{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1175,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 3.3722460551119577,
      "learning_rate": 1.6949152542372883e-07,
      "loss": 0.4424,
      "step": 1
    },
    {
      "epoch": 0.02,
      "grad_norm": 3.3291982249982532,
      "learning_rate": 8.474576271186441e-07,
      "loss": 0.521,
      "step": 5
    },
    {
      "epoch": 0.04,
      "grad_norm": 3.366718903705411,
      "learning_rate": 1.6949152542372882e-06,
      "loss": 0.5267,
      "step": 10
    },
    {
      "epoch": 0.06,
      "grad_norm": 3.743534573457503,
      "learning_rate": 2.5423728813559323e-06,
      "loss": 0.5054,
      "step": 15
    },
    {
      "epoch": 0.09,
      "grad_norm": 2.6726105105201854,
      "learning_rate": 3.3898305084745763e-06,
      "loss": 0.5138,
      "step": 20
    },
    {
      "epoch": 0.11,
      "grad_norm": 2.5592230066929766,
      "learning_rate": 4.23728813559322e-06,
      "loss": 0.4656,
      "step": 25
    },
    {
      "epoch": 0.13,
      "grad_norm": 2.5555759369628186,
      "learning_rate": 5.084745762711865e-06,
      "loss": 0.4712,
      "step": 30
    },
    {
      "epoch": 0.15,
      "grad_norm": 2.5046298800000617,
      "learning_rate": 5.932203389830509e-06,
      "loss": 0.4611,
      "step": 35
    },
    {
      "epoch": 0.17,
      "grad_norm": 2.7136428613817194,
      "learning_rate": 6.779661016949153e-06,
      "loss": 0.4541,
      "step": 40
    },
    {
      "epoch": 0.19,
      "grad_norm": 3.010241923008894,
      "learning_rate": 7.627118644067797e-06,
      "loss": 0.4781,
      "step": 45
    },
    {
      "epoch": 0.21,
      "grad_norm": 2.475959902357205,
      "learning_rate": 8.47457627118644e-06,
      "loss": 0.442,
      "step": 50
    },
    {
      "epoch": 0.23,
      "grad_norm": 3.128773493576812,
      "learning_rate": 9.322033898305085e-06,
      "loss": 0.426,
      "step": 55
    },
    {
      "epoch": 0.26,
      "grad_norm": 3.1009164137521252,
      "learning_rate": 1.016949152542373e-05,
      "loss": 0.4958,
      "step": 60
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.808266593808473,
      "learning_rate": 1.1016949152542374e-05,
      "loss": 0.46,
      "step": 65
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.6760608859412054,
      "learning_rate": 1.1864406779661018e-05,
      "loss": 0.4929,
      "step": 70
    },
    {
      "epoch": 0.32,
      "grad_norm": 3.900783318495709,
      "learning_rate": 1.2711864406779661e-05,
      "loss": 0.4618,
      "step": 75
    },
    {
      "epoch": 0.34,
      "grad_norm": 2.524216832199052,
      "learning_rate": 1.3559322033898305e-05,
      "loss": 0.4588,
      "step": 80
    },
    {
      "epoch": 0.36,
      "grad_norm": 60.979865346510024,
      "learning_rate": 1.440677966101695e-05,
      "loss": 0.6492,
      "step": 85
    },
    {
      "epoch": 0.38,
      "grad_norm": 15.672327487548742,
      "learning_rate": 1.5254237288135594e-05,
      "loss": 0.6082,
      "step": 90
    },
    {
      "epoch": 0.4,
      "grad_norm": 4.6849329984217345,
      "learning_rate": 1.6101694915254237e-05,
      "loss": 0.5659,
      "step": 95
    },
    {
      "epoch": 0.43,
      "grad_norm": 7.354203430908385,
      "learning_rate": 1.694915254237288e-05,
      "loss": 0.5671,
      "step": 100
    },
    {
      "epoch": 0.45,
      "grad_norm": 13.233658769725766,
      "learning_rate": 1.7796610169491526e-05,
      "loss": 0.6546,
      "step": 105
    },
    {
      "epoch": 0.47,
      "grad_norm": 8.816187325352404,
      "learning_rate": 1.864406779661017e-05,
      "loss": 0.6247,
      "step": 110
    },
    {
      "epoch": 0.49,
      "grad_norm": 110.07956321249138,
      "learning_rate": 1.9491525423728814e-05,
      "loss": 0.6283,
      "step": 115
    },
    {
      "epoch": 0.51,
      "grad_norm": 9.519856297335076,
      "learning_rate": 1.999982332362188e-05,
      "loss": 0.5585,
      "step": 120
    },
    {
      "epoch": 0.53,
      "grad_norm": 7.0894189196386375,
      "learning_rate": 1.999783578606323e-05,
      "loss": 0.5264,
      "step": 125
    },
    {
      "epoch": 0.55,
      "grad_norm": 5.230200581327651,
      "learning_rate": 1.999364030586835e-05,
      "loss": 0.5317,
      "step": 130
    },
    {
      "epoch": 0.57,
      "grad_norm": 4.219668396076119,
      "learning_rate": 1.9987237809575722e-05,
      "loss": 0.5372,
      "step": 135
    },
    {
      "epoch": 0.6,
      "grad_norm": 5.10590490035955,
      "learning_rate": 1.997862971112581e-05,
      "loss": 0.5236,
      "step": 140
    },
    {
      "epoch": 0.62,
      "grad_norm": 9.22741963642746,
      "learning_rate": 1.9967817911548796e-05,
      "loss": 0.5351,
      "step": 145
    },
    {
      "epoch": 0.64,
      "grad_norm": 3.2116021264594767,
