      "epoch": 8.18639798488665,
      "grad_norm": 3.511242151260376,
      "learning_rate": 8.082744702320889e-05,
      "loss": 6.409,
      "step": 19500
    },
    {
      "epoch": 8.39630562552477,
      "grad_norm": 3.865196466445923,
      "learning_rate": 7.98183652875883e-05,
      "loss": 6.4229,
      "step": 20000
    },
    {
      "epoch": 8.39630562552477,
      "eval_accuracy": 0.15258565055136228,
      "eval_loss": 6.8691325187683105,
      "eval_runtime": 1.0321,
      "eval_samples_per_second": 464.086,
      "eval_steps_per_second": 4.844,
      "step": 20000
    },
    {
      "epoch": 8.606213266162888,
      "grad_norm": 2.6308882236480713,
      "learning_rate": 7.880928355196771e-05,
      "loss": 6.4247,
      "step": 20500
    },
    {
      "epoch": 8.816120906801007,
      "grad_norm": 3.646284341812134,
      "learning_rate": 7.780020181634713e-05,
      "loss": 6.4395,
      "step": 21000
    },
    {
      "epoch": 9.026028547439127,
      "grad_norm": 3.824458599090576,
      "learning_rate": 7.679112008072655e-05,
      "loss": 6.4206,
      "step": 21500
    },
    {
      "epoch": 9.235936188077247,
      "grad_norm": 2.4145076274871826,
      "learning_rate": 7.578203834510595e-05,
      "loss": 6.328,
      "step": 22000
    },
    {
      "epoch": 9.445843828715365,
      "grad_norm": 3.590956449508667,
      "learning_rate": 7.477295660948537e-05,
      "loss": 6.3363,
      "step": 22500
    },
    {
      "epoch": 9.655751469353484,
      "grad_norm": 3.603163957595825,
      "learning_rate": 7.376387487386479e-05,
      "loss": 6.3484,
      "step": 23000
    },
    {
      "epoch": 9.865659109991604,
      "grad_norm": 4.103058338165283,
      "learning_rate": 7.27547931382442e-05,
      "loss": 6.3553,
      "step": 23500
    },
    {
      "epoch": 10.075566750629722,
      "grad_norm": 3.4144155979156494,
      "learning_rate": 7.174571140262362e-05,
      "loss": 6.3237,
      "step": 24000
    },
    {
      "epoch": 10.285474391267842,
      "grad_norm": 3.46001935005188,
      "learning_rate": 7.073662966700304e-05,
      "loss": 6.2735,
      "step": 24500
    },
    {
      "epoch": 10.495382031905962,
      "grad_norm": 4.731631278991699,
      "learning_rate": 6.972754793138244e-05,
      "loss": 6.2794,
      "step": 25000
    },
    {
      "epoch": 10.705289672544081,
      "grad_norm": 4.480548858642578,
      "learning_rate": 6.871846619576186e-05,
      "loss": 6.2838,
      "step": 25500
    },
    {
      "epoch": 10.9151973131822,
      "grad_norm": 2.398155450820923,
      "learning_rate": 6.770938446014128e-05,
      "loss": 6.2959,
      "step": 26000
    },
    {
      "epoch": 11.125104953820319,
      "grad_norm": 4.368534564971924,
      "learning_rate": 6.670030272452068e-05,
      "loss": 6.2515,
      "step": 26500
    },
    {
      "epoch": 11.335012594458439,
      "grad_norm": 4.595712661743164,
      "learning_rate": 6.56912209889001e-05,
      "loss": 6.2241,
      "step": 27000
    },
    {
      "epoch": 11.544920235096557,
      "grad_norm": 4.37913179397583,
      "learning_rate": 6.468213925327952e-05,
      "loss": 6.2297,
      "step": 27500
    },
    {
      "epoch": 11.754827875734676,
      "grad_norm": 3.8384788036346436,
      "learning_rate": 6.367305751765893e-05,
      "loss": 6.2311,
      "step": 28000
    },
    {
      "epoch": 11.964735516372796,
      "grad_norm": 2.7434074878692627,
      "learning_rate": 6.266397578203835e-05,
      "loss": 6.2375,
      "step": 28500
    },
    {
      "epoch": 12.174643157010916,
      "grad_norm": 3.8050191402435303,
      "learning_rate": 6.165489404641777e-05,
      "loss": 6.1907,
      "step": 29000
    },
    {
      "epoch": 12.384550797649034,
      "grad_norm": 4.27536153793335,
      "learning_rate": 6.064581231079718e-05,
      "loss": 6.1765,
      "step": 29500
    },
    {
