{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9977426636568847,
  "eval_steps": 50,
  "global_step": 996,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.015048908954100828,
      "grad_norm": 18.47044613613669,
      "learning_rate": 5e-07,
      "loss": 1.7355,
      "step": 5
    },
    {
      "epoch": 0.030097817908201655,
      "grad_norm": 12.86199157181735,
      "learning_rate": 1e-06,
      "loss": 1.5959,
      "step": 10
    },
    {
      "epoch": 0.045146726862302484,
      "grad_norm": 7.557418173894333,
      "learning_rate": 9.999365521737421e-07,
      "loss": 1.2852,
      "step": 15
    },
    {
      "epoch": 0.06019563581640331,
      "grad_norm": 4.370427039727891,
      "learning_rate": 9.99746224797475e-07,
      "loss": 1.1474,
      "step": 20
    },
    {
      "epoch": 0.07524454477050414,
      "grad_norm": 3.7509821509698265,
      "learning_rate": 9.99429066174632e-07,
      "loss": 1.0775,
      "step": 25
    },
    {
      "epoch": 0.09029345372460497,
      "grad_norm": 3.4862474442533076,
      "learning_rate": 9.989851567973138e-07,
      "loss": 1.0181,
      "step": 30
    },
    {
      "epoch": 0.1053423626787058,
      "grad_norm": 3.6303816391691863,
      "learning_rate": 9.984146093258608e-07,
      "loss": 1.0061,
      "step": 35
    },
    {
      "epoch": 0.12039127163280662,
      "grad_norm": 3.379335415407791,
      "learning_rate": 9.9771756856026e-07,
      "loss": 0.9556,
      "step": 40
    },
    {
      "epoch": 0.13544018058690746,
      "grad_norm": 3.757214111120403,
      "learning_rate": 9.968942114033973e-07,
      "loss": 0.9456,
      "step": 45
    },
    {
      "epoch": 0.1504890895410083,
      "grad_norm": 3.403304414146967,
      "learning_rate": 9.959447468161596e-07,
      "loss": 0.9297,
      "step": 50
    },
    {
      "epoch": 0.1504890895410083,
      "eval_loss": 0.939862847328186,
      "eval_runtime": 164.2799,
      "eval_samples_per_second": 57.524,
      "eval_steps_per_second": 0.901,
      "step": 50
    },
    {
      "epoch": 0.1655379984951091,
      "grad_norm": 3.5204364554568266,
      "learning_rate": 9.948694157644042e-07,
      "loss": 0.9204,
      "step": 55
    },
    {
      "epoch": 0.18058690744920994,
      "grad_norm": 3.318644271218084,
      "learning_rate": 9.936684911578017e-07,
      "loss": 0.8933,
      "step": 60
    },
    {
      "epoch": 0.19563581640331076,
      "grad_norm": 3.7154382360141494,
      "learning_rate": 9.923422777805751e-07,
      "loss": 0.9219,
      "step": 65
    },
    {
      "epoch": 0.2106847253574116,
      "grad_norm": 3.6290063947255926,
      "learning_rate": 9.908911122141486e-07,
      "loss": 0.8946,
      "step": 70
    },
    {
      "epoch": 0.22573363431151242,
      "grad_norm": 3.6808742110099466,
      "learning_rate": 9.893153627517248e-07,
      "loss": 0.902,
      "step": 75
    },
    {
      "epoch": 0.24078254326561324,
      "grad_norm": 3.5301295406592255,
      "learning_rate": 9.876154293048163e-07,
      "loss": 0.8924,
      "step": 80
    },
    {
      "epoch": 0.2558314522197141,
      "grad_norm": 3.3462246475893123,
      "learning_rate": 9.857917433017508e-07,
      "loss": 0.8958,
      "step": 85
    },
    {
      "epoch": 0.2708803611738149,
      "grad_norm": 3.487142176575527,
      "learning_rate": 9.838447675781793e-07,
      "loss": 0.8881,
      "step": 90
    },
    {
      "epoch": 0.28592927012791575,
      "grad_norm": 3.3151772418217402,
      "learning_rate": 9.817749962596114e-07,
      "loss": 0.882,
      "step": 95
    },
    {
      "epoch": 0.3009781790820166,
      "grad_norm": 3.4456571870576362,
      "learning_rate": 9.795829546360113e-07,
      "loss": 0.8675,
      "step": 100
    },
    {
      "epoch": 0.3009781790820166,
      "eval_loss": 0.8835354447364807,
      "eval_runtime": 164.3323,
      "eval_samples_per_second": 57.505,
      "eval_steps_per_second": 0.901,
      "step": 100
    },
    {
      "epoch": 0.3160270880361174,
      "grad_norm": 3.68930117031382,
      "learning_rate": 9.77269199028483e-07,
      "loss": 0.8656,
      "step": 105
    },
    {
      "epoch": 0.3310759969902182,
      "grad_norm": 3.7770729679447816,
      "learning_rate": 9.748343166480822e-07,
      "loss": 0.8907,
      "step": 110
    },
    {
      "epoch": 0.34612490594431905,
      "grad_norm": 3.5934806142622104,
      "learning_rate": 9.722789254467854e-07,
      "loss": 0.8678,
      "step": 115
    },
    {
      "epoch": 0.3611738148984199,
      "grad_norm": 3.4907829794302447,
      "learning_rate": 9.696036739606606e-07,
      "loss": 0.8634,
      "step": 120
    },
    {
      "epoch": 0.3762227238525207,
      "grad_norm": 3.7214337808863798,
      "learning_rate": 9.668092411452735e-07,
      "loss": 0.8789,
      "step": 125
    },
    {
