{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9996249062265566,
  "eval_steps": 80,
  "global_step": 1066,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0046886721680420105,
      "grad_norm": 131.48858057089643,
      "learning_rate": 5.5e-07,
      "loss": 1.2775,
      "step": 5
    },
    {
      "epoch": 0.009377344336084021,
      "grad_norm": 33.55585411699123,
      "learning_rate": 1.1e-06,
      "loss": 1.3472,
      "step": 10
    },
    {
      "epoch": 0.014066016504126031,
      "grad_norm": 10.392565895406074,
      "learning_rate": 1.6499999999999999e-06,
      "loss": 1.1186,
      "step": 15
    },
    {
      "epoch": 0.018754688672168042,
      "grad_norm": 13.30940048870882,
      "learning_rate": 2.2e-06,
      "loss": 1.1381,
      "step": 20
    },
    {
      "epoch": 0.023443360840210054,
      "grad_norm": 25.46432441054564,
      "learning_rate": 2.75e-06,
      "loss": 1.1455,
      "step": 25
    },
    {
      "epoch": 0.028132033008252063,
      "grad_norm": 26.00538399730873,
      "learning_rate": 3.2999999999999997e-06,
      "loss": 1.049,
      "step": 30
    },
    {
      "epoch": 0.032820705176294075,
      "grad_norm": 9.35384218444162,
      "learning_rate": 3.8499999999999996e-06,
      "loss": 0.8965,
      "step": 35
    },
    {
      "epoch": 0.037509377344336084,
      "grad_norm": 2.9429262000270113,
      "learning_rate": 4.4e-06,
      "loss": 1.1271,
      "step": 40
    },
    {
      "epoch": 0.04219804951237809,
      "grad_norm": 3.2422986577839716,
      "learning_rate": 4.95e-06,
      "loss": 1.0061,
      "step": 45
    },
    {
      "epoch": 0.04688672168042011,
      "grad_norm": 6.503070007858331,
      "learning_rate": 5.5e-06,
      "loss": 0.9871,
      "step": 50
    },
    {
      "epoch": 0.05157539384846212,
      "grad_norm": 5.434874712126211,
      "learning_rate": 5.414406436166232e-06,
      "loss": 1.0584,
      "step": 55
    },
    {
      "epoch": 0.056264066016504126,
      "grad_norm": 2.394040425022367,
      "learning_rate": 5.32986463435603e-06,
      "loss": 1.0672,
      "step": 60
    },
    {
      "epoch": 0.060952738184546135,
      "grad_norm": 2.7812207561170776,
      "learning_rate": 5.246366801851234e-06,
      "loss": 0.9798,
      "step": 65
    },
    {
      "epoch": 0.06564141035258815,
      "grad_norm": 3.4321299409489927,
      "learning_rate": 5.163905165275343e-06,
      "loss": 1.0684,
      "step": 70
    },
    {
      "epoch": 0.07033008252063015,
      "grad_norm": 2.4331799242183005,
      "learning_rate": 5.082471970641763e-06,
      "loss": 1.1214,
      "step": 75
    },
    {
      "epoch": 0.07501875468867217,
      "grad_norm": 2.66350051316822,
      "learning_rate": 5.002059483402411e-06,
      "loss": 1.0422,
      "step": 80
    },
    {
      "epoch": 0.07501875468867217,
      "eval_loss": 1.0044387578964233,
      "eval_runtime": 21.7707,
      "eval_samples_per_second": 9.187,
      "eval_steps_per_second": 2.297,
      "step": 80
    },
    {
      "epoch": 0.07970742685671418,
      "grad_norm": 2.457085516495112,
      "learning_rate": 4.922659988496696e-06,
      "loss": 1.004,
      "step": 85
    },
    {
      "epoch": 0.08439609902475619,
      "grad_norm": 2.673304521252877,
      "learning_rate": 4.844265790400869e-06,
      "loss": 1.1774,
      "step": 90
    },
    {
      "epoch": 0.0890847711927982,
      "grad_norm": 2.9697389911450105,
      "learning_rate": 4.766869213177739e-06,
      "loss": 1.029,
      "step": 95
    },
    {
      "epoch": 0.09377344336084022,
      "grad_norm": 2.7485846803270175,
      "learning_rate": 4.690462600526791e-06,
      "loss": 1.0735,
      "step": 100
    },
    {
      "epoch": 0.09846211552888222,
      "grad_norm": 2.3597539988620047,
      "learning_rate": 4.615038315834675e-06,
      "loss": 1.0426,
      "step": 105
    },
    {
      "epoch": 0.10315078769692423,
      "grad_norm": 10.287138782877946,
      "learning_rate": 4.5405887422260886e-06,
      "loss": 1.1257,
      "step": 110
    },
    {
      "epoch": 0.10783945986496624,
      "grad_norm": 6.22099162659968,
      "learning_rate": 4.467106282615065e-06,
      "loss": 1.0203,
      "step": 115
    },
    {
      "epoch": 0.11252813203300825,
      "grad_norm": 6.918830274001748,
      "learning_rate": 4.394583359756651e-06,
      "loss": 1.1145,
      "step": 120
    },
    {
      "epoch": 0.11721680420105027,
      "grad_norm": 5.8742101902111,
      "learning_rate": 4.323012416298999e-06,
      "loss": 1.0943,
      "step": 125
    },
    {
      "epoch": 0.12190547636909227,
      "grad_norm": 2.6642490667255525,
      "learning_rate": 4.252385914835873e-06,
      "loss": 1.087,
      "step": 130
    },
    {
      "epoch": 0.12659414853713427,
      "grad_norm": 2.6965562685918023,
      "learning_rate": 4.182696337959566e-06,
      "loss": 1.0873,
      "step": 135
    },
    {
      "epoch": 0.1312828207051763,
      "grad_norm": 2.951044122767734,