      "learning_rate": 1.9954804798544748e-05,
      "loss": 0.477,
      "step": 150
    },
    {
      "epoch": 0.66,
      "grad_norm": 4.7279336245257255,
      "learning_rate": 1.993959324595634e-05,
      "loss": 0.5063,
      "step": 155
    },
    {
      "epoch": 0.68,
      "grad_norm": 2.6783717640693663,
      "learning_rate": 1.9922186613134152e-05,
      "loss": 0.4547,
      "step": 160
    },
    {
      "epoch": 0.7,
      "grad_norm": 4.6102800388264,
      "learning_rate": 1.9902588744194815e-05,
      "loss": 0.4618,
      "step": 165
    },
    {
      "epoch": 0.72,
      "grad_norm": 2.5223162544544246,
      "learning_rate": 1.9880803967172048e-05,
      "loss": 0.4532,
      "step": 170
    },
    {
      "epoch": 0.74,
      "grad_norm": 2.2521576224935687,
      "learning_rate": 1.985683709306085e-05,
      "loss": 0.4627,
      "step": 175
    },
    {
      "epoch": 0.77,
      "grad_norm": 2.363613216496382,
      "learning_rate": 1.983069341475504e-05,
      "loss": 0.4266,
      "step": 180
    },
    {
      "epoch": 0.79,
      "grad_norm": 2.319077028125871,
      "learning_rate": 1.9802378705878354e-05,
      "loss": 0.4294,
      "step": 185
    },
    {
      "epoch": 0.81,
      "grad_norm": 2.5983523553598675,
      "learning_rate": 1.9771899219509388e-05,
      "loss": 0.4463,
      "step": 190
    },
    {
      "epoch": 0.83,
      "grad_norm": 2.697869053891185,
      "learning_rate": 1.9739261686800662e-05,
      "loss": 0.3995,
      "step": 195
    },
    {
      "epoch": 0.85,
      "grad_norm": 2.2299084670054166,
      "learning_rate": 1.9704473315492072e-05,
      "loss": 0.4145,
      "step": 200
    },
    {
      "epoch": 0.87,
      "grad_norm": 2.144013684497221,
      "learning_rate": 1.966754178831916e-05,
      "loss": 0.4047,
      "step": 205
    },
    {
      "epoch": 0.89,
      "grad_norm": 2.165636444751405,
      "learning_rate": 1.962847526131642e-05,
      "loss": 0.429,
      "step": 210
    },
    {
      "epoch": 0.91,
      "grad_norm": 2.5009764776602066,
      "learning_rate": 1.9587282362016083e-05,
      "loss": 0.3987,
      "step": 215
    },
    {
      "epoch": 0.94,
      "grad_norm": 3.2454133915018946,
      "learning_rate": 1.9543972187542833e-05,
      "loss": 0.4196,
      "step": 220
    },
    {
      "epoch": 0.96,
      "grad_norm": 2.273670978657276,
      "learning_rate": 1.9498554302604768e-05,
      "loss": 0.4029,
      "step": 225
    },
    {
      "epoch": 0.98,
      "grad_norm": 3.004323662088254,
      "learning_rate": 1.9451038737381078e-05,
      "loss": 0.4268,
      "step": 230
    },
    {
      "epoch": 1.0,
      "grad_norm": 2.3784052825284845,
      "learning_rate": 1.940143598530701e-05,
      "loss": 0.4066,
      "step": 235
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.7733179926872253,
      "eval_runtime": 347.9521,
      "eval_samples_per_second": 21.575,
      "eval_steps_per_second": 0.339,
      "step": 235
    },
    {
      "epoch": 1.02,
      "grad_norm": 2.2553533504002683,
      "learning_rate": 1.9349757000756442e-05,
      "loss": 0.2356,
      "step": 240
    },
    {
      "epoch": 1.04,
      "grad_norm": 2.5450743747403717,
      "learning_rate": 1.929601319662271e-05,
      "loss": 0.2119,
      "step": 245
    },
    {
      "epoch": 1.06,
      "grad_norm": 3.778262007789548,
      "learning_rate": 1.9240216441798145e-05,
      "loss": 0.2295,
      "step": 250
    },
    {
      "epoch": 1.09,
      "grad_norm": 2.2858585923056536,
      "learning_rate": 1.918237905855295e-05,
      "loss": 0.2251,
      "step": 255
    },
    {
      "epoch": 1.11,
      "grad_norm": 1.7784488699701833,
      "learning_rate": 1.91225138198139e-05,
      "loss": 0.2145,
      "step": 260
    },
    {
      "epoch": 1.13,
      "grad_norm": 1.8254402227157958,
      "learning_rate": 1.906063394634356e-05,
      "loss": 0.2191,
      "step": 265
    },
    {
      "epoch": 1.15,
      "grad_norm": 1.9464300006155193,
      "learning_rate": 1.899675310382057e-05,
      "loss": 0.2273,
      "step": 270
    },
    {
      "epoch": 1.17,
      "grad_norm": 2.2936148112741757,
      "learning_rate": 1.8930885399821693e-05,
      "loss": 0.2246,
      "step": 275
    },
    {
      "epoch": 1.19,
      "grad_norm": 2.3799646925913245,
      "learning_rate": 1.8863045380706275e-05,
      "loss": 0.2303,
      "step": 280
    },
    {
      "epoch": 1.21,
      "grad_norm": 2.190554922996115,
      "learning_rate": 1.879324802840379e-05,
      "loss": 0.2318,
      "step": 285
    },
    {
      "epoch": 1.23,
      "grad_norm": 2.7452325011160035,
      "learning_rate": 1.8721508757105203e-05,
      "loss": 0.2276,
      "step": 290
    },
    {
      "epoch": 1.26,
      "grad_norm": 1.8163744421498897,
      "learning_rate": 1.864784340985887e-05,
      "loss": 0.24,
      "step": 295