      "epoch": 0.3912716328066215,
      "grad_norm": 3.510784238627902,
      "learning_rate": 9.638963362033756e-07,
      "loss": 0.8592,
      "step": 130
    },
    {
      "epoch": 0.40632054176072235,
      "grad_norm": 3.3785797508677837,
      "learning_rate": 9.608656984049132e-07,
      "loss": 0.8438,
      "step": 135
    },
    {
      "epoch": 0.4213694507148232,
      "grad_norm": 3.3458509178809854,
      "learning_rate": 9.577180968994081e-07,
      "loss": 0.8544,
      "step": 140
    },
    {
      "epoch": 0.436418359668924,
      "grad_norm": 3.4920402952738656,
      "learning_rate": 9.544543305207546e-07,
      "loss": 0.8518,
      "step": 145
    },
    {
      "epoch": 0.45146726862302483,
      "grad_norm": 3.2912233930311627,
      "learning_rate": 9.510752275844809e-07,
      "loss": 0.8457,
      "step": 150
    },
    {
      "epoch": 0.45146726862302483,
      "eval_loss": 0.857115626335144,
      "eval_runtime": 164.1562,
      "eval_samples_per_second": 57.567,
      "eval_steps_per_second": 0.902,
      "step": 150
    },
    {
      "epoch": 0.46651617757712566,
      "grad_norm": 3.553467076040635,
      "learning_rate": 9.475816456775312e-07,
      "loss": 0.8595,
      "step": 155
    },
    {
      "epoch": 0.4815650865312265,
      "grad_norm": 3.3686846297763116,
      "learning_rate": 9.439744714406166e-07,
      "loss": 0.8392,
      "step": 160
    },
    {
      "epoch": 0.4966139954853273,
      "grad_norm": 3.45480960814811,
      "learning_rate": 9.402546203431947e-07,
      "loss": 0.8323,
      "step": 165
    },
    {
      "epoch": 0.5116629044394282,
      "grad_norm": 3.459267285158168,
      "learning_rate": 9.364230364511295e-07,
      "loss": 0.8355,
      "step": 170
    },
    {
      "epoch": 0.526711813393529,
      "grad_norm": 3.4121749327302378,
      "learning_rate": 9.324806921870975e-07,
      "loss": 0.8537,
      "step": 175
    },
    {
      "epoch": 0.5417607223476298,
      "grad_norm": 3.5893222373958524,
      "learning_rate": 9.284285880837946e-07,
      "loss": 0.828,
      "step": 180
    },
    {
      "epoch": 0.5568096313017307,
      "grad_norm": 3.7134185725142936,
      "learning_rate": 9.242677525300088e-07,
      "loss": 0.8346,
      "step": 185
    },
    {
      "epoch": 0.5718585402558315,
      "grad_norm": 3.550721543432503,
      "learning_rate": 9.199992415096259e-07,
      "loss": 0.832,
      "step": 190
    },
    {
      "epoch": 0.5869074492099323,
      "grad_norm": 3.7325735346120728,
      "learning_rate": 9.156241383336278e-07,
      "loss": 0.8156,
      "step": 195
    },
    {
      "epoch": 0.6019563581640331,
      "grad_norm": 3.6426263387830535,
      "learning_rate": 9.111435533651595e-07,
      "loss": 0.8411,
      "step": 200
    },
    {
      "epoch": 0.6019563581640331,
      "eval_loss": 0.8422114253044128,
      "eval_runtime": 164.0752,
      "eval_samples_per_second": 57.596,
      "eval_steps_per_second": 0.902,
      "step": 200
    },
    {
      "epoch": 0.617005267118134,
      "grad_norm": 3.978063196207494,
      "learning_rate": 9.065586237377274e-07,
      "loss": 0.8255,
      "step": 205
    },
    {
      "epoch": 0.6320541760722348,
      "grad_norm": 3.536724916036319,
      "learning_rate": 9.018705130666049e-07,
      "loss": 0.8355,
      "step": 210
    },
    {
      "epoch": 0.6471030850263356,
      "grad_norm": 3.7576883800972243,
      "learning_rate": 8.970804111535175e-07,
      "loss": 0.8244,
      "step": 215
    },
    {
      "epoch": 0.6621519939804364,
      "grad_norm": 3.413329114828355,
      "learning_rate": 8.921895336846812e-07,
      "loss": 0.8336,
      "step": 220
    },
    {
      "epoch": 0.6772009029345373,
      "grad_norm": 3.477984800656738,
      "learning_rate": 8.871991219222712e-07,
      "loss": 0.8152,
      "step": 225
    },
    {
      "epoch": 0.6922498118886381,
      "grad_norm": 3.6780798309273233,
      "learning_rate": 8.821104423894014e-07,
      "loss": 0.8235,
      "step": 230
    },
    {
      "epoch": 0.7072987208427389,
      "grad_norm": 3.4820310206949383,
      "learning_rate": 8.769247865486915e-07,
      "loss": 0.8226,
      "step": 235
    },
    {
      "epoch": 0.7223476297968398,
      "grad_norm": 3.3478327176800224,
      "learning_rate": 8.716434704745046e-07,
      "loss": 0.8341,
      "step": 240
    },
    {
      "epoch": 0.7373965387509406,
      "grad_norm": 3.540341298435034,
      "learning_rate": 8.662678345189396e-07,
      "loss": 0.825,
      "step": 245
    },
    {
      "epoch": 0.7524454477050414,
      "grad_norm": 3.658518292411783,
      "learning_rate": 8.607992429716608e-07,
      "loss": 0.8133,
      "step": 250
    },
    {
      "epoch": 0.7524454477050414,
      "eval_loss": 0.828486442565918,
      "eval_runtime": 164.1931,
      "eval_samples_per_second": 57.554,