      "learning_rate": 4.113936188314245e-06,
      "loss": 0.8704,
      "step": 140
    },
    {
      "epoch": 0.1359714928732183,
      "grad_norm": 2.101235354697959,
      "learning_rate": 4.046097988649726e-06,
      "loss": 1.1235,
      "step": 145
    },
    {
      "epoch": 0.1406601650412603,
      "grad_norm": 3.2330279427425617,
      "learning_rate": 3.979174281875685e-06,
      "loss": 0.997,
      "step": 150
    },
    {
      "epoch": 0.14534883720930233,
      "grad_norm": 3.3889833141888728,
      "learning_rate": 3.9131576311163e-06,
      "loss": 1.0179,
      "step": 155
    },
    {
      "epoch": 0.15003750937734434,
      "grad_norm": 4.973298096388189,
      "learning_rate": 3.848040619765356e-06,
      "loss": 0.9127,
      "step": 160
    },
    {
      "epoch": 0.15003750937734434,
      "eval_loss": 0.9606707692146301,
      "eval_runtime": 20.2418,
      "eval_samples_per_second": 9.881,
      "eval_steps_per_second": 2.47,
      "step": 160
    },
    {
      "epoch": 0.15472618154538634,
      "grad_norm": 8.29768606535662,
      "learning_rate": 3.7838158515417857e-06,
      "loss": 1.0887,
      "step": 165
    },
    {
      "epoch": 0.15941485371342837,
      "grad_norm": 3.7810993101840147,
      "learning_rate": 3.7204759505456866e-06,
      "loss": 0.8966,
      "step": 170
    },
    {
      "epoch": 0.16410352588147037,
      "grad_norm": 2.644220736384783,
      "learning_rate": 3.65801356131479e-06,
      "loss": 1.0543,
      "step": 175
    },
    {
      "epoch": 0.16879219804951237,
      "grad_norm": 3.614698732279852,
      "learning_rate": 3.596421348881407e-06,
      "loss": 0.9387,
      "step": 180
    },
    {
      "epoch": 0.1734808702175544,
      "grad_norm": 2.8015834886932787,
      "learning_rate": 3.535691998829856e-06,
      "loss": 1.0746,
      "step": 185
    },
    {
      "epoch": 0.1781695423855964,
      "grad_norm": 2.564293653383082,
      "learning_rate": 3.4758182173543725e-06,
      "loss": 0.8235,
      "step": 190
    },
    {
      "epoch": 0.1828582145536384,
      "grad_norm": 3.118218626122133,
      "learning_rate": 3.4167927313175065e-06,
      "loss": 0.9392,
      "step": 195
    },
    {
      "epoch": 0.18754688672168043,
      "grad_norm": 3.1320015313206078,
      "learning_rate": 3.358608288309036e-06,
      "loss": 0.9951,
      "step": 200
    },
    {
      "epoch": 0.19223555888972244,
      "grad_norm": 2.8693840620672026,
      "learning_rate": 3.3012576567053635e-06,
      "loss": 1.0835,
      "step": 205
    },
    {
      "epoch": 0.19692423105776444,
      "grad_norm": 3.961730546242666,
      "learning_rate": 3.2447336257294427e-06,
      "loss": 0.8606,
      "step": 210
    },
    {
      "epoch": 0.20161290322580644,
      "grad_norm": 3.380982676943891,
      "learning_rate": 3.189029005511225e-06,
      "loss": 1.0039,
      "step": 215
    },
    {
      "epoch": 0.20630157539384847,
      "grad_norm": 3.072769902074738,
      "learning_rate": 3.134136627148626e-06,
      "loss": 0.9235,
      "step": 220
    },
    {
      "epoch": 0.21099024756189047,
      "grad_norm": 2.245002916223262,
      "learning_rate": 3.080049342769041e-06,
      "loss": 1.06,
      "step": 225
    },
    {
      "epoch": 0.21567891972993247,
      "grad_norm": 2.541443132900714,
      "learning_rate": 3.026760025591393e-06,
      "loss": 0.9432,
      "step": 230
    },
    {
      "epoch": 0.2203675918979745,
      "grad_norm": 3.123240987262189,
      "learning_rate": 2.97426156998874e-06,
      "loss": 1.0341,
      "step": 235
    },
    {
      "epoch": 0.2250562640660165,
      "grad_norm": 2.404173594460732,
      "learning_rate": 2.9225468915514425e-06,
      "loss": 0.8605,
      "step": 240
    },
    {
      "epoch": 0.2250562640660165,
      "eval_loss": 0.944010317325592,
      "eval_runtime": 20.3476,
      "eval_samples_per_second": 9.829,
      "eval_steps_per_second": 2.457,
      "step": 240
    },
    {
      "epoch": 0.2297449362340585,
      "grad_norm": 2.839083125403713,
      "learning_rate": 2.8716089271509e-06,
      "loss": 0.986,
      "step": 245
    },
    {
      "epoch": 0.23443360840210054,
      "grad_norm": 3.109434158155652,
      "learning_rate": 2.8214406350038632e-06,
      "loss": 0.8936,
      "step": 250
    },
    {
      "epoch": 0.23912228057014254,
      "grad_norm": 2.5603812627943325,
      "learning_rate": 2.772034994737337e-06,
      "loss": 1.0787,
      "step": 255
    },
    {
      "epoch": 0.24381095273818454,
      "grad_norm": 2.843918945845048,
      "learning_rate": 2.7233850074540736e-06,
      "loss": 0.8681,
      "step": 260
    },
    {
      "epoch": 0.24849962490622657,
      "grad_norm": 3.295967379586164,
      "learning_rate": 2.6754836957986757e-06,
      "loss": 1.063,
      "step": 265
    },
    {
      "epoch": 0.25318829707426854,
      "grad_norm": 2.6450503587985867,
      "learning_rate": 2.6283241040243133e-06,
      "loss": 0.9345,
      "step": 270
    },
    {