    },
    {
      "epoch": 1.28,
      "grad_norm": 1.885101115960248,
      "learning_rate": 1.8572268255071718e-05,
      "loss": 0.2399,
      "step": 300
    },
    {
      "epoch": 1.3,
      "grad_norm": 1.760073718971802,
      "learning_rate": 1.8494799982916512e-05,
      "loss": 0.1955,
      "step": 305
    },
    {
      "epoch": 1.32,
      "grad_norm": 1.8829516569820812,
      "learning_rate": 1.8415455701645942e-05,
      "loss": 0.2035,
      "step": 310
    },
    {
      "epoch": 1.34,
      "grad_norm": 2.146880306353409,
      "learning_rate": 1.833425293381443e-05,
      "loss": 0.2271,
      "step": 315
    },
    {
      "epoch": 1.36,
      "grad_norm": 2.0534275082799684,
      "learning_rate": 1.8251209612408375e-05,
      "loss": 0.197,
      "step": 320
    },
    {
      "epoch": 1.38,
      "grad_norm": 2.0857973185832743,
      "learning_rate": 1.816634407688583e-05,
      "loss": 0.2337,
      "step": 325
    },
    {
      "epoch": 1.4,
      "grad_norm": 1.7319986338118718,
      "learning_rate": 1.807967506912636e-05,
      "loss": 0.1909,
      "step": 330
    },
    {
      "epoch": 1.43,
      "grad_norm": 1.7148628603629519,
      "learning_rate": 1.799122172929206e-05,
      "loss": 0.2103,
      "step": 335
    },
    {
      "epoch": 1.45,
      "grad_norm": 2.0404565885174786,
      "learning_rate": 1.7901003591600575e-05,
      "loss": 0.2045,
      "step": 340
    },
    {
      "epoch": 1.47,
      "grad_norm": 2.132532760084515,
      "learning_rate": 1.780904058001116e-05,
      "loss": 0.2234,
      "step": 345
    },
    {
      "epoch": 1.49,
      "grad_norm": 2.2388854318437774,
      "learning_rate": 1.7715353003824613e-05,
      "loss": 0.222,
      "step": 350
    },
    {
      "epoch": 1.51,
      "grad_norm": 2.1625096071245986,
      "learning_rate": 1.761996155319811e-05,
      "loss": 0.2278,
      "step": 355
    },
    {
      "epoch": 1.53,
      "grad_norm": 1.8994643849098976,
      "learning_rate": 1.7522887294575978e-05,
      "loss": 0.2114,
      "step": 360
    },
    {
      "epoch": 1.55,
      "grad_norm": 2.0232265086023213,
      "learning_rate": 1.742415166603733e-05,
      "loss": 0.2276,
      "step": 365
    },
    {
      "epoch": 1.57,
      "grad_norm": 2.0393231755488737,
      "learning_rate": 1.7323776472561625e-05,
      "loss": 0.2065,
      "step": 370
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.7156091041056394,
      "learning_rate": 1.7221783881213222e-05,
      "loss": 0.2201,
      "step": 375
    },
    {
      "epoch": 1.62,
      "grad_norm": 2.0249954720249645,
      "learning_rate": 1.7118196416245947e-05,
      "loss": 0.1986,
      "step": 380
    },
    {
      "epoch": 1.64,
      "grad_norm": 1.860444641374077,
      "learning_rate": 1.701303695412881e-05,
      "loss": 0.2224,
      "step": 385
    },
    {
      "epoch": 1.66,
      "grad_norm": 9.195021611366267,
      "learning_rate": 1.6906328718493906e-05,
      "loss": 0.2152,
      "step": 390
    },
    {
      "epoch": 1.68,
      "grad_norm": 2.790540417351318,
      "learning_rate": 1.679809527500765e-05,
      "loss": 0.2014,
      "step": 395
    },
    {
      "epoch": 1.7,
      "grad_norm": 1.9401261875054212,
      "learning_rate": 1.6688360526166514e-05,
      "loss": 0.2305,
      "step": 400
    },
    {
      "epoch": 1.72,
      "grad_norm": 2.3499169909762903,
      "learning_rate": 1.657714870601833e-05,
      "loss": 0.2163,
      "step": 405
    },
    {
      "epoch": 1.74,
      "grad_norm": 1.889841729135842,
      "learning_rate": 1.646448437481039e-05,
      "loss": 0.2285,
      "step": 410
    },
    {
      "epoch": 1.77,
      "grad_norm": 1.9100327330280613,
      "learning_rate": 1.635039241356553e-05,
      "loss": 0.2066,
      "step": 415
    },
    {
      "epoch": 1.79,
      "grad_norm": 2.0395782940541913,
      "learning_rate": 1.6234898018587336e-05,
      "loss": 0.2159,
      "step": 420
    },
    {
      "epoch": 1.81,
      "grad_norm": 2.0672756907905647,
      "learning_rate": 1.611802669589575e-05,
      "loss": 0.2211,
      "step": 425
    },
    {
      "epoch": 1.83,
      "grad_norm": 1.8644628412570703,
      "learning_rate": 1.5999804255594262e-05,
      "loss": 0.2123,
      "step": 430
    },
    {
      "epoch": 1.85,
      "grad_norm": 2.109246262952444,
      "learning_rate": 1.5880256806169954e-05,
      "loss": 0.2179,
      "step": 435
    },
    {
      "epoch": 1.87,
      "grad_norm": 2.0987747021618537,
      "learning_rate": 1.5759410748727663e-05,
      "loss": 0.2165,
      "step": 440
    },
    {
      "epoch": 1.89,
      "grad_norm": 1.8615315847750686,
      "learning_rate": 1.563729277115947e-05,
      "loss": 0.2115,
      "step": 445
    },
    {
      "epoch": 1.91,
      "grad_norm": 1.91470907138453,
      "learning_rate": 1.551392984225094e-05,