      "epoch": 12.594458438287154,
      "grad_norm": 2.882713794708252,
      "learning_rate": 5.963673057517659e-05,
      "loss": 6.1843,
      "step": 30000
    },
    {
      "epoch": 12.594458438287154,
      "eval_accuracy": 0.15521728180876962,
      "eval_loss": 6.946802616119385,
      "eval_runtime": 1.0123,
      "eval_samples_per_second": 473.2,
      "eval_steps_per_second": 4.939,
      "step": 30000
    },
    {
      "epoch": 12.804366078925273,
      "grad_norm": 3.9159255027770996,
      "learning_rate": 5.8627648839556e-05,
      "loss": 6.1809,
      "step": 30500
    },
    {
      "epoch": 13.014273719563391,
      "grad_norm": 4.446341037750244,
      "learning_rate": 5.7618567103935416e-05,
      "loss": 6.1833,
      "step": 31000
    },
    {
      "epoch": 13.224181360201511,
      "grad_norm": 4.760090351104736,
      "learning_rate": 5.660948536831484e-05,
      "loss": 6.1413,
      "step": 31500
    },
    {
      "epoch": 13.43408900083963,
      "grad_norm": 3.024639129638672,
      "learning_rate": 5.5600403632694254e-05,
      "loss": 6.1389,
      "step": 32000
    },
    {
      "epoch": 13.64399664147775,
      "grad_norm": 2.604367971420288,
      "learning_rate": 5.4591321897073666e-05,
      "loss": 6.1454,
      "step": 32500
    },
    {
      "epoch": 13.853904282115868,
      "grad_norm": 3.731123208999634,
      "learning_rate": 5.358224016145308e-05,
      "loss": 6.1442,
      "step": 33000
    },
    {
      "epoch": 14.063811922753988,
      "grad_norm": 3.5284948348999023,
      "learning_rate": 5.257315842583249e-05,
      "loss": 6.1332,
      "step": 33500
    },
    {
      "epoch": 14.273719563392108,
      "grad_norm": 3.301119089126587,
      "learning_rate": 5.15640766902119e-05,
      "loss": 6.1089,
      "step": 34000
    },
    {
      "epoch": 14.483627204030226,
      "grad_norm": 3.2660069465637207,
      "learning_rate": 5.055499495459133e-05,
      "loss": 6.1084,
      "step": 34500
    },
    {
      "epoch": 14.693534844668346,
      "grad_norm": 3.465550661087036,
      "learning_rate": 4.954591321897074e-05,
      "loss": 6.1077,
      "step": 35000
    },
    {
      "epoch": 14.903442485306465,
      "grad_norm": 1.9607808589935303,
      "learning_rate": 4.853683148335015e-05,
      "loss": 6.108,
      "step": 35500
    },
    {
      "epoch": 15.113350125944585,
      "grad_norm": 3.2005231380462646,
      "learning_rate": 4.7527749747729566e-05,
      "loss": 6.096,
      "step": 36000
    },
    {
      "epoch": 15.323257766582703,
      "grad_norm": 3.318370819091797,
      "learning_rate": 4.6518668012108985e-05,
      "loss": 6.0825,
      "step": 36500
    },
    {
      "epoch": 15.533165407220823,
      "grad_norm": 3.47308349609375,
      "learning_rate": 4.55095862764884e-05,
      "loss": 6.0779,
      "step": 37000
    },
    {
      "epoch": 15.743073047858942,
      "grad_norm": 3.5470447540283203,
      "learning_rate": 4.450050454086781e-05,
      "loss": 6.0808,
      "step": 37500
    },
    {
      "epoch": 15.95298068849706,
      "grad_norm": 3.67008638381958,
      "learning_rate": 4.349142280524723e-05,
      "loss": 6.074,
      "step": 38000
    },
    {
      "epoch": 16.162888329135182,
      "grad_norm": 2.1680643558502197,
      "learning_rate": 4.248234106962664e-05,
      "loss": 6.0605,
      "step": 38500
    },
    {
      "epoch": 16.3727959697733,
      "grad_norm": 2.086988925933838,
      "learning_rate": 4.147325933400605e-05,
      "loss": 6.0539,
      "step": 39000
    },
    {
      "epoch": 16.582703610411418,
      "grad_norm": 3.1543734073638916,
      "learning_rate": 4.046417759838547e-05,
      "loss": 6.0574,
      "step": 39500
    },
    {
      "epoch": 16.79261125104954,
      "grad_norm": 2.6747941970825195,
      "learning_rate": 3.9455095862764884e-05,
      "loss": 6.0459,
      "step": 40000
    },
    {