      "epoch": 0.25787696924231057,
      "grad_norm": 2.6669681551281936,
      "learning_rate": 2.5818992980600576e-06,
      "loss": 0.9366,
      "step": 275
    },
    {
      "epoch": 0.2625656414103526,
      "grad_norm": 2.365910075512645,
      "learning_rate": 2.5362023655788563e-06,
      "loss": 0.9222,
      "step": 280
    },
    {
      "epoch": 0.2672543135783946,
      "grad_norm": 2.4773250316490336,
      "learning_rate": 2.491226416066151e-06,
      "loss": 0.9816,
      "step": 285
    },
    {
      "epoch": 0.2719429857464366,
      "grad_norm": 2.051844651359742,
      "learning_rate": 2.4469645808891426e-06,
      "loss": 1.0592,
      "step": 290
    },
    {
      "epoch": 0.27663165791447863,
      "grad_norm": 2.7110963822173875,
      "learning_rate": 2.40341001336673e-06,
      "loss": 0.9458,
      "step": 295
    },
    {
      "epoch": 0.2813203300825206,
      "grad_norm": 2.1769151172554047,
      "learning_rate": 2.3605558888401135e-06,
      "loss": 0.8555,
      "step": 300
    },
    {
      "epoch": 0.28600900225056264,
      "grad_norm": 2.626727731539155,
      "learning_rate": 2.318395404744094e-06,
      "loss": 1.1516,
      "step": 305
    },
    {
      "epoch": 0.29069767441860467,
      "grad_norm": 2.306595833971501,
      "learning_rate": 2.276921780679061e-06,
      "loss": 0.8737,
      "step": 310
    },
    {
      "epoch": 0.29538634658664664,
      "grad_norm": 2.569147248252582,
      "learning_rate": 2.2361282584836925e-06,
      "loss": 1.0032,
      "step": 315
    },
    {
      "epoch": 0.30007501875468867,
      "grad_norm": 1.9670956149750123,
      "learning_rate": 2.1960081023083778e-06,
      "loss": 0.9068,
      "step": 320
    },
    {
      "epoch": 0.30007501875468867,
      "eval_loss": 0.9342896938323975,
      "eval_runtime": 20.3147,
      "eval_samples_per_second": 9.845,
      "eval_steps_per_second": 2.461,
      "step": 320
    },
    {
      "epoch": 0.3047636909227307,
      "grad_norm": 2.0859945243647395,
      "learning_rate": 2.156554598689365e-06,
      "loss": 1.0191,
      "step": 325
    },
    {
      "epoch": 0.3094523630907727,
      "grad_norm": 1.9655466100226573,
      "learning_rate": 2.117761056623659e-06,
      "loss": 0.9328,
      "step": 330
    },
    {
      "epoch": 0.3141410352588147,
      "grad_norm": 2.581123721405676,
      "learning_rate": 2.0796208076446752e-06,
      "loss": 1.0101,
      "step": 335
    },
    {
      "epoch": 0.31882970742685673,
      "grad_norm": 2.443562264938451,
      "learning_rate": 2.0421272058986607e-06,
      "loss": 0.9154,
      "step": 340
    },
    {
      "epoch": 0.3235183795948987,
      "grad_norm": 2.3271716313484987,
      "learning_rate": 2.0052736282219008e-06,
      "loss": 1.044,
      "step": 345
    },
    {
      "epoch": 0.32820705176294074,
      "grad_norm": 2.4236414480748834,
      "learning_rate": 1.9690534742187182e-06,
      "loss": 1.0251,
      "step": 350
    },
    {
      "epoch": 0.33289572393098277,
      "grad_norm": 2.5311887907377253,
      "learning_rate": 1.9334601663402865e-06,
      "loss": 0.9126,
      "step": 355
    },
    {
      "epoch": 0.33758439609902474,
      "grad_norm": 2.19719594303758,
      "learning_rate": 1.898487149964267e-06,
      "loss": 0.9877,
      "step": 360
    },
    {
      "epoch": 0.34227306826706677,
      "grad_norm": 3.0147002131335805,
      "learning_rate": 1.8641278934752799e-06,
      "loss": 0.9057,
      "step": 365
    },
    {
      "epoch": 0.3469617404351088,
      "grad_norm": 2.134323364449907,
      "learning_rate": 1.8303758883462328e-06,
      "loss": 1.0681,
      "step": 370
    },
    {
      "epoch": 0.3516504126031508,
      "grad_norm": 2.3509776870701424,
      "learning_rate": 1.7972246492205194e-06,
      "loss": 0.9568,
      "step": 375
    },
    {
      "epoch": 0.3563390847711928,
      "grad_norm": 2.1090286048701814,
      "learning_rate": 1.7646677139950976e-06,
      "loss": 1.032,
      "step": 380
    },
    {
      "epoch": 0.36102775693923483,
      "grad_norm": 2.36149114890146,
      "learning_rate": 1.7326986439044696e-06,
      "loss": 0.9867,
      "step": 385
    },
    {
      "epoch": 0.3657164291072768,
      "grad_norm": 2.255024349073518,
      "learning_rate": 1.701311023605583e-06,
      "loss": 0.8747,
      "step": 390
    },
    {
      "epoch": 0.37040510127531884,
      "grad_norm": 2.6380764656009923,
      "learning_rate": 1.6704984612636572e-06,
      "loss": 0.9224,
      "step": 395
    },
    {
      "epoch": 0.37509377344336087,
      "grad_norm": 2.4870934478859126,
      "learning_rate": 1.6402545886389659e-06,
      "loss": 0.9147,
      "step": 400
    },
    {
      "epoch": 0.37509377344336087,
      "eval_loss": 0.9293374419212341,
      "eval_runtime": 20.325,
      "eval_samples_per_second": 9.84,
      "eval_steps_per_second": 2.46,
      "step": 400
    },
    {
      "epoch": 0.37978244561140284,
      "grad_norm": 2.5774447639043236,