      "loss": 0.2226,
      "step": 450
    },
    {
      "epoch": 1.94,
      "grad_norm": 2.1673896851739,
      "learning_rate": 1.5389349205725244e-05,
      "loss": 0.2034,
      "step": 455
    },
    {
      "epoch": 1.96,
      "grad_norm": 1.9561220925261051,
      "learning_rate": 1.5263578374226607e-05,
      "loss": 0.2079,
      "step": 460
    },
    {
      "epoch": 1.98,
      "grad_norm": 1.8490634437024727,
      "learning_rate": 1.5136645123244366e-05,
      "loss": 0.2011,
      "step": 465
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.8370535270484316,
      "learning_rate": 1.5008577484978966e-05,
      "loss": 0.1901,
      "step": 470
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6736623048782349,
      "eval_runtime": 345.8336,
      "eval_samples_per_second": 21.707,
      "eval_steps_per_second": 0.341,
      "step": 470
    },
    {
      "epoch": 2.02,
      "grad_norm": 3.2591869308084505,
      "learning_rate": 1.4879403742151283e-05,
      "loss": 0.0766,
      "step": 475
    },
    {
      "epoch": 2.04,
      "grad_norm": 1.5781740604686478,
      "learning_rate": 1.4749152421756596e-05,
      "loss": 0.0809,
      "step": 480
    },
    {
      "epoch": 2.06,
      "grad_norm": 1.6789209629444652,
      "learning_rate": 1.4617852288764624e-05,
      "loss": 0.0728,
      "step": 485
    },
    {
      "epoch": 2.09,
      "grad_norm": 1.524275026933269,
      "learning_rate": 1.4485532339767036e-05,
      "loss": 0.0671,
      "step": 490
    },
    {
      "epoch": 2.11,
      "grad_norm": 1.911681176460919,
      "learning_rate": 1.4352221796573758e-05,
      "loss": 0.0829,
      "step": 495
    },
    {
      "epoch": 2.13,
      "grad_norm": 1.5314000006841326,
      "learning_rate": 1.421795009975957e-05,
      "loss": 0.0739,
      "step": 500
    },
    {
      "epoch": 2.15,
      "grad_norm": 1.8497162969927008,
      "learning_rate": 1.4082746902162414e-05,
      "loss": 0.0782,
      "step": 505
    },
    {
      "epoch": 2.17,
      "grad_norm": 1.5153675955770316,
      "learning_rate": 1.3946642062334765e-05,
      "loss": 0.0717,
      "step": 510
    },
    {
      "epoch": 2.19,
      "grad_norm": 1.4798583086240507,
      "learning_rate": 1.3809665637949636e-05,
      "loss": 0.0683,
      "step": 515
    },
    {
      "epoch": 2.21,
      "grad_norm": 1.5185665743912256,
      "learning_rate": 1.3671847879162562e-05,
      "loss": 0.0647,
      "step": 520
    },
    {
      "epoch": 2.23,
      "grad_norm": 1.6655269792696328,
      "learning_rate": 1.3533219221931102e-05,
      "loss": 0.0756,
      "step": 525
    },
    {
      "epoch": 2.26,
      "grad_norm": 1.4623074396711568,
      "learning_rate": 1.3393810281293294e-05,
      "loss": 0.079,
      "step": 530
    },
    {
      "epoch": 2.28,
      "grad_norm": 1.4769143347541074,
      "learning_rate": 1.3253651844606571e-05,
      "loss": 0.0644,
      "step": 535
    },
    {
      "epoch": 2.3,
      "grad_norm": 1.5924771297209446,
      "learning_rate": 1.311277486474862e-05,
      "loss": 0.0676,
      "step": 540
    },
    {
      "epoch": 2.32,
      "grad_norm": 1.4963368374266843,
      "learning_rate": 1.2971210453281675e-05,
      "loss": 0.0685,
      "step": 545
    },
    {
      "epoch": 2.34,
      "grad_norm": 1.3660066623187117,
      "learning_rate": 1.2828989873581786e-05,
      "loss": 0.0643,
      "step": 550
    },
    {
      "epoch": 2.36,
      "grad_norm": 1.5417608458745726,
      "learning_rate": 1.268614453393454e-05,
      "loss": 0.0704,
      "step": 555
    },
    {
      "epoch": 2.38,
      "grad_norm": 1.3386793668448367,
      "learning_rate": 1.2542705980598813e-05,
      "loss": 0.0725,
      "step": 560
    },
    {
      "epoch": 2.4,
      "grad_norm": 4.538383693449207,
      "learning_rate": 1.2398705890839988e-05,
      "loss": 0.0724,
      "step": 565
    },
    {
      "epoch": 2.43,
      "grad_norm": 1.5161704188772478,
      "learning_rate": 1.2254176065934332e-05,
      "loss": 0.069,
      "step": 570
    },
    {
      "epoch": 2.45,
      "grad_norm": 1.5190424960322728,
      "learning_rate": 1.2109148424145897e-05,
      "loss": 0.0675,
      "step": 575
    },
    {
      "epoch": 2.47,
      "grad_norm": 1.6418241298329592,
      "learning_rate": 1.1963654993677645e-05,
      "loss": 0.0706,
      "step": 580
    },
    {
      "epoch": 2.49,
      "grad_norm": 2.7746288115630415,
      "learning_rate": 1.1817727905598268e-05,
      "loss": 0.0661,
      "step": 585
    },
    {
      "epoch": 2.51,
      "grad_norm": 1.613430655848529,
      "learning_rate": 1.1671399386746301e-05,
      "loss": 0.0685,
      "step": 590
    },
    {
      "epoch": 2.53,
      "grad_norm": 1.652377714904139,
      "learning_rate": 1.1524701752613074e-05,
      "loss": 0.0685,
      "step": 595