      "eval_steps_per_second": 0.901,
      "step": 250
    },
    {
      "epoch": 0.7674943566591422,
      "grad_norm": 3.3210348798020797,
      "learning_rate": 8.55239083713654e-07,
      "loss": 0.8267,
      "step": 255
    },
    {
      "epoch": 0.782543265613243,
      "grad_norm": 3.2928105286016867,
      "learning_rate": 8.495887678649932e-07,
      "loss": 0.8336,
      "step": 260
    },
    {
      "epoch": 0.7975921745673439,
      "grad_norm": 3.7993668860558567,
      "learning_rate": 8.438497294267116e-07,
      "loss": 0.8371,
      "step": 265
    },
    {
      "epoch": 0.8126410835214447,
      "grad_norm": 3.7491461546658456,
      "learning_rate": 8.38023424916864e-07,
      "loss": 0.83,
      "step": 270
    },
    {
      "epoch": 0.8276899924755455,
      "grad_norm": 3.335971213466453,
      "learning_rate": 8.321113330008756e-07,
      "loss": 0.8231,
      "step": 275
    },
    {
      "epoch": 0.8427389014296464,
      "grad_norm": 3.50538153645489,
      "learning_rate": 8.261149541162691e-07,
      "loss": 0.8283,
      "step": 280
    },
    {
      "epoch": 0.8577878103837472,
      "grad_norm": 3.405552551438381,
      "learning_rate": 8.20035810091867e-07,
      "loss": 0.8333,
      "step": 285
    },
    {
      "epoch": 0.872836719337848,
      "grad_norm": 3.5518417970408036,
      "learning_rate": 8.13875443761565e-07,
      "loss": 0.8215,
      "step": 290
    },
    {
      "epoch": 0.8878856282919488,
      "grad_norm": 3.346948128373629,
      "learning_rate": 8.076354185727734e-07,
      "loss": 0.8127,
      "step": 295
    },
    {
      "epoch": 0.9029345372460497,
      "grad_norm": 3.472824409308143,
      "learning_rate": 8.013173181896282e-07,
      "loss": 0.8114,
      "step": 300
    },
    {
      "epoch": 0.9029345372460497,
      "eval_loss": 0.8188709616661072,
      "eval_runtime": 163.9398,
      "eval_samples_per_second": 57.643,
      "eval_steps_per_second": 0.903,
      "step": 300
    },
    {
      "epoch": 0.9179834462001505,
      "grad_norm": 3.535029645974378,
      "learning_rate": 7.94922746091071e-07,
      "loss": 0.8245,
      "step": 305
    },
    {
      "epoch": 0.9330323551542513,
      "grad_norm": 3.350512530278651,
      "learning_rate": 7.884533251638999e-07,
      "loss": 0.8031,
      "step": 310
    },
    {
      "epoch": 0.9480812641083521,
      "grad_norm": 3.1272224593692903,
      "learning_rate": 7.819106972908949e-07,
      "loss": 0.792,
      "step": 315
    },
    {
      "epoch": 0.963130173062453,
      "grad_norm": 3.3082625072971066,
      "learning_rate": 7.752965229341219e-07,
      "loss": 0.8073,
      "step": 320
    },
    {
      "epoch": 0.9781790820165538,
      "grad_norm": 3.6472654100125674,
      "learning_rate": 7.686124807135228e-07,
      "loss": 0.8091,
      "step": 325
    },
    {
      "epoch": 0.9932279909706546,
      "grad_norm": 3.4227050519519184,
      "learning_rate": 7.618602669808957e-07,
      "loss": 0.812,
      "step": 330
    },
    {
      "epoch": 1.0082768999247556,
      "grad_norm": 3.63643107405423,
      "learning_rate": 7.550415953893756e-07,
      "loss": 0.7794,
      "step": 335
    },
    {
      "epoch": 1.0233258088788564,
      "grad_norm": 3.5111108810888307,
      "learning_rate": 7.481581964585244e-07,
      "loss": 0.7534,
      "step": 340
    },
    {
      "epoch": 1.0383747178329572,
      "grad_norm": 3.738034258550528,
      "learning_rate": 7.412118171351395e-07,
      "loss": 0.7473,
      "step": 345
    },
    {
      "epoch": 1.053423626787058,
      "grad_norm": 3.511742644224616,
      "learning_rate": 7.342042203498951e-07,
      "loss": 0.7379,
      "step": 350
    },
    {
      "epoch": 1.053423626787058,
      "eval_loss": 0.8137179613113403,
      "eval_runtime": 163.9091,
      "eval_samples_per_second": 57.654,
      "eval_steps_per_second": 0.903,
      "step": 350
    },
    {
      "epoch": 1.0684725357411589,
      "grad_norm": 3.482412652291007,
      "learning_rate": 7.271371845699241e-07,
      "loss": 0.757,
      "step": 355
    },
    {
      "epoch": 1.0835214446952597,
      "grad_norm": 3.57693306476351,
      "learning_rate": 7.200125033474598e-07,
      "loss": 0.7405,
      "step": 360
    },
    {
      "epoch": 1.0985703536493605,
      "grad_norm": 3.551987518468492,
      "learning_rate": 7.128319848646477e-07,
      "loss": 0.7424,
      "step": 365
    },
    {
      "epoch": 1.1136192626034613,
      "grad_norm": 3.644980029199069,
      "learning_rate": 7.055974514746445e-07,
      "loss": 0.7557,
      "step": 370
    },
    {
      "epoch": 1.1286681715575622,
      "grad_norm": 3.661260327693092,
      "learning_rate": 6.983107392391202e-07,
      "loss": 0.7595,
      "step": 375
    },
    {
      "epoch": 1.143717080511663,
      "grad_norm": 3.6520213924338867,