      "learning_rate": 1.610573061174586e-06,
      "loss": 0.9285,
      "step": 405
    },
    {
      "epoch": 0.38447111777944487,
      "grad_norm": 2.2678788716470817,
      "learning_rate": 1.5814475580851346e-06,
      "loss": 0.9994,
      "step": 410
    },
    {
      "epoch": 0.38915978994748684,
      "grad_norm": 2.8722747503884114,
      "learning_rate": 1.5528717824465089e-06,
      "loss": 0.8864,
      "step": 415
    },
    {
      "epoch": 0.3938484621155289,
      "grad_norm": 2.436535560556849,
      "learning_rate": 1.5248394612866496e-06,
      "loss": 1.1302,
      "step": 420
    },
    {
      "epoch": 0.3985371342835709,
      "grad_norm": 1.9720574518399083,
      "learning_rate": 1.4973443456773522e-06,
      "loss": 0.9394,
      "step": 425
    },
    {
      "epoch": 0.4032258064516129,
      "grad_norm": 2.6156706265956258,
      "learning_rate": 1.4703802108271373e-06,
      "loss": 0.7922,
      "step": 430
    },
    {
      "epoch": 0.4079144786196549,
      "grad_norm": 2.295310925977984,
      "learning_rate": 1.4439408561752077e-06,
      "loss": 0.9312,
      "step": 435
    },
    {
      "epoch": 0.41260315078769694,
      "grad_norm": 2.147812597078933,
      "learning_rate": 1.4180201054865116e-06,
      "loss": 0.9859,
      "step": 440
    },
    {
      "epoch": 0.4172918229557389,
      "grad_norm": 2.029009456836816,
      "learning_rate": 1.392611806947934e-06,
      "loss": 1.0307,
      "step": 445
    },
    {
      "epoch": 0.42198049512378094,
      "grad_norm": 2.590837380389044,
      "learning_rate": 1.3677098332656357e-06,
      "loss": 0.92,
      "step": 450
    },
    {
      "epoch": 0.42666916729182297,
      "grad_norm": 2.3763009955883505,
      "learning_rate": 1.3433080817635696e-06,
      "loss": 1.0955,
      "step": 455
    },
    {
      "epoch": 0.43135783945986494,
      "grad_norm": 2.80941534379534,
      "learning_rate": 1.3194004744831898e-06,
      "loss": 0.903,
      "step": 460
    },
    {
      "epoch": 0.436046511627907,
      "grad_norm": 2.14656443429063,
      "learning_rate": 1.2959809582843855e-06,
      "loss": 0.8284,
      "step": 465
    },
    {
      "epoch": 0.440735183795949,
      "grad_norm": 2.7676121711339423,
      "learning_rate": 1.273043504947661e-06,
      "loss": 1.0462,
      "step": 470
    },
    {
      "epoch": 0.445423855963991,
      "grad_norm": 2.44008408182776,
      "learning_rate": 1.2505821112775862e-06,
      "loss": 0.979,
      "step": 475
    },
    {
      "epoch": 0.450112528132033,
      "grad_norm": 2.584889518443889,
      "learning_rate": 1.2285907992075474e-06,
      "loss": 1.0192,
      "step": 480
    },
    {
      "epoch": 0.450112528132033,
      "eval_loss": 0.9250276684761047,
      "eval_runtime": 20.5471,
      "eval_samples_per_second": 9.734,
      "eval_steps_per_second": 2.433,
      "step": 480
    },
    {
      "epoch": 0.45480120030007504,
      "grad_norm": 2.496373694809835,
      "learning_rate": 1.207063615905829e-06,
      "loss": 0.9037,
      "step": 485
    },
    {
      "epoch": 0.459489872468117,
      "grad_norm": 2.371873747415523,
      "learning_rate": 1.1859946338830404e-06,
      "loss": 1.0312,
      "step": 490
    },
    {
      "epoch": 0.46417854463615904,
      "grad_norm": 2.2570285962672894,
      "learning_rate": 1.1653779511009372e-06,
      "loss": 0.9113,
      "step": 495
    },
    {
      "epoch": 0.46886721680420107,
      "grad_norm": 2.6279198777243606,
      "learning_rate": 1.145207691082648e-06,
      "loss": 0.8337,
      "step": 500
    },
    {
      "epoch": 0.47355588897224304,
      "grad_norm": 2.7761777412837767,
      "learning_rate": 1.1254780030243539e-06,
      "loss": 0.9602,
      "step": 505
    },
    {
      "epoch": 0.4782445611402851,
      "grad_norm": 2.5445962149324663,
      "learning_rate": 1.1061830619084358e-06,
      "loss": 0.9804,
      "step": 510
    },
    {
      "epoch": 0.4829332333083271,
      "grad_norm": 2.3633364319931514,
      "learning_rate": 1.087317068618139e-06,
      "loss": 0.977,
      "step": 515
    },
    {
      "epoch": 0.4876219054763691,
      "grad_norm": 1.9986528249443352,
      "learning_rate": 1.0688742500537784e-06,
      "loss": 0.9425,
      "step": 520
    },
    {
      "epoch": 0.4923105776444111,
      "grad_norm": 2.4501412442964803,
      "learning_rate": 1.0508488592505175e-06,
      "loss": 0.9995,
      "step": 525
    },
    {
      "epoch": 0.49699924981245314,
      "grad_norm": 2.4394994412302786,
      "learning_rate": 1.0332351754977698e-06,
      "loss": 0.9329,
      "step": 530
    },
    {
      "epoch": 0.5016879219804952,
      "grad_norm": 2.2074019259444544,
      "learning_rate": 1.016027504460246e-06,
      "loss": 0.9203,
      "step": 535
    },
    {
      "epoch": 0.5063765941485371,
      "grad_norm": 2.610216953517823,
      "learning_rate": 9.992201783006927e-07,
      "loss": 1.0291,
      "step": 540
    },
    {
      "epoch": 0.5110652663165791,