    },
    {
      "epoch": 2.55,
      "grad_norm": 1.6108396859937457,
      "learning_rate": 1.13776674002061e-05,
      "loss": 0.0736,
      "step": 600
    },
    {
      "epoch": 2.57,
      "grad_norm": 1.4308175997531103,
      "learning_rate": 1.1230328800894437e-05,
      "loss": 0.0699,
      "step": 605
    },
    {
      "epoch": 2.6,
      "grad_norm": 1.529984546400154,
      "learning_rate": 1.108271849323767e-05,
      "loss": 0.0647,
      "step": 610
    },
    {
      "epoch": 2.62,
      "grad_norm": 1.7364294117046162,
      "learning_rate": 1.09348690758e-05,
      "loss": 0.0691,
      "step": 615
    },
    {
      "epoch": 2.64,
      "grad_norm": 1.4526326385257828,
      "learning_rate": 1.0786813199951145e-05,
      "loss": 0.0623,
      "step": 620
    },
    {
      "epoch": 2.66,
      "grad_norm": 1.5442010646463833,
      "learning_rate": 1.0638583562655498e-05,
      "loss": 0.0673,
      "step": 625
    },
    {
      "epoch": 2.68,
      "grad_norm": 1.4797041462885991,
      "learning_rate": 1.0490212899251308e-05,
      "loss": 0.0686,
      "step": 630
    },
    {
      "epoch": 2.7,
      "grad_norm": 1.650135612201066,
      "learning_rate": 1.0341733976221313e-05,
      "loss": 0.0715,
      "step": 635
    },
    {
      "epoch": 2.72,
      "grad_norm": 1.5265081306931187,
      "learning_rate": 1.0193179583956523e-05,
      "loss": 0.0642,
      "step": 640
    },
    {
      "epoch": 2.74,
      "grad_norm": 1.4887311962873813,
      "learning_rate": 1.0044582529514739e-05,
      "loss": 0.075,
      "step": 645
    },
    {
      "epoch": 2.77,
      "grad_norm": 1.4387050627018103,
      "learning_rate": 9.89597562937536e-06,
      "loss": 0.06,
      "step": 650
    },
    {
      "epoch": 2.79,
      "grad_norm": 1.58228556181931,
      "learning_rate": 9.747391702192132e-06,
      "loss": 0.0613,
      "step": 655
    },
    {
      "epoch": 2.81,
      "grad_norm": 1.424709805936743,
      "learning_rate": 9.598863561545404e-06,
      "loss": 0.0728,
      "step": 660
    },
    {
      "epoch": 2.83,
      "grad_norm": 1.3925788280661726,
      "learning_rate": 9.45042400869551e-06,
      "loss": 0.0638,
      "step": 665
    },
    {
      "epoch": 2.85,
      "grad_norm": 1.8579081866696643,
      "learning_rate": 9.302105825338876e-06,
      "loss": 0.0729,
      "step": 670
    },
    {
      "epoch": 2.87,
      "grad_norm": 1.4419650630198477,
      "learning_rate": 9.153941766368439e-06,
      "loss": 0.0605,
      "step": 675
    },
    {
      "epoch": 2.89,
      "grad_norm": 1.4587070966850302,
      "learning_rate": 9.005964552639983e-06,
      "loss": 0.0637,
      "step": 680
    },
    {
      "epoch": 2.91,
      "grad_norm": 1.400937857775751,
      "learning_rate": 8.858206863746018e-06,
      "loss": 0.0644,
      "step": 685
    },
    {
      "epoch": 2.94,
      "grad_norm": 1.4194625294878935,
      "learning_rate": 8.71070133079872e-06,
      "loss": 0.0708,
      "step": 690
    },
    {
      "epoch": 2.96,
      "grad_norm": 1.511069294541966,
      "learning_rate": 8.563480529223638e-06,
      "loss": 0.0667,
      "step": 695
    },
    {
      "epoch": 2.98,
      "grad_norm": 1.3579712613828467,
      "learning_rate": 8.41657697156566e-06,
      "loss": 0.0527,
      "step": 700
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.4627979346880684,
      "learning_rate": 8.270023100308865e-06,
      "loss": 0.068,
      "step": 705
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6575748324394226,
      "eval_runtime": 346.2698,
      "eval_samples_per_second": 21.68,
      "eval_steps_per_second": 0.341,
      "step": 705
    },
    {
      "epoch": 3.02,
      "grad_norm": 0.9187425982230547,
      "learning_rate": 8.123851280711877e-06,
      "loss": 0.0216,
      "step": 710
    },
    {
      "epoch": 3.04,
      "grad_norm": 1.0704321956753566,
      "learning_rate": 7.978093793660234e-06,
      "loss": 0.0187,
      "step": 715
    },
    {
      "epoch": 3.06,
      "grad_norm": 0.9253831765543917,
      "learning_rate": 7.832782828537437e-06,
      "loss": 0.0191,
      "step": 720
    },
    {
      "epoch": 3.09,
      "grad_norm": 0.9546574193581036,
      "learning_rate": 7.68795047611615e-06,
      "loss": 0.0186,
      "step": 725
    },
    {
      "epoch": 3.11,
      "grad_norm": 0.9527598588762864,
      "learning_rate": 7.543628721471234e-06,
      "loss": 0.0191,
      "step": 730
    },
    {
      "epoch": 3.13,
      "grad_norm": 1.0142369552642194,
      "learning_rate": 7.399849436916076e-06,
      "loss": 0.017,
      "step": 735
    },
    {
      "epoch": 3.15,
      "grad_norm": 0.8516277331105468,
      "learning_rate": 7.256644374963857e-06,
      "loss": 0.0169,
      "step": 740
    },
    {
      "epoch": 3.17,
      "grad_norm": 0.6998532587968802,