      "learning_rate": 6.909736974622826e-07,
      "loss": 0.7411,
      "step": 380
    },
    {
      "epoch": 1.1587659894657638,
      "grad_norm": 3.341094037190979,
      "learning_rate": 6.835881882215395e-07,
      "loss": 0.7415,
      "step": 385
    },
    {
      "epoch": 1.1738148984198646,
      "grad_norm": 3.4808910667039803,
      "learning_rate": 6.761560858949192e-07,
      "loss": 0.7317,
      "step": 390
    },
    {
      "epoch": 1.1888638073739655,
      "grad_norm": 3.6539416517236076,
      "learning_rate": 6.686792766853705e-07,
      "loss": 0.7537,
      "step": 395
    },
    {
      "epoch": 1.2039127163280663,
      "grad_norm": 3.4648582888257002,
      "learning_rate": 6.611596581420599e-07,
      "loss": 0.7416,
      "step": 400
    },
    {
      "epoch": 1.2039127163280663,
      "eval_loss": 0.8082437515258789,
      "eval_runtime": 163.785,
      "eval_samples_per_second": 57.698,
      "eval_steps_per_second": 0.904,
      "step": 400
    },
    {
      "epoch": 1.2189616252821671,
      "grad_norm": 3.3756134153833433,
      "learning_rate": 6.53599138678791e-07,
      "loss": 0.7295,
      "step": 405
    },
    {
      "epoch": 1.234010534236268,
      "grad_norm": 3.5366994024401093,
      "learning_rate": 6.459996370896652e-07,
      "loss": 0.7419,
      "step": 410
    },
    {
      "epoch": 1.2490594431903688,
      "grad_norm": 3.6126358863182118,
      "learning_rate": 6.383630820621081e-07,
      "loss": 0.7348,
      "step": 415
    },
    {
      "epoch": 1.2641083521444696,
      "grad_norm": 3.329040929496611,
      "learning_rate": 6.306914116873862e-07,
      "loss": 0.7456,
      "step": 420
    },
    {
      "epoch": 1.2791572610985704,
      "grad_norm": 3.389626369376778,
      "learning_rate": 6.22986572968736e-07,
      "loss": 0.7399,
      "step": 425
    },
    {
      "epoch": 1.2942061700526712,
      "grad_norm": 3.640732106094957,
      "learning_rate": 6.152505213272307e-07,
      "loss": 0.7478,
      "step": 430
    },
    {
      "epoch": 1.309255079006772,
      "grad_norm": 3.678659612980298,
      "learning_rate": 6.074852201055121e-07,
      "loss": 0.7348,
      "step": 435
    },
    {
      "epoch": 1.324303987960873,
      "grad_norm": 3.456540220241952,
      "learning_rate": 5.996926400695113e-07,
      "loss": 0.7367,
      "step": 440
    },
    {
      "epoch": 1.3393528969149737,
      "grad_norm": 3.598565981484491,
      "learning_rate": 5.918747589082852e-07,
      "loss": 0.7452,
      "step": 445
    },
    {
      "epoch": 1.3544018058690745,
      "grad_norm": 3.6204936371770584,
      "learning_rate": 5.840335607320963e-07,
      "loss": 0.7623,
      "step": 450
    },
    {
      "epoch": 1.3544018058690745,
      "eval_loss": 0.8033472299575806,
      "eval_runtime": 163.9716,
      "eval_samples_per_second": 57.632,
      "eval_steps_per_second": 0.903,
      "step": 450
    },
    {
      "epoch": 1.3694507148231754,
      "grad_norm": 3.5876675484160927,
      "learning_rate": 5.761710355688627e-07,
      "loss": 0.7431,
      "step": 455
    },
    {
      "epoch": 1.3844996237772762,
      "grad_norm": 3.619019215155952,
      "learning_rate": 5.682891788591065e-07,
      "loss": 0.7464,
      "step": 460
    },
    {
      "epoch": 1.399548532731377,
      "grad_norm": 3.5444023840644237,
      "learning_rate": 5.603899909495283e-07,
      "loss": 0.7357,
      "step": 465
    },
    {
      "epoch": 1.4145974416854779,
      "grad_norm": 3.4447315407410266,
      "learning_rate": 5.52475476585336e-07,
      "loss": 0.735,
      "step": 470
    },
    {
      "epoch": 1.4296463506395787,
      "grad_norm": 3.8423796002300983,
      "learning_rate": 5.445476444014591e-07,
      "loss": 0.7318,
      "step": 475
    },
    {
      "epoch": 1.4446952595936795,
      "grad_norm": 3.594078636376061,
      "learning_rate": 5.366085064127734e-07,
      "loss": 0.7484,
      "step": 480
    },
    {
      "epoch": 1.4597441685477803,
      "grad_norm": 3.607424426488508,
      "learning_rate": 5.286600775034699e-07,
      "loss": 0.7435,
      "step": 485
    },
    {
      "epoch": 1.4747930775018812,
      "grad_norm": 3.7328475918869715,
      "learning_rate": 5.207043749156944e-07,
      "loss": 0.7378,
      "step": 490
    },
    {
      "epoch": 1.489841986455982,
      "grad_norm": 3.443608577698693,
      "learning_rate": 5.127434177375893e-07,
      "loss": 0.7418,
      "step": 495
    },
    {
      "epoch": 1.5048908954100828,
      "grad_norm": 3.5186913080533637,
      "learning_rate": 5.047792263908659e-07,
      "loss": 0.7158,
      "step": 500
    },
    {
      "epoch": 1.5048908954100828,
      "eval_loss": 0.7987317442893982,
      "eval_runtime": 163.9794,
      "eval_samples_per_second": 57.629,
      "eval_steps_per_second": 0.903,