      "grad_norm": 2.718383436345948,
      "learning_rate": 9.828075558043617e-07,
      "loss": 0.9292,
      "step": 545
    },
    {
      "epoch": 0.5157539384846211,
      "grad_norm": 3.6229517535213143,
      "learning_rate": 9.667840225052484e-07,
      "loss": 1.0165,
      "step": 550
    },
    {
      "epoch": 0.5204426106526632,
      "grad_norm": 2.566929441984276,
      "learning_rate": 9.511439908141446e-07,
      "loss": 1.012,
      "step": 555
    },
    {
      "epoch": 0.5251312828207052,
      "grad_norm": 2.2133821736785015,
      "learning_rate": 9.358819001485473e-07,
      "loss": 0.8303,
      "step": 560
    },
    {
      "epoch": 0.5251312828207052,
      "eval_loss": 0.9191934466362,
      "eval_runtime": 20.5275,
      "eval_samples_per_second": 9.743,
      "eval_steps_per_second": 2.436,
      "step": 560
    },
    {
      "epoch": 0.5298199549887472,
      "grad_norm": 2.530396363584249,
      "learning_rate": 9.209922170644708e-07,
      "loss": 1.022,
      "step": 565
    },
    {
      "epoch": 0.5345086271567892,
      "grad_norm": 3.0761740344436017,
      "learning_rate": 9.06469435390206e-07,
      "loss": 0.9437,
      "step": 570
    },
    {
      "epoch": 0.5391972993248312,
      "grad_norm": 2.679955680751534,
      "learning_rate": 8.923080763620794e-07,
      "loss": 0.9776,
      "step": 575
    },
    {
      "epoch": 0.5438859714928732,
      "grad_norm": 2.477813948279551,
      "learning_rate": 8.785026887622588e-07,
      "loss": 0.8858,
      "step": 580
    },
    {
      "epoch": 0.5485746436609152,
      "grad_norm": 3.021641574122816,
      "learning_rate": 8.650478490586582e-07,
      "loss": 0.9392,
      "step": 585
    },
    {
      "epoch": 0.5532633158289573,
      "grad_norm": 3.0622821568316985,
      "learning_rate": 8.519381615469985e-07,
      "loss": 1.0067,
      "step": 590
    },
    {
      "epoch": 0.5579519879969993,
      "grad_norm": 2.0240771327720606,
      "learning_rate": 8.391682584950767e-07,
      "loss": 0.8645,
      "step": 595
    },
    {
      "epoch": 0.5626406601650412,
      "grad_norm": 2.4592599613454453,
      "learning_rate": 8.267328002892997e-07,
      "loss": 0.8116,
      "step": 600
    },
    {
      "epoch": 0.5673293323330832,
      "grad_norm": 2.3279060484215304,
      "learning_rate": 8.146264755835511e-07,
      "loss": 1.0685,
      "step": 605
    },
    {
      "epoch": 0.5720180045011253,
      "grad_norm": 2.1915343114528394,
      "learning_rate": 8.028440014504431e-07,
      "loss": 1.0312,
      "step": 610
    },
    {
      "epoch": 0.5767066766691673,
      "grad_norm": 2.353311242466667,
      "learning_rate": 7.913801235350256e-07,
      "loss": 0.9753,
      "step": 615
    },
    {
      "epoch": 0.5813953488372093,
      "grad_norm": 1.9822254771453949,
      "learning_rate": 7.80229616211014e-07,
      "loss": 1.0626,
      "step": 620
    },
    {
      "epoch": 0.5860840210052514,
      "grad_norm": 2.229570765456582,
      "learning_rate": 7.693872827396111e-07,
      "loss": 0.8915,
      "step": 625
    },
    {
      "epoch": 0.5907726931732933,
      "grad_norm": 2.1333212031989017,
      "learning_rate": 7.58847955430991e-07,
      "loss": 0.9256,
      "step": 630
    },
    {
      "epoch": 0.5954613653413353,
      "grad_norm": 2.3285962864619107,
      "learning_rate": 7.486064958085216e-07,
      "loss": 0.8844,
      "step": 635
    },
    {
      "epoch": 0.6001500375093773,
      "grad_norm": 2.4239895162113654,
      "learning_rate": 7.386577947758049e-07,
      "loss": 1.0284,
      "step": 640
    },
    {
      "epoch": 0.6001500375093773,
      "eval_loss": 0.9292100667953491,
      "eval_runtime": 20.2551,
      "eval_samples_per_second": 9.874,
      "eval_steps_per_second": 2.469,
      "step": 640
    },
    {
      "epoch": 0.6048387096774194,
      "grad_norm": 2.592229243089013,
      "learning_rate": 7.289967727866171e-07,
      "loss": 0.8607,
      "step": 645
    },
    {
      "epoch": 0.6095273818454614,
      "grad_norm": 2.557096114560489,
      "learning_rate": 7.196183800178289e-07,
      "loss": 1.0461,
      "step": 650
    },
    {
      "epoch": 0.6142160540135033,
      "grad_norm": 2.7258364788939278,
      "learning_rate": 7.105175965454019e-07,
      "loss": 0.8923,
      "step": 655
    },
    {
      "epoch": 0.6189047261815454,
      "grad_norm": 2.500897088452433,
      "learning_rate": 7.016894325235454e-07,
      "loss": 0.908,
      "step": 660
    },
    {
      "epoch": 0.6235933983495874,
      "grad_norm": 3.4364862124732873,
      "learning_rate": 6.931289283671353e-07,
      "loss": 0.8488,
      "step": 665
    },
    {
      "epoch": 0.6282820705176294,
      "grad_norm": 2.6544071650230405,
      "learning_rate": 6.84831154937491e-07,
      "loss": 0.815,
      "step": 670
    },
    {
      "epoch": 0.6329707426856714,
      "grad_norm": 2.9017321933643356,
      "learning_rate": 6.767912137316187e-07,
      "loss": 1.0641,
      "step": 675