      "learning_rate": 7.11404516131526e-06,
      "loss": 0.0167,
      "step": 745
    },
    {
      "epoch": 3.19,
      "grad_norm": 0.9597297439764481,
      "learning_rate": 6.9720832878741776e-06,
      "loss": 0.0195,
      "step": 750
    },
    {
      "epoch": 3.21,
      "grad_norm": 0.8162423625488969,
      "learning_rate": 6.8307901057929735e-06,
      "loss": 0.0196,
      "step": 755
    },
    {
      "epoch": 3.23,
      "grad_norm": 1.0809022540934834,
      "learning_rate": 6.690196818548846e-06,
      "loss": 0.0202,
      "step": 760
    },
    {
      "epoch": 3.26,
      "grad_norm": 0.9451137823905494,
      "learning_rate": 6.550334475052767e-06,
      "loss": 0.0173,
      "step": 765
    },
    {
      "epoch": 3.28,
      "grad_norm": 1.0194236278105107,
      "learning_rate": 6.411233962792593e-06,
      "loss": 0.0217,
      "step": 770
    },
    {
      "epoch": 3.3,
      "grad_norm": 0.9428518118269854,
      "learning_rate": 6.2729260010117995e-06,
      "loss": 0.0205,
      "step": 775
    },
    {
      "epoch": 3.32,
      "grad_norm": 0.8298491758909723,
      "learning_rate": 6.135441133925382e-06,
      "loss": 0.0156,
      "step": 780
    },
    {
      "epoch": 3.34,
      "grad_norm": 0.9988001166823972,
      "learning_rate": 5.998809723974407e-06,
      "loss": 0.018,
      "step": 785
    },
    {
      "epoch": 3.36,
      "grad_norm": 0.8874287620178792,
      "learning_rate": 5.863061945120719e-06,
      "loss": 0.0172,
      "step": 790
    },
    {
      "epoch": 3.38,
      "grad_norm": 0.9021252322116792,
      "learning_rate": 5.728227776183244e-06,
      "loss": 0.0184,
      "step": 795
    },
    {
      "epoch": 3.4,
      "grad_norm": 0.899901563579598,
      "learning_rate": 5.594336994217416e-06,
      "loss": 0.0207,
      "step": 800
    },
    {
      "epoch": 3.43,
      "grad_norm": 0.7294818546149708,
      "learning_rate": 5.461419167939145e-06,
      "loss": 0.0154,
      "step": 805
    },
    {
      "epoch": 3.45,
      "grad_norm": 0.9595599651370788,
      "learning_rate": 5.329503651194805e-06,
      "loss": 0.0182,
      "step": 810
    },
    {
      "epoch": 3.47,
      "grad_norm": 0.7263044743513145,
      "learning_rate": 5.198619576478678e-06,
      "loss": 0.017,
      "step": 815
    },
    {
      "epoch": 3.49,
      "grad_norm": 0.9300662832706075,
      "learning_rate": 5.068795848499257e-06,
      "loss": 0.0182,
      "step": 820
    },
    {
      "epoch": 3.51,
      "grad_norm": 1.044597812236674,
      "learning_rate": 4.940061137795876e-06,
      "loss": 0.0154,
      "step": 825
    },
    {
      "epoch": 3.53,
      "grad_norm": 0.9405246884194852,
      "learning_rate": 4.812443874407059e-06,
      "loss": 0.0187,
      "step": 830
    },
    {
      "epoch": 3.55,
      "grad_norm": 0.8135204435661318,
      "learning_rate": 4.685972241591956e-06,
      "loss": 0.0156,
      "step": 835
    },
    {
      "epoch": 3.57,
      "grad_norm": 0.806855367473174,
      "learning_rate": 4.560674169606317e-06,
      "loss": 0.0165,
      "step": 840
    },
    {
      "epoch": 3.6,
      "grad_norm": 1.1283393882651565,
      "learning_rate": 4.436577329534291e-06,
      "loss": 0.0161,
      "step": 845
    },
    {
      "epoch": 3.62,
      "grad_norm": 1.0316895831074073,
      "learning_rate": 4.3137091271775e-06,
      "loss": 0.018,
      "step": 850
    },
    {
      "epoch": 3.64,
      "grad_norm": 0.83713135691551,
      "learning_rate": 4.192096697002686e-06,
      "loss": 0.0154,
      "step": 855
    },
    {
      "epoch": 3.66,
      "grad_norm": 0.8862197834996656,
      "learning_rate": 4.0717668961492725e-06,
      "loss": 0.014,
      "step": 860
    },
    {
      "epoch": 3.68,
      "grad_norm": 0.9409064835179552,
      "learning_rate": 3.9527462984981954e-06,
      "loss": 0.0185,
      "step": 865
    },
    {
      "epoch": 3.7,
      "grad_norm": 0.8141414364863905,
      "learning_rate": 3.8350611888032474e-06,
      "loss": 0.0167,
      "step": 870
    },
    {
      "epoch": 3.72,
      "grad_norm": 0.9491167080432446,
      "learning_rate": 3.718737556886316e-06,
      "loss": 0.0166,
      "step": 875
    },
    {
      "epoch": 3.74,
      "grad_norm": 0.8939541066316343,
      "learning_rate": 3.6038010918977308e-06,
      "loss": 0.0157,
      "step": 880
    },
    {
      "epoch": 3.77,
      "grad_norm": 1.1762213442708835,
      "learning_rate": 3.490277176643033e-06,
      "loss": 0.0176,
      "step": 885
    },
    {
      "epoch": 3.79,
      "grad_norm": 0.8409814896300148,
      "learning_rate": 3.378190881977359e-06,
      "loss": 0.0152,
      "step": 890
    },
    {
      "epoch": 3.81,
      "grad_norm": 1.0408829490759663,
      "learning_rate": 3.2675669612687565e-06,
      "loss": 0.0191,
      "step": 895
    },
    {
      "epoch": 3.83,