      "step": 500
    },
    {
      "epoch": 1.5199398043641836,
      "grad_norm": 3.865883308748703,
      "learning_rate": 4.968138221180401e-07,
      "loss": 0.7431,
      "step": 505
    },
    {
      "epoch": 1.5349887133182845,
      "grad_norm": 3.710215468774283,
      "learning_rate": 4.888492264694565e-07,
      "loss": 0.7007,
      "step": 510
    },
    {
      "epoch": 1.5500376222723853,
      "grad_norm": 3.6132015362670145,
      "learning_rate": 4.808874607902397e-07,
      "loss": 0.7364,
      "step": 515
    },
    {
      "epoch": 1.565086531226486,
      "grad_norm": 3.7199884506401593,
      "learning_rate": 4.7293054570729126e-07,
      "loss": 0.7267,
      "step": 520
    },
    {
      "epoch": 1.580135440180587,
      "grad_norm": 3.9105644710376457,
      "learning_rate": 4.649805006164743e-07,
      "loss": 0.7298,
      "step": 525
    },
    {
      "epoch": 1.5951843491346878,
      "grad_norm": 3.5526438869005292,
      "learning_rate": 4.5703934317010727e-07,
      "loss": 0.7429,
      "step": 530
    },
    {
      "epoch": 1.6102332580887886,
      "grad_norm": 3.502575386275301,
      "learning_rate": 4.491090887649024e-07,
      "loss": 0.7337,
      "step": 535
    },
    {
      "epoch": 1.6252821670428894,
      "grad_norm": 3.794160344449478,
      "learning_rate": 4.4119175003047407e-07,
      "loss": 0.7299,
      "step": 540
    },
    {
      "epoch": 1.6403310759969902,
      "grad_norm": 3.2770795614619237,
      "learning_rate": 4.3328933631855195e-07,
      "loss": 0.7321,
      "step": 545
    },
    {
      "epoch": 1.655379984951091,
      "grad_norm": 3.6119103240995263,
      "learning_rate": 4.2540385319302524e-07,
      "loss": 0.7322,
      "step": 550
    },
    {
      "epoch": 1.655379984951091,
      "eval_loss": 0.7954748272895813,
      "eval_runtime": 164.5155,
      "eval_samples_per_second": 57.441,
      "eval_steps_per_second": 0.9,
      "step": 550
    },
    {
      "epoch": 1.670428893905192,
      "grad_norm": 3.5947026375482944,
      "learning_rate": 4.175373019209468e-07,
      "loss": 0.7457,
      "step": 555
    },
    {
      "epoch": 1.6854778028592927,
      "grad_norm": 3.761848976466262,
      "learning_rate": 4.0969167896463046e-07,
      "loss": 0.7386,
      "step": 560
    },
    {
      "epoch": 1.7005267118133935,
      "grad_norm": 3.439361811338972,
      "learning_rate": 4.018689754749648e-07,
      "loss": 0.724,
      "step": 565
    },
    {
      "epoch": 1.7155756207674944,
      "grad_norm": 3.4555454207422858,
      "learning_rate": 3.9407117678607756e-07,
      "loss": 0.7441,
      "step": 570
    },
    {
      "epoch": 1.7306245297215952,
      "grad_norm": 3.5617425595455177,
      "learning_rate": 3.8630026191147405e-07,
      "loss": 0.7249,
      "step": 575
    },
    {
      "epoch": 1.745673438675696,
      "grad_norm": 3.448724415418216,
      "learning_rate": 3.78558203041782e-07,
      "loss": 0.6972,
      "step": 580
    },
    {
      "epoch": 1.7607223476297968,
      "grad_norm": 3.6943962689825827,
      "learning_rate": 3.7084696504422525e-07,
      "loss": 0.7302,
      "step": 585
    },
    {
      "epoch": 1.7757712565838977,
      "grad_norm": 3.6561881604475945,
      "learning_rate": 3.6316850496395855e-07,
      "loss": 0.7424,
      "step": 590
    },
    {
      "epoch": 1.7908201655379985,
      "grad_norm": 3.625068299425613,
      "learning_rate": 3.555247715273867e-07,
      "loss": 0.7252,
      "step": 595
    },
    {
      "epoch": 1.8058690744920993,
      "grad_norm": 3.519787303173951,
      "learning_rate": 3.4791770464759347e-07,
      "loss": 0.7148,
      "step": 600
    },
    {
      "epoch": 1.8058690744920993,
      "eval_loss": 0.7919732332229614,
      "eval_runtime": 163.9475,
      "eval_samples_per_second": 57.64,
      "eval_steps_per_second": 0.903,
      "step": 600
    },
    {
      "epoch": 1.8209179834462002,
      "grad_norm": 3.8300888994459306,
      "learning_rate": 3.4034923493201007e-07,
      "loss": 0.7377,
      "step": 605
    },
    {
      "epoch": 1.835966892400301,
      "grad_norm": 3.400875766754663,
      "learning_rate": 3.3282128319244237e-07,
      "loss": 0.7385,
      "step": 610
    },
    {
      "epoch": 1.8510158013544018,
      "grad_norm": 3.6182045725169325,
      "learning_rate": 3.2533575995758694e-07,
      "loss": 0.7197,
      "step": 615
    },
    {
      "epoch": 1.8660647103085026,
      "grad_norm": 3.904119144772108,
      "learning_rate": 3.178945649881543e-07,
      "loss": 0.7018,
      "step": 620
    },
    {
      "epoch": 1.8811136192626035,
      "grad_norm": 3.8565228264688822,
      "learning_rate": 3.1049958679472645e-07,
      "loss": 0.7246,
      "step": 625
    },
    {
      "epoch": 1.8961625282167043,
      "grad_norm": 3.422783362651409,