      "epoch": 16.79261125104954,
      "eval_accuracy": 0.15201300885948188,
      "eval_loss": 7.121280193328857,
      "eval_runtime": 1.0303,
      "eval_samples_per_second": 464.924,
      "eval_steps_per_second": 4.853,
      "step": 40000
    },
    {
      "epoch": 17.002518891687657,
      "grad_norm": 2.871965169906616,
      "learning_rate": 3.84460141271443e-05,
      "loss": 6.0494,
      "step": 40500
    },
    {
      "epoch": 17.212426532325775,
      "grad_norm": 3.4283244609832764,
      "learning_rate": 3.7436932391523715e-05,
      "loss": 6.0311,
      "step": 41000
    },
    {
      "epoch": 17.422334172963897,
      "grad_norm": 4.8513312339782715,
      "learning_rate": 3.642785065590313e-05,
      "loss": 6.0281,
      "step": 41500
    },
    {
      "epoch": 17.632241813602015,
      "grad_norm": 1.9108192920684814,
      "learning_rate": 3.541876892028255e-05,
      "loss": 6.0286,
      "step": 42000
    },
    {
      "epoch": 17.842149454240133,
      "grad_norm": 2.6659042835235596,
      "learning_rate": 3.440968718466196e-05,
      "loss": 6.0288,
      "step": 42500
    },
    {
      "epoch": 18.052057094878254,
      "grad_norm": 3.0580365657806396,
      "learning_rate": 3.340060544904137e-05,
      "loss": 6.0219,
      "step": 43000
    },
    {
      "epoch": 18.261964735516372,
      "grad_norm": 2.2417285442352295,
      "learning_rate": 3.239152371342079e-05,
      "loss": 6.0067,
      "step": 43500
    },
    {
      "epoch": 18.471872376154494,
      "grad_norm": 4.560697555541992,
      "learning_rate": 3.13824419778002e-05,
      "loss": 6.0078,
      "step": 44000
    },
    {
      "epoch": 18.68178001679261,
      "grad_norm": 4.196490287780762,
      "learning_rate": 3.0373360242179615e-05,
      "loss": 6.0028,
      "step": 44500
    },
    {
      "epoch": 18.89168765743073,
      "grad_norm": 2.619422197341919,
      "learning_rate": 2.9364278506559034e-05,
      "loss": 6.0101,
      "step": 45000
    },
    {
      "epoch": 19.10159529806885,
      "grad_norm": 3.3644564151763916,
      "learning_rate": 2.8355196770938446e-05,
      "loss": 6.0012,
      "step": 45500
    },
    {
      "epoch": 19.31150293870697,
      "grad_norm": 1.0323830842971802,
      "learning_rate": 2.734611503531786e-05,
      "loss": 5.9849,
      "step": 46000
    },
    {
      "epoch": 19.521410579345087,
      "grad_norm": 2.431551218032837,
      "learning_rate": 2.6337033299697278e-05,
      "loss": 5.9878,
      "step": 46500
    },
    {
      "epoch": 19.73131821998321,
      "grad_norm": 2.245572328567505,
      "learning_rate": 2.532795156407669e-05,
      "loss": 5.9863,
      "step": 47000
    },
    {
      "epoch": 19.941225860621326,
      "grad_norm": 4.539727687835693,
      "learning_rate": 2.4318869828456106e-05,
      "loss": 5.9824,
      "step": 47500
    },
    {
      "epoch": 20.151133501259444,
      "grad_norm": 2.2890708446502686,
      "learning_rate": 2.330978809283552e-05,
      "loss": 5.9756,
      "step": 48000
    },
    {
      "epoch": 20.361041141897566,
      "grad_norm": 1.448447823524475,
      "learning_rate": 2.2300706357214934e-05,
      "loss": 5.9709,
      "step": 48500
    },
    {
      "epoch": 20.570948782535684,
      "grad_norm": 2.7866880893707275,
      "learning_rate": 2.129162462159435e-05,
      "loss": 5.9634,
      "step": 49000
    },
    {
      "epoch": 20.7808564231738,
      "grad_norm": 2.5933969020843506,
      "learning_rate": 2.0282542885973765e-05,
      "loss": 5.9651,
      "step": 49500
    },
    {
      "epoch": 20.990764063811923,
      "grad_norm": 3.509547472000122,
      "learning_rate": 1.927346115035318e-05,
      "loss": 5.9642,
      "step": 50000
    },
    {
      "epoch": 20.990764063811923,
      "eval_accuracy": 0.15266212465304063,
      "eval_loss": 7.094726085662842,
      "eval_runtime": 1.0342,
      "eval_samples_per_second": 463.157,