    },
    {
      "epoch": 0.6376594148537135,
      "grad_norm": 3.2461446427637854,
      "learning_rate": 6.690042370750264e-07,
      "loss": 0.9388,
      "step": 680
    },
    {
      "epoch": 0.6423480870217554,
      "grad_norm": 2.5936370960610398,
      "learning_rate": 6.614653883182271e-07,
      "loss": 0.83,
      "step": 685
    },
    {
      "epoch": 0.6470367591897974,
      "grad_norm": 3.027451608426823,
      "learning_rate": 6.541698620370481e-07,
      "loss": 0.9852,
      "step": 690
    },
    {
      "epoch": 0.6517254313578394,
      "grad_norm": 2.8965181885799933,
      "learning_rate": 6.471128842368711e-07,
      "loss": 0.8914,
      "step": 695
    },
    {
      "epoch": 0.6564141035258815,
      "grad_norm": 2.9030689776107,
      "learning_rate": 6.402897125609332e-07,
      "loss": 0.9833,
      "step": 700
    },
    {
      "epoch": 0.6611027756939235,
      "grad_norm": 2.6265372108049303,
      "learning_rate": 6.336956365028259e-07,
      "loss": 1.0902,
      "step": 705
    },
    {
      "epoch": 0.6657914478619655,
      "grad_norm": 2.186738270502514,
      "learning_rate": 6.273259776233337e-07,
      "loss": 0.8316,
      "step": 710
    },
    {
      "epoch": 0.6704801200300075,
      "grad_norm": 10.89006237530139,
      "learning_rate": 6.211760897717641e-07,
      "loss": 1.0283,
      "step": 715
    },
    {
      "epoch": 0.6751687921980495,
      "grad_norm": 2.7796220586562344,
      "learning_rate": 6.152413593119235e-07,
      "loss": 0.9183,
      "step": 720
    },
    {
      "epoch": 0.6751687921980495,
      "eval_loss": 0.9389083385467529,
      "eval_runtime": 20.3673,
      "eval_samples_per_second": 9.82,
      "eval_steps_per_second": 2.455,
      "step": 720
    },
    {
      "epoch": 0.6798574643660915,
      "grad_norm": 2.906995236208513,
      "learning_rate": 6.095172053529076e-07,
      "loss": 1.034,
      "step": 725
    },
    {
      "epoch": 0.6845461365341335,
      "grad_norm": 2.3367410989611943,
      "learning_rate": 6.039990799848741e-07,
      "loss": 0.898,
      "step": 730
    },
    {
      "epoch": 0.6892348087021756,
      "grad_norm": 2.3266907504669976,
      "learning_rate": 5.986824685199863e-07,
      "loss": 0.7855,
      "step": 735
    },
    {
      "epoch": 0.6939234808702176,
      "grad_norm": 2.1566038379464842,
      "learning_rate": 5.935628897387149e-07,
      "loss": 1.0417,
      "step": 740
    },
    {
      "epoch": 0.6986121530382595,
      "grad_norm": 2.3519944231674685,
      "learning_rate": 5.886358961416999e-07,
      "loss": 0.9567,
      "step": 745
    },
    {
      "epoch": 0.7033008252063015,
      "grad_norm": 2.5749570284624457,
      "learning_rate": 5.838970742073876e-07,
      "loss": 0.7935,
      "step": 750
    },
    {
      "epoch": 0.7079894973743436,
      "grad_norm": 2.0032057277402844,
      "learning_rate": 5.793420446556638e-07,
      "loss": 0.9967,
      "step": 755
    },
    {
      "epoch": 0.7126781695423856,
      "grad_norm": 2.8692030696523703,
      "learning_rate": 5.74966462717722e-07,
      "loss": 1.0786,
      "step": 760
    },
    {
      "epoch": 0.7173668417104276,
      "grad_norm": 2.1378421674086994,
      "learning_rate": 5.707660184124143e-07,
      "loss": 1.115,
      "step": 765
    },
    {
      "epoch": 0.7220555138784697,
      "grad_norm": 2.344543312884903,
      "learning_rate": 5.667364368293497e-07,
      "loss": 0.9502,
      "step": 770
    },
    {
      "epoch": 0.7267441860465116,
      "grad_norm": 1.9323329729740275,
      "learning_rate": 5.6287347841902e-07,
      "loss": 0.9964,
      "step": 775
    },
    {
      "epoch": 0.7314328582145536,
      "grad_norm": 2.948582260278745,
      "learning_rate": 5.591729392902467e-07,
      "loss": 1.0077,
      "step": 780
    },
    {
      "epoch": 0.7361215303825956,
      "grad_norm": 2.1985962635350442,
      "learning_rate": 5.556306515152638e-07,
      "loss": 0.7058,
      "step": 785
    },
    {
      "epoch": 0.7408102025506377,
      "grad_norm": 4.099442910663203,
      "learning_rate": 5.522424834427688e-07,
      "loss": 0.856,
      "step": 790
    },
    {
      "epoch": 0.7454988747186797,
      "grad_norm": 2.3260610179486787,
      "learning_rate": 5.490043400192936e-07,
      "loss": 0.9852,
      "step": 795
    },
    {
      "epoch": 0.7501875468867217,
      "grad_norm": 2.2249724666107693,
      "learning_rate": 5.459121631192727e-07,
      "loss": 0.9897,
      "step": 800
    },
    {
      "epoch": 0.7501875468867217,
      "eval_loss": 0.9337027072906494,
      "eval_runtime": 20.543,
      "eval_samples_per_second": 9.736,
      "eval_steps_per_second": 2.434,
      "step": 800
    },
    {
      "epoch": 0.7548762190547637,
      "grad_norm": 2.3238862226735244,
      "learning_rate": 5.429619318842062e-07,
      "loss": 0.9693,
      "step": 805
    },
    {
      "epoch": 0.7595648912228057,
      "grad_norm": 2.5983979780721005,
      "learning_rate": 5.401496630713439e-07,