      "learning_rate": 3.031527021584701e-07,
      "loss": 0.7336,
      "step": 630
    },
    {
      "epoch": 1.911211437170805,
      "grad_norm": 3.675919325226495,
      "learning_rate": 2.9585577565482484e-07,
      "loss": 0.7169,
      "step": 635
    },
    {
      "epoch": 1.926260346124906,
      "grad_norm": 3.726914018119805,
      "learning_rate": 2.886106591802908e-07,
      "loss": 0.7259,
      "step": 640
    },
    {
      "epoch": 1.9413092550790068,
      "grad_norm": 3.5805352508835333,
      "learning_rate": 2.814191914824332e-07,
      "loss": 0.7168,
      "step": 645
    },
    {
      "epoch": 1.9563581640331076,
      "grad_norm": 3.60649172674426,
      "learning_rate": 2.7428319769322415e-07,
      "loss": 0.71,
      "step": 650
    },
    {
      "epoch": 1.9563581640331076,
      "eval_loss": 0.7893925309181213,
      "eval_runtime": 164.1237,
      "eval_samples_per_second": 57.579,
      "eval_steps_per_second": 0.902,
      "step": 650
    },
    {
      "epoch": 1.9714070729872084,
      "grad_norm": 3.614812754262687,
      "learning_rate": 2.672044888658399e-07,
      "loss": 0.7041,
      "step": 655
    },
    {
      "epoch": 1.9864559819413092,
      "grad_norm": 3.7073968028896473,
      "learning_rate": 2.6018486151503213e-07,
      "loss": 0.7285,
      "step": 660
    },
    {
      "epoch": 2.00150489089541,
      "grad_norm": 3.647866320725029,
      "learning_rate": 2.532260971611867e-07,
      "loss": 0.7113,
      "step": 665
    },
    {
      "epoch": 2.016553799849511,
      "grad_norm": 3.699708580061935,
      "learning_rate": 2.4632996187819034e-07,
      "loss": 0.6883,
      "step": 670
    },
    {
      "epoch": 2.0316027088036117,
      "grad_norm": 3.936748692781072,
      "learning_rate": 2.394982058452165e-07,
      "loss": 0.6756,
      "step": 675
    },
    {
      "epoch": 2.0466516177577128,
      "grad_norm": 3.716518462078895,
      "learning_rate": 2.3273256290254402e-07,
      "loss": 0.6762,
      "step": 680
    },
    {
      "epoch": 2.0617005267118134,
      "grad_norm": 3.702081977847743,
      "learning_rate": 2.2603475011152517e-07,
      "loss": 0.6895,
      "step": 685
    },
    {
      "epoch": 2.0767494356659144,
      "grad_norm": 3.605178078795864,
      "learning_rate": 2.1940646731880885e-07,
      "loss": 0.6607,
      "step": 690
    },
    {
      "epoch": 2.091798344620015,
      "grad_norm": 3.7196889182496733,
      "learning_rate": 2.1284939672493506e-07,
      "loss": 0.6789,
      "step": 695
    },
    {
      "epoch": 2.106847253574116,
      "grad_norm": 3.7090327815341753,
      "learning_rate": 2.0636520245740708e-07,
      "loss": 0.6651,
      "step": 700
    },
    {
      "epoch": 2.106847253574116,
      "eval_loss": 0.7947185039520264,
      "eval_runtime": 164.1101,
      "eval_samples_per_second": 57.583,
      "eval_steps_per_second": 0.902,
      "step": 700
    },
    {
      "epoch": 2.1218961625282167,
      "grad_norm": 3.736696447237514,
      "learning_rate": 1.9995553014834986e-07,
      "loss": 0.6809,
      "step": 705
    },
    {
      "epoch": 2.1369450714823177,
      "grad_norm": 3.6458918889335865,
      "learning_rate": 1.9362200651686406e-07,
      "loss": 0.6864,
      "step": 710
    },
    {
      "epoch": 2.1519939804364183,
      "grad_norm": 4.002063868688003,
      "learning_rate": 1.873662389561771e-07,
      "loss": 0.6987,
      "step": 715
    },
    {
      "epoch": 2.1670428893905194,
      "grad_norm": 3.8698599662856714,
      "learning_rate": 1.8118981512570254e-07,
      "loss": 0.6801,
      "step": 720
    },
    {
      "epoch": 2.18209179834462,
      "grad_norm": 3.6230492953936677,
      "learning_rate": 1.750943025481046e-07,
      "loss": 0.6826,
      "step": 725
    },
    {
      "epoch": 2.197140707298721,
      "grad_norm": 3.7726470306057953,
      "learning_rate": 1.6908124821147517e-07,
      "loss": 0.6904,
      "step": 730
    },
    {
      "epoch": 2.2121896162528216,
      "grad_norm": 3.787910233434616,
      "learning_rate": 1.631521781767214e-07,
      "loss": 0.6798,
      "step": 735
    },
    {
      "epoch": 2.2272385252069227,
      "grad_norm": 3.9836210259763023,
      "learning_rate": 1.5730859719026535e-07,
      "loss": 0.6776,
      "step": 740
    },
    {
      "epoch": 2.2422874341610233,
      "grad_norm": 3.763373496609544,
      "learning_rate": 1.5155198830215144e-07,
      "loss": 0.6805,
      "step": 745
    },
    {
      "epoch": 2.2573363431151243,
      "grad_norm": 3.8572405199189466,
      "learning_rate": 1.4588381248966185e-07,
      "loss": 0.6684,
      "step": 750
    },
    {
      "epoch": 2.2573363431151243,
      "eval_loss": 0.7936417460441589,
      "eval_runtime": 163.9206,
      "eval_samples_per_second": 57.65,
      "eval_steps_per_second": 0.903,
      "step": 750