      "eval_steps_per_second": 4.835,
      "step": 50000
    },
    {
      "epoch": 21.20067170445004,
      "grad_norm": 1.0806323289871216,
      "learning_rate": 1.8264379414732593e-05,
      "loss": 5.9559,
      "step": 50500
    },
    {
      "epoch": 21.410579345088163,
      "grad_norm": 0.8813648819923401,
      "learning_rate": 1.725529767911201e-05,
      "loss": 5.955,
      "step": 51000
    },
    {
      "epoch": 21.62048698572628,
      "grad_norm": 1.093647837638855,
      "learning_rate": 1.6246215943491424e-05,
      "loss": 5.9522,
      "step": 51500
    },
    {
      "epoch": 21.8303946263644,
      "grad_norm": 1.5033223628997803,
      "learning_rate": 1.5237134207870837e-05,
      "loss": 5.951,
      "step": 52000
    },
    {
      "epoch": 22.04030226700252,
      "grad_norm": 1.4466809034347534,
      "learning_rate": 1.4228052472250252e-05,
      "loss": 5.9494,
      "step": 52500
    },
    {
      "epoch": 22.250209907640638,
      "grad_norm": 0.8350287079811096,
      "learning_rate": 1.3218970736629668e-05,
      "loss": 5.9399,
      "step": 53000
    },
    {
      "epoch": 22.460117548278756,
      "grad_norm": 3.2869601249694824,
      "learning_rate": 1.2209889001009082e-05,
      "loss": 5.9397,
      "step": 53500
    },
    {
      "epoch": 22.670025188916878,
      "grad_norm": 1.1864923238754272,
      "learning_rate": 1.1200807265388496e-05,
      "loss": 5.9406,
      "step": 54000
    },
    {
      "epoch": 22.879932829554996,
      "grad_norm": 3.4119696617126465,
      "learning_rate": 1.0191725529767912e-05,
      "loss": 5.9373,
      "step": 54500
    },
    {
      "epoch": 23.089840470193113,
      "grad_norm": 0.8252732157707214,
      "learning_rate": 9.182643794147326e-06,
      "loss": 5.9326,
      "step": 55000
    },
    {
      "epoch": 23.299748110831235,
      "grad_norm": 2.515281915664673,
      "learning_rate": 8.173562058526741e-06,
      "loss": 5.9291,
      "step": 55500
    },
    {
      "epoch": 23.509655751469353,
      "grad_norm": 3.4073526859283447,
      "learning_rate": 7.164480322906156e-06,
      "loss": 5.9293,
      "step": 56000
    },
    {
      "epoch": 23.719563392107474,
      "grad_norm": 4.885205268859863,
      "learning_rate": 6.155398587285571e-06,
      "loss": 5.9243,
      "step": 56500
    },
    {
      "epoch": 23.929471032745592,
      "grad_norm": 0.7963102459907532,
      "learning_rate": 5.146316851664986e-06,
      "loss": 5.9269,
      "step": 57000
    },
    {
      "epoch": 24.13937867338371,
      "grad_norm": 0.8043511509895325,
      "learning_rate": 4.1372351160444e-06,
      "loss": 5.9217,
      "step": 57500
    },
    {
      "epoch": 24.349286314021832,
      "grad_norm": 0.8935407400131226,
      "learning_rate": 3.1281533804238145e-06,
      "loss": 5.921,
      "step": 58000
    },
    {
      "epoch": 24.55919395465995,
      "grad_norm": 1.7087383270263672,
      "learning_rate": 2.1190716448032293e-06,
      "loss": 5.9166,
      "step": 58500
    },
    {
      "epoch": 24.769101595298068,
      "grad_norm": 1.0903595685958862,
      "learning_rate": 1.109989909182644e-06,
      "loss": 5.9158,
      "step": 59000
    },
    {
      "epoch": 24.97900923593619,
      "grad_norm": 3.5094573497772217,
      "learning_rate": 1.0090817356205853e-07,
      "loss": 5.9159,
      "step": 59500
    },
    {
      "epoch": 25.0,
      "step": 59550,
      "total_flos": 1.5155177772432384e+18,
      "train_loss": 6.511859622830408,
      "train_runtime": 24815.6833,
      "train_samples_per_second": 230.337,
      "train_steps_per_second": 2.4
    }
  ],
  "logging_steps": 500,
  "max_steps": 59550,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 10000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5155177772432384e+18,
  "train_batch_size": 96,
  "trial_name": null,
  "trial_params": null
}