      "grad_norm": 0.9464969671060768,
      "learning_rate": 3.158429844931611e-06,
      "loss": 0.0156,
      "step": 900
    },
    {
      "epoch": 3.85,
      "grad_norm": 0.8183518112165244,
      "learning_rate": 3.0508036350313553e-06,
      "loss": 0.0155,
      "step": 905
    },
    {
      "epoch": 3.87,
      "grad_norm": 1.031893449336409,
      "learning_rate": 2.9447120999617363e-06,
      "loss": 0.0151,
      "step": 910
    },
    {
      "epoch": 3.89,
      "grad_norm": 0.9770328880326357,
      "learning_rate": 2.8401786691957632e-06,
      "loss": 0.0156,
      "step": 915
    },
    {
      "epoch": 3.91,
      "grad_norm": 0.9350326565557209,
      "learning_rate": 2.737226428111471e-06,
      "loss": 0.0176,
      "step": 920
    },
    {
      "epoch": 3.94,
      "grad_norm": 0.7498411603912395,
      "learning_rate": 2.635878112893717e-06,
      "loss": 0.0147,
      "step": 925
    },
    {
      "epoch": 3.96,
      "grad_norm": 0.890762978868864,
      "learning_rate": 2.5361561055130625e-06,
      "loss": 0.0179,
      "step": 930
    },
    {
      "epoch": 3.98,
      "grad_norm": 1.153428874102931,
      "learning_rate": 2.4380824287829073e-06,
      "loss": 0.0159,
      "step": 935
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.89001029327834,
      "learning_rate": 2.3416787414959097e-06,
      "loss": 0.0183,
      "step": 940
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.7071638107299805,
      "eval_runtime": 348.8132,
      "eval_samples_per_second": 21.522,
      "eval_steps_per_second": 0.338,
      "step": 940
    },
    {
      "epoch": 4.02,
      "grad_norm": 0.3600823717570703,
      "learning_rate": 2.246966333640823e-06,
      "loss": 0.0048,
      "step": 945
    },
    {
      "epoch": 4.04,
      "grad_norm": 0.33280820804988087,
      "learning_rate": 2.153966121700769e-06,
      "loss": 0.0049,
      "step": 950
    },
    {
      "epoch": 4.06,
      "grad_norm": 0.48812206732400915,
      "learning_rate": 2.0626986440340036e-06,
      "loss": 0.0058,
      "step": 955
    },
    {
      "epoch": 4.09,
      "grad_norm": 0.3662559033659471,
      "learning_rate": 1.973184056338173e-06,
      "loss": 0.0039,
      "step": 960
    },
    {
      "epoch": 4.11,
      "grad_norm": 0.3710554752839552,
      "learning_rate": 1.8854421271990964e-06,
      "loss": 0.0037,
      "step": 965
    },
    {
      "epoch": 4.13,
      "grad_norm": 0.3360503768539956,
      "learning_rate": 1.7994922337250276e-06,
      "loss": 0.0041,
      "step": 970
    },
    {
      "epoch": 4.15,
      "grad_norm": 0.32232522758989207,
      "learning_rate": 1.7153533572673708e-06,
      "loss": 0.0042,
      "step": 975
    },
    {
      "epoch": 4.17,
      "grad_norm": 0.3933577657073139,
      "learning_rate": 1.633044079228817e-06,
      "loss": 0.0042,
      "step": 980
    },
    {
      "epoch": 4.19,
      "grad_norm": 0.4133692185032901,
      "learning_rate": 1.5525825769597625e-06,
      "loss": 0.0042,
      "step": 985
    },
    {
      "epoch": 4.21,
      "grad_norm": 0.311775019674446,
      "learning_rate": 1.4739866197440046e-06,
      "loss": 0.004,
      "step": 990
    },
    {
      "epoch": 4.23,
      "grad_norm": 0.49759561364679256,
      "learning_rate": 1.3972735648745295e-06,
      "loss": 0.0049,
      "step": 995
    },
    {
      "epoch": 4.26,
      "grad_norm": 0.4477509727419261,
      "learning_rate": 1.3224603538202929e-06,
      "loss": 0.0041,
      "step": 1000
    },
    {
      "epoch": 4.28,
      "grad_norm": 0.4461256466898462,
      "learning_rate": 1.2495635084848356e-06,
      "loss": 0.0038,
      "step": 1005
    },
    {
      "epoch": 4.3,
      "grad_norm": 0.40708673174578486,
      "learning_rate": 1.1785991275575426e-06,
      "loss": 0.0039,
      "step": 1010
    },
    {
      "epoch": 4.32,
      "grad_norm": 0.4494486001323653,
      "learning_rate": 1.1095828829583844e-06,
      "loss": 0.0049,
      "step": 1015
    },
    {
      "epoch": 4.34,
      "grad_norm": 0.3778130066090292,
      "learning_rate": 1.0425300163768903e-06,
      "loss": 0.0042,
      "step": 1020
    },
    {
      "epoch": 4.36,
      "grad_norm": 0.3983653765993126,
      "learning_rate": 9.77455335906149e-07,
      "loss": 0.0053,
      "step": 1025
    },
    {
      "epoch": 4.38,
      "grad_norm": 0.3604607140996719,
      "learning_rate": 9.143732127725591e-07,
      "loss": 0.0039,
      "step": 1030
    },
    {
      "epoch": 4.4,
      "grad_norm": 0.6021048871397348,
      "learning_rate": 8.532975781620511e-07,
      "loss": 0.0035,
      "step": 1035
    },
    {
      "epoch": 4.43,
      "grad_norm": 0.3771138444026374,
      "learning_rate": 7.942419201435014e-07,
      "loss": 0.0041,
      "step": 1040
    },
    {
      "epoch": 4.45,
      "grad_norm": 0.40288900755654594,