    },
    {
      "epoch": 2.272385252069225,
      "grad_norm": 3.8388172371691924,
      "learning_rate": 1.4030550828653354e-07,
      "loss": 0.7012,
      "step": 755
    },
    {
      "epoch": 2.287434161023326,
      "grad_norm": 3.8523292354815823,
      "learning_rate": 1.3481849141786977e-07,
      "loss": 0.6778,
      "step": 760
    },
    {
      "epoch": 2.3024830699774266,
      "grad_norm": 3.937046633996997,
      "learning_rate": 1.294241544408425e-07,
      "loss": 0.6775,
      "step": 765
    },
    {
      "epoch": 2.3175319789315276,
      "grad_norm": 3.953239589293332,
      "learning_rate": 1.241238663912727e-07,
      "loss": 0.6868,
      "step": 770
    },
    {
      "epoch": 2.3325808878856282,
      "grad_norm": 4.089219915268593,
      "learning_rate": 1.1891897243618183e-07,
      "loss": 0.6694,
      "step": 775
    },
    {
      "epoch": 2.3476297968397293,
      "grad_norm": 3.846961987007966,
      "learning_rate": 1.1381079353239915e-07,
      "loss": 0.6779,
      "step": 780
    },
    {
      "epoch": 2.36267870579383,
      "grad_norm": 3.7397027273861885,
      "learning_rate": 1.0880062609131485e-07,
      "loss": 0.6726,
      "step": 785
    },
    {
      "epoch": 2.377727614747931,
      "grad_norm": 3.921668306186131,
      "learning_rate": 1.0388974164986247e-07,
      "loss": 0.6796,
      "step": 790
    },
    {
      "epoch": 2.3927765237020315,
      "grad_norm": 3.8468336671320853,
      "learning_rate": 9.907938654781306e-08,
      "loss": 0.6835,
      "step": 795
    },
    {
      "epoch": 2.4078254326561326,
      "grad_norm": 3.7410679007377854,
      "learning_rate": 9.437078161146589e-08,
      "loss": 0.6826,
      "step": 800
    },
    {
      "epoch": 2.4078254326561326,
      "eval_loss": 0.7927989363670349,
      "eval_runtime": 167.9966,
      "eval_samples_per_second": 56.251,
      "eval_steps_per_second": 0.881,
      "step": 800
    },
    {
      "epoch": 2.422874341610233,
      "grad_norm": 4.079995858636958,
      "learning_rate": 8.976512184381246e-08,
      "loss": 0.6795,
      "step": 805
    },
    {
      "epoch": 2.4379232505643342,
      "grad_norm": 3.735129398089214,
      "learning_rate": 8.526357612125573e-08,
      "loss": 0.6898,
      "step": 810
    },
    {
      "epoch": 2.452972159518435,
      "grad_norm": 3.8390441333226817,
      "learning_rate": 8.086728689695921e-08,
      "loss": 0.7046,
      "step": 815
    },
    {
      "epoch": 2.468021068472536,
      "grad_norm": 3.9049522755901362,
      "learning_rate": 7.657736991090263e-08,
      "loss": 0.661,
      "step": 820
    },
    {
      "epoch": 2.4830699774266365,
      "grad_norm": 3.880401857930606,
      "learning_rate": 7.239491390671631e-08,
      "loss": 0.6742,
      "step": 825
    },
    {
      "epoch": 2.4981188863807375,
      "grad_norm": 3.912207039826943,
      "learning_rate": 6.832098035536759e-08,
      "loss": 0.6672,
      "step": 830
    },
    {
      "epoch": 2.513167795334838,
      "grad_norm": 3.922109823682805,
      "learning_rate": 6.435660318576935e-08,
      "loss": 0.6754,
      "step": 835
    },
    {
      "epoch": 2.528216704288939,
      "grad_norm": 3.8537847324852224,
      "learning_rate": 6.0502788522377e-08,
      "loss": 0.6824,
      "step": 840
    },
    {
      "epoch": 2.54326561324304,
      "grad_norm": 4.033607009118544,
      "learning_rate": 5.676051442984325e-08,
      "loss": 0.6716,
      "step": 845
    },
    {
      "epoch": 2.558314522197141,
      "grad_norm": 3.6539091883922423,
      "learning_rate": 5.313073066479379e-08,
      "loss": 0.6711,
      "step": 850
    },
    {
      "epoch": 2.558314522197141,
      "eval_loss": 0.7919635772705078,
      "eval_runtime": 164.1041,
      "eval_samples_per_second": 57.585,
      "eval_steps_per_second": 0.902,
      "step": 850
    },
    {
      "epoch": 2.5733634311512414,
      "grad_norm": 3.876589664826218,
      "learning_rate": 4.961435843478751e-08,
      "loss": 0.6806,
      "step": 855
    },
    {
      "epoch": 2.5884123401053425,
      "grad_norm": 3.655653854351189,
      "learning_rate": 4.621229016452155e-08,
      "loss": 0.6535,
      "step": 860
    },
    {
      "epoch": 2.603461249059443,
      "grad_norm": 4.079997158323887,
      "learning_rate": 4.2925389269341916e-08,
      "loss": 0.673,
      "step": 865
    },
    {
      "epoch": 2.618510158013544,
      "grad_norm": 3.8325795147951207,
      "learning_rate": 3.975448993611652e-08,
      "loss": 0.6709,
      "step": 870
    },
    {
      "epoch": 2.6335590669676447,
      "grad_norm": 3.778742993294259,
      "learning_rate": 3.67003969115251e-08,
      "loss": 0.6619,
      "step": 875
    },
    {
      "epoch": 2.648607975921746,
      "grad_norm": 4.10366570220581,
      "learning_rate": 3.376388529782215e-08,