      "loss": 0.9363,
      "step": 810
    },
    {
      "epoch": 0.7642535633908477,
      "grad_norm": 3.203392479652258,
      "learning_rate": 5.374714114123462e-07,
      "loss": 0.9007,
      "step": 815
    },
    {
      "epoch": 0.7689422355588897,
      "grad_norm": 3.0116714434819523,
      "learning_rate": 5.34923269982403e-07,
      "loss": 0.8538,
      "step": 820
    },
    {
      "epoch": 0.7736309077269318,
      "grad_norm": 2.735036803752382,
      "learning_rate": 5.325013705803326e-07,
      "loss": 0.9429,
      "step": 825
    },
    {
      "epoch": 0.7783195798949737,
      "grad_norm": 2.7192900196409693,
      "learning_rate": 5.302018841202155e-07,
      "loss": 0.862,
      "step": 830
    },
    {
      "epoch": 0.7830082520630157,
      "grad_norm": 3.310126483594716,
      "learning_rate": 5.28021021035156e-07,
      "loss": 0.8743,
      "step": 835
    },
    {
      "epoch": 0.7876969242310577,
      "grad_norm": 3.5788231445835623,
      "learning_rate": 5.25955031693814e-07,
      "loss": 0.8728,
      "step": 840
    },
    {
      "epoch": 0.7923855963990998,
      "grad_norm": 2.3031948200618784,
      "learning_rate": 5.240002068303935e-07,
      "loss": 0.9804,
      "step": 845
    },
    {
      "epoch": 0.7970742685671418,
      "grad_norm": 2.27441145876717,
      "learning_rate": 5.22152877988829e-07,
      "loss": 0.9592,
      "step": 850
    },
    {
      "epoch": 0.8017629407351838,
      "grad_norm": 2.399760421263353,
      "learning_rate": 5.204094179819663e-07,
      "loss": 1.1113,
      "step": 855
    },
    {
      "epoch": 0.8064516129032258,
      "grad_norm": 2.786650153581553,
      "learning_rate": 5.187662413666055e-07,
      "loss": 0.896,
      "step": 860
    },
    {
      "epoch": 0.8111402850712678,
      "grad_norm": 2.3375248528715886,
      "learning_rate": 5.137741469209312e-07,
      "loss": 0.9484,
      "step": 865
    },
    {
      "epoch": 0.8158289572393098,
      "grad_norm": 2.107239092918484,
      "learning_rate": 5.034812576940423e-07,
      "loss": 0.9107,
      "step": 870
    },
    {
      "epoch": 0.8205176294073518,
      "grad_norm": 3.2120652639502136,
      "learning_rate": 4.931883684671535e-07,
      "loss": 1.0274,
      "step": 875
    },
    {
      "epoch": 0.8252063015753939,
      "grad_norm": 2.1470972788703353,
      "learning_rate": 4.828954792402647e-07,
      "loss": 1.0209,
      "step": 880
    },
    {
      "epoch": 0.8252063015753939,
      "eval_loss": 0.93896484375,
      "eval_runtime": 20.3111,
      "eval_samples_per_second": 9.847,
      "eval_steps_per_second": 2.462,
      "step": 880
    },
    {
      "epoch": 0.8298949737434359,
      "grad_norm": 2.0424483656754724,
      "learning_rate": 4.7260259001337577e-07,
      "loss": 0.9472,
      "step": 885
    },
    {
      "epoch": 0.8345836459114778,
      "grad_norm": 2.5878530403298434,
      "learning_rate": 4.6230970078648696e-07,
      "loss": 0.9458,
      "step": 890
    },
    {
      "epoch": 0.8392723180795199,
      "grad_norm": 2.497909901428293,
      "learning_rate": 4.5201681155959816e-07,
      "loss": 0.7928,
      "step": 895
    },
    {
      "epoch": 0.8439609902475619,
      "grad_norm": 2.1704945555054462,
      "learning_rate": 4.417239223327093e-07,
      "loss": 0.8992,
      "step": 900
    },
    {
      "epoch": 0.8486496624156039,
      "grad_norm": 1.9807380776486385,
      "learning_rate": 4.314310331058205e-07,
      "loss": 1.0681,
      "step": 905
    },
    {
      "epoch": 0.8533383345836459,
      "grad_norm": 2.2057348642236625,
      "learning_rate": 4.2113814387893164e-07,
      "loss": 1.057,
      "step": 910
    },
    {
      "epoch": 0.858027006751688,
      "grad_norm": 2.434197081864654,
      "learning_rate": 4.108452546520428e-07,
      "loss": 0.9609,
      "step": 915
    },
    {
      "epoch": 0.8627156789197299,
      "grad_norm": 1.641371038860168,
      "learning_rate": 4.00552365425154e-07,
      "loss": 0.9775,
      "step": 920
    },
    {
      "epoch": 0.8674043510877719,
      "grad_norm": 2.207714901698735,
      "learning_rate": 3.9025947619826517e-07,
      "loss": 0.7712,
      "step": 925
    },
    {
      "epoch": 0.872093023255814,
      "grad_norm": 2.011772170698376,
      "learning_rate": 3.7996658697137626e-07,
      "loss": 0.9408,
      "step": 930
    },
    {
      "epoch": 0.876781695423856,
      "grad_norm": 2.5252414435017747,
      "learning_rate": 3.6967369774448745e-07,
      "loss": 0.8363,
      "step": 935
    },
    {
      "epoch": 0.881470367591898,
      "grad_norm": 2.3365392350792287,
      "learning_rate": 3.5938080851759865e-07,
      "loss": 0.9977,
      "step": 940
    },
    {
      "epoch": 0.88615903975994,
      "grad_norm": 2.3882731527862346,
      "learning_rate": 3.490879192907099e-07,
      "loss": 0.9326,
      "step": 945
    },
    {
      "epoch": 0.890847711927982,
      "grad_norm": 2.180049227770247,