      "learning_rate": 7.372192806899947e-07,
      "loss": 0.0045,
      "step": 1045
    },
    {
      "epoch": 4.47,
      "grad_norm": 0.4475530083250148,
      "learning_rate": 6.822422527986161e-07,
      "loss": 0.0043,
      "step": 1050
    },
    {
      "epoch": 4.49,
      "grad_norm": 0.3791478244996085,
      "learning_rate": 6.293229777093779e-07,
      "loss": 0.0046,
      "step": 1055
    },
    {
      "epoch": 4.51,
      "grad_norm": 0.3778438936995471,
      "learning_rate": 5.784731422239276e-07,
      "loss": 0.0038,
      "step": 1060
    },
    {
      "epoch": 4.53,
      "grad_norm": 0.376192442816014,
      "learning_rate": 5.297039761246137e-07,
      "loss": 0.0038,
      "step": 1065
    },
    {
      "epoch": 4.55,
      "grad_norm": 0.2960720923070219,
      "learning_rate": 4.830262496944693e-07,
      "loss": 0.0037,
      "step": 1070
    },
    {
      "epoch": 4.57,
      "grad_norm": 0.37323225545474764,
      "learning_rate": 4.384502713386918e-07,
      "loss": 0.0038,
      "step": 1075
    },
    {
      "epoch": 4.6,
      "grad_norm": 0.33848258454167024,
      "learning_rate": 3.9598588530810335e-07,
      "loss": 0.0037,
      "step": 1080
    },
    {
      "epoch": 4.62,
      "grad_norm": 0.4828649331052619,
      "learning_rate": 3.5564246952512817e-07,
      "loss": 0.0034,
      "step": 1085
    },
    {
      "epoch": 4.64,
      "grad_norm": 0.43489507591573195,
      "learning_rate": 3.174289335127612e-07,
      "loss": 0.0039,
      "step": 1090
    },
    {
      "epoch": 4.66,
      "grad_norm": 0.3646806476787612,
      "learning_rate": 2.8135371642695865e-07,
      "loss": 0.0042,
      "step": 1095
    },
    {
      "epoch": 4.68,
      "grad_norm": 0.5325215204533521,
      "learning_rate": 2.474247851929246e-07,
      "loss": 0.0048,
      "step": 1100
    },
    {
      "epoch": 4.7,
      "grad_norm": 0.40238125679038056,
      "learning_rate": 2.1564963274568028e-07,
      "loss": 0.0051,
      "step": 1105
    },
    {
      "epoch": 4.72,
      "grad_norm": 0.38702990805573456,
      "learning_rate": 1.860352763753004e-07,
      "loss": 0.0042,
      "step": 1110
    },
    {
      "epoch": 4.74,
      "grad_norm": 0.3852959902302661,
      "learning_rate": 1.585882561772112e-07,
      "loss": 0.0044,
      "step": 1115
    },
    {
      "epoch": 4.77,
      "grad_norm": 0.44717232901584403,
      "learning_rate": 1.333146336078528e-07,
      "loss": 0.0043,
      "step": 1120
    },
    {
      "epoch": 4.79,
      "grad_norm": 0.372462847171052,
      "learning_rate": 1.1021999014606322e-07,
      "loss": 0.0038,
      "step": 1125
    },
    {
      "epoch": 4.81,
      "grad_norm": 0.2928988966778955,
      "learning_rate": 8.930942606044434e-08,
      "loss": 0.0034,
      "step": 1130
    },
    {
      "epoch": 4.83,
      "grad_norm": 0.39889536364254635,
      "learning_rate": 7.058755928301631e-08,
      "loss": 0.0051,
      "step": 1135
    },
    {
      "epoch": 4.85,
      "grad_norm": 0.40786489098741363,
      "learning_rate": 5.405852438937764e-08,
      "loss": 0.0034,
      "step": 1140
    },
    {
      "epoch": 4.87,
      "grad_norm": 0.3349150763788282,
      "learning_rate": 3.972597168562131e-08,
      "loss": 0.0066,
      "step": 1145
    },
    {
      "epoch": 4.89,
      "grad_norm": 0.3597591461189066,
      "learning_rate": 2.7593066402189506e-08,
      "loss": 0.0037,
      "step": 1150
    },
    {
      "epoch": 4.91,
      "grad_norm": 0.4040878903041599,
      "learning_rate": 1.7662487994862808e-08,
      "loss": 0.0039,
      "step": 1155
    },
    {
      "epoch": 4.94,
      "grad_norm": 0.3992498945384279,
      "learning_rate": 9.93642955301799e-09,
      "loss": 0.0039,
      "step": 1160
    },
    {
      "epoch": 4.96,
      "grad_norm": 0.3821243876748289,
      "learning_rate": 4.4165973153076355e-09,
      "loss": 0.0044,
      "step": 1165
    },
    {
      "epoch": 4.98,
      "grad_norm": 0.3430681816581551,
      "learning_rate": 1.1042102928460196e-09,
      "loss": 0.0045,
      "step": 1170
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.32719513702973596,
      "learning_rate": 0.0,
      "loss": 0.004,
      "step": 1175
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.7659260034561157,
      "eval_runtime": 346.6983,
      "eval_samples_per_second": 21.653,
      "eval_steps_per_second": 0.34,
      "step": 1175
    },
    {
      "epoch": 5.0,
      "step": 1175,
      "total_flos": 246021095424000.0,
      "train_loss": 0.1592100728509274,
      "train_runtime": 9397.2382,
      "train_samples_per_second": 3.994,
      "train_steps_per_second": 0.125
    }
  ],
  "logging_steps": 5,
  "max_steps": 1175,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 100,
  "total_flos": 246021095424000.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}