      "loss": 0.6848,
      "step": 880
    },
    {
      "epoch": 2.6636568848758464,
      "grad_norm": 3.8699319815585236,
      "learning_rate": 3.094570035612226e-08,
      "loss": 0.6853,
      "step": 885
    },
    {
      "epoch": 2.6787057938299474,
      "grad_norm": 4.043194008918937,
      "learning_rate": 2.8246557317259723e-08,
      "loss": 0.6745,
      "step": 890
    },
    {
      "epoch": 2.693754702784048,
      "grad_norm": 3.99392087783422,
      "learning_rate": 2.5667141200268694e-08,
      "loss": 0.6577,
      "step": 895
    },
    {
      "epoch": 2.708803611738149,
      "grad_norm": 3.6631248500731814,
      "learning_rate": 2.3208106638531842e-08,
      "loss": 0.6757,
      "step": 900
    },
    {
      "epoch": 2.708803611738149,
      "eval_loss": 0.7914727926254272,
      "eval_runtime": 164.3819,
      "eval_samples_per_second": 57.488,
      "eval_steps_per_second": 0.9,
      "step": 900
    },
    {
      "epoch": 2.7238525206922497,
      "grad_norm": 3.9578522639967537,
      "learning_rate": 2.087007771363969e-08,
      "loss": 0.6736,
      "step": 905
    },
    {
      "epoch": 2.7389014296463507,
      "grad_norm": 3.739728228760654,
      "learning_rate": 1.8653647797004236e-08,
      "loss": 0.6729,
      "step": 910
    },
    {
      "epoch": 2.7539503386004514,
      "grad_norm": 3.8722809533250326,
      "learning_rate": 1.655937939926655e-08,
      "loss": 0.6841,
      "step": 915
    },
    {
      "epoch": 2.7689992475545524,
      "grad_norm": 3.8241600199550945,
      "learning_rate": 1.4587804027536454e-08,
      "loss": 0.6917,
      "step": 920
    },
    {
      "epoch": 2.784048156508653,
      "grad_norm": 3.7138354028978506,
      "learning_rate": 1.2739422050500436e-08,
      "loss": 0.6751,
      "step": 925
    },
    {
      "epoch": 2.799097065462754,
      "grad_norm": 3.790111519043104,
      "learning_rate": 1.101470257143261e-08,
      "loss": 0.6738,
      "step": 930
    },
    {
      "epoch": 2.8141459744168547,
      "grad_norm": 3.561873269533578,
      "learning_rate": 9.414083309140453e-09,
      "loss": 0.6952,
      "step": 935
    },
    {
      "epoch": 2.8291948833709557,
      "grad_norm": 3.6893556293107364,
      "learning_rate": 7.93797048687539e-09,
      "loss": 0.6761,
      "step": 940
    },
    {
      "epoch": 2.8442437923250563,
      "grad_norm": 3.8751112944911412,
      "learning_rate": 6.5867387292369295e-09,
      "loss": 0.6932,
      "step": 945
    },
    {
      "epoch": 2.8592927012791574,
      "grad_norm": 4.004750596943983,
      "learning_rate": 5.360730967096272e-09,
      "loss": 0.6791,
      "step": 950
    },
    {
      "epoch": 2.8592927012791574,
      "eval_loss": 0.7912743091583252,
      "eval_runtime": 164.3927,
      "eval_samples_per_second": 57.484,
      "eval_steps_per_second": 0.9,
      "step": 950
    },
    {
      "epoch": 2.874341610233258,
      "grad_norm": 3.8028887659361637,
      "learning_rate": 4.260258350563317e-09,
      "loss": 0.6716,
      "step": 955
    },
    {
      "epoch": 2.889390519187359,
      "grad_norm": 3.8634306165906107,
      "learning_rate": 3.285600170019609e-09,
      "loss": 0.6917,
      "step": 960
    },
    {
      "epoch": 2.9044394281414596,
      "grad_norm": 3.6280717262306363,
      "learning_rate": 2.437003785236702e-09,
      "loss": 0.665,
      "step": 965
    },
    {
      "epoch": 2.9194883370955607,
      "grad_norm": 4.013775538099351,
      "learning_rate": 1.714684562598545e-09,
      "loss": 0.6845,
      "step": 970
    },
    {
      "epoch": 2.9345372460496613,
      "grad_norm": 3.8102208297791,
      "learning_rate": 1.1188258204433144e-09,
      "loss": 0.6673,
      "step": 975
    },
    {
      "epoch": 2.9495861550037623,
      "grad_norm": 3.649482883492926,
      "learning_rate": 6.49578782538851e-10,
      "loss": 0.6772,
      "step": 980
    },
    {
      "epoch": 2.964635063957863,
      "grad_norm": 3.9407869896383483,
      "learning_rate": 3.070625397031401e-10,
      "loss": 0.6743,
      "step": 985
    },
    {
      "epoch": 2.979683972911964,
      "grad_norm": 3.857935656292572,
      "learning_rate": 9.136401958059759e-11,
      "loss": 0.6633,
      "step": 990
    },
    {
      "epoch": 2.9947328818660646,
      "grad_norm": 3.934124028791183,
      "learning_rate": 2.5379645800516215e-12,
      "loss": 0.672,
      "step": 995
    },
    {
      "epoch": 2.9977426636568847,
      "step": 996,
      "total_flos": 5872633472090112.0,
      "train_loss": 0.7687147306988996,
      "train_runtime": 16194.6861,
      "train_samples_per_second": 15.755,
      "train_steps_per_second": 0.062
    }
  ],
  "logging_steps": 5,
  "max_steps": 996,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 5872633472090112.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}