      "learning_rate": 3.38795030063821e-07,
      "loss": 0.9909,
      "step": 950
    },
    {
      "epoch": 0.895536384096024,
      "grad_norm": 2.4546116544168965,
      "learning_rate": 3.285021408369321e-07,
      "loss": 0.8207,
      "step": 955
    },
    {
      "epoch": 0.900225056264066,
      "grad_norm": 3.062713753398875,
      "learning_rate": 3.1820925161004326e-07,
      "loss": 0.9118,
      "step": 960
    },
    {
      "epoch": 0.900225056264066,
      "eval_loss": 0.9374144077301025,
      "eval_runtime": 20.3928,
      "eval_samples_per_second": 9.807,
      "eval_steps_per_second": 2.452,
      "step": 960
    },
    {
      "epoch": 0.904913728432108,
      "grad_norm": 2.9305276635268633,
      "learning_rate": 3.0791636238315446e-07,
      "loss": 0.8345,
      "step": 965
    },
    {
      "epoch": 0.9096024006001501,
      "grad_norm": 2.9508689565092046,
      "learning_rate": 2.9762347315626565e-07,
      "loss": 0.7999,
      "step": 970
    },
    {
      "epoch": 0.9142910727681921,
      "grad_norm": 3.9095096230897237,
      "learning_rate": 2.873305839293768e-07,
      "loss": 0.8812,
      "step": 975
    },
    {
      "epoch": 0.918979744936234,
      "grad_norm": 2.954610613967905,
      "learning_rate": 2.7703769470248794e-07,
      "loss": 0.9124,
      "step": 980
    },
    {
      "epoch": 0.923668417104276,
      "grad_norm": 2.3909984231204446,
      "learning_rate": 2.6674480547559913e-07,
      "loss": 0.8418,
      "step": 985
    },
    {
      "epoch": 0.9283570892723181,
      "grad_norm": 2.407545423074375,
      "learning_rate": 2.5645191624871027e-07,
      "loss": 0.9194,
      "step": 990
    },
    {
      "epoch": 0.9330457614403601,
      "grad_norm": 2.868767477778071,
      "learning_rate": 2.4615902702182147e-07,
      "loss": 0.9541,
      "step": 995
    },
    {
      "epoch": 0.9377344336084021,
      "grad_norm": 2.1293635803564066,
      "learning_rate": 2.3586613779493258e-07,
      "loss": 0.8087,
      "step": 1000
    },
    {
      "epoch": 0.9424231057764441,
      "grad_norm": 2.0557494272602264,
      "learning_rate": 2.255732485680438e-07,
      "loss": 0.9081,
      "step": 1005
    },
    {
      "epoch": 0.9471117779444861,
      "grad_norm": 2.1864930223853056,
      "learning_rate": 2.1528035934115495e-07,
      "loss": 0.859,
      "step": 1010
    },
    {
      "epoch": 0.9518004501125281,
      "grad_norm": 2.05122621453041,
      "learning_rate": 2.0498747011426614e-07,
      "loss": 0.8725,
      "step": 1015
    },
    {
      "epoch": 0.9564891222805701,
      "grad_norm": 3.2726273841892923,
      "learning_rate": 1.9469458088737728e-07,
      "loss": 0.7904,
      "step": 1020
    },
    {
      "epoch": 0.9611777944486122,
      "grad_norm": 2.259537550017741,
      "learning_rate": 1.8440169166048842e-07,
      "loss": 1.0669,
      "step": 1025
    },
    {
      "epoch": 0.9658664666166542,
      "grad_norm": 2.406310953068651,
      "learning_rate": 1.7410880243359964e-07,
      "loss": 0.8434,
      "step": 1030
    },
    {
      "epoch": 0.9705551387846961,
      "grad_norm": 1.9259292180053644,
      "learning_rate": 1.6381591320671076e-07,
      "loss": 1.0234,
      "step": 1035
    },
    {
      "epoch": 0.9752438109527382,
      "grad_norm": 2.0241322819114504,
      "learning_rate": 1.5352302397982195e-07,
      "loss": 0.9077,
      "step": 1040
    },
    {
      "epoch": 0.9752438109527382,
      "eval_loss": 0.9355975389480591,
      "eval_runtime": 20.3082,
      "eval_samples_per_second": 9.848,
      "eval_steps_per_second": 2.462,
      "step": 1040
    },
    {
      "epoch": 0.9799324831207802,
      "grad_norm": 1.9205593868577922,
      "learning_rate": 1.432301347529331e-07,
      "loss": 0.7646,
      "step": 1045
    },
    {
      "epoch": 0.9846211552888222,
      "grad_norm": 3.2012232274694834,
      "learning_rate": 1.329372455260443e-07,
      "loss": 0.8593,
      "step": 1050
    },
    {
      "epoch": 0.9893098274568642,
      "grad_norm": 2.3484948634638205,
      "learning_rate": 1.2264435629915543e-07,
      "loss": 1.0972,
      "step": 1055
    },
    {
      "epoch": 0.9939984996249063,
      "grad_norm": 2.7160625332606685,
      "learning_rate": 1.1235146707226661e-07,
      "loss": 0.7634,
      "step": 1060
    },
    {
      "epoch": 0.9986871717929482,
      "grad_norm": 2.3964216948732173,
      "learning_rate": 1.0205857784537777e-07,
      "loss": 0.9783,
      "step": 1065
    },
    {
      "epoch": 0.9996249062265566,
      "step": 1066,
      "total_flos": 1.5465201990107136e+17,
      "train_loss": 0.9640034305221815,
      "train_runtime": 13404.8173,
      "train_samples_per_second": 2.386,
      "train_steps_per_second": 0.08
    }
  ],
  "logging_steps": 5,
  "max_steps": 1066,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1066,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5465201990107136e+17,
  "train_batch_size": 3,
  "trial_name": null,
  "trial_params": null
}