{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.999972803176589,
  "eval_steps": 1000,
  "global_step": 9192,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005439364682205118,
      "grad_norm": 0.7295230031013489,
      "learning_rate": 9.946692776327242e-05,
      "loss": 0.2937,
      "step": 50
    },
    {
      "epoch": 0.010878729364410237,
      "grad_norm": 0.6098010540008545,
      "learning_rate": 9.892297650130549e-05,
      "loss": 0.0771,
      "step": 100
    },
    {
      "epoch": 0.016318094046615356,
      "grad_norm": 0.7312902808189392,
      "learning_rate": 9.837902523933856e-05,
      "loss": 0.0662,
      "step": 150
    },
    {
      "epoch": 0.021757458728820473,
      "grad_norm": 0.531105637550354,
      "learning_rate": 9.783507397737163e-05,
      "loss": 0.0591,
      "step": 200
    },
    {
      "epoch": 0.02719682341102559,
      "grad_norm": 0.3680162727832794,
      "learning_rate": 9.72911227154047e-05,
      "loss": 0.0593,
      "step": 250
    },
    {
      "epoch": 0.03263618809323071,
      "grad_norm": 0.5968620777130127,
      "learning_rate": 9.674717145343778e-05,
      "loss": 0.0534,
      "step": 300
    },
    {
      "epoch": 0.03807555277543583,
      "grad_norm": 0.31027188897132874,
      "learning_rate": 9.620322019147084e-05,
      "loss": 0.0471,
      "step": 350
    },
    {
      "epoch": 0.043514917457640946,
      "grad_norm": 0.3172386884689331,
      "learning_rate": 9.565926892950392e-05,
      "loss": 0.0486,
      "step": 400
    },
    {
      "epoch": 0.04895428213984607,
      "grad_norm": 0.3238416314125061,
      "learning_rate": 9.5115317667537e-05,
      "loss": 0.0483,
      "step": 450
    },
    {
      "epoch": 0.05439364682205118,
      "grad_norm": 0.5560753345489502,
      "learning_rate": 9.457136640557006e-05,
      "loss": 0.0392,
      "step": 500
    },
    {
      "epoch": 0.0598330115042563,
      "grad_norm": 0.4322672188282013,
      "learning_rate": 9.402741514360314e-05,
      "loss": 0.0395,
      "step": 550
    },
    {
      "epoch": 0.06527237618646142,
      "grad_norm": 0.29789817333221436,
      "learning_rate": 9.348346388163621e-05,
      "loss": 0.0396,
      "step": 600
    },
    {
      "epoch": 0.07071174086866654,
      "grad_norm": 0.2505125403404236,
      "learning_rate": 9.293951261966927e-05,
      "loss": 0.0312,
      "step": 650
    },
    {
      "epoch": 0.07615110555087166,
      "grad_norm": 0.4463261663913727,
      "learning_rate": 9.239556135770236e-05,
      "loss": 0.0382,
      "step": 700
    },
    {
      "epoch": 0.08159047023307678,
      "grad_norm": 0.30546480417251587,
      "learning_rate": 9.185161009573543e-05,
      "loss": 0.0373,
      "step": 750
    },
    {
      "epoch": 0.08702983491528189,
      "grad_norm": 0.1717098206281662,
      "learning_rate": 9.130765883376849e-05,
      "loss": 0.0405,
      "step": 800
    },
    {
      "epoch": 0.09246919959748702,
      "grad_norm": 0.2477688491344452,
      "learning_rate": 9.076370757180158e-05,
      "loss": 0.0364,
      "step": 850
    },
    {
      "epoch": 0.09790856427969213,
      "grad_norm": 0.4148896038532257,
      "learning_rate": 9.021975630983465e-05,
      "loss": 0.0327,
      "step": 900
    },
    {
      "epoch": 0.10334792896189725,
      "grad_norm": 0.36908024549484253,
      "learning_rate": 8.967580504786771e-05,
      "loss": 0.0307,
      "step": 950
    },
    {
      "epoch": 0.10878729364410236,
      "grad_norm": 0.612694263458252,
      "learning_rate": 8.913185378590078e-05,
      "loss": 0.0291,
      "step": 1000
    },
    {
      "epoch": 0.10878729364410236,
      "eval_loss": 0.031227048486471176,
      "eval_runtime": 623.3239,
      "eval_samples_per_second": 26.218,
      "eval_steps_per_second": 3.278,
      "step": 1000
    },
    {
      "epoch": 0.11422665832630749,
      "grad_norm": 0.13878417015075684,
      "learning_rate": 8.858790252393387e-05,
      "loss": 0.0314,
      "step": 1050
    },
    {
      "epoch": 0.1196660230085126,
      "grad_norm": 0.3287246823310852,
      "learning_rate": 8.804395126196693e-05,
      "loss": 0.0289,
      "step": 1100
    },
    {
      "epoch": 0.12510538769071772,
      "grad_norm": 0.2922157943248749,
      "learning_rate": 8.75e-05,
      "loss": 0.0327,
      "step": 1150
    },
    {
      "epoch": 0.13054475237292285,
      "grad_norm": 0.2656305432319641,
      "learning_rate": 8.695604873803309e-05,
      "loss": 0.0306,
      "step": 1200
    },
    {
      "epoch": 0.13598411705512797,
      "grad_norm": 0.1364014893770218,
      "learning_rate": 8.641209747606614e-05,
      "loss": 0.0296,
      "step": 1250
    },
    {
      "epoch": 0.14142348173733307,
      "grad_norm": 0.19203521311283112,
      "learning_rate": 8.586814621409922e-05,
      "loss": 0.0305,
      "step": 1300
    },
    {
      "epoch": 0.1468628464195382,
      "grad_norm": 0.1480610966682434,
      "learning_rate": 8.53241949521323e-05,
      "loss": 0.0217,
      "step": 1350
    },
    {
      "epoch": 0.15230221110174333,
      "grad_norm": 0.04526514559984207,
      "learning_rate": 8.478024369016536e-05,
      "loss": 0.0216,
      "step": 1400
    },
    {
      "epoch": 0.15774157578394843,
      "grad_norm": 0.7239785194396973,
      "learning_rate": 8.423629242819843e-05,
      "loss": 0.0293,
      "step": 1450
    },
    {
      "epoch": 0.16318094046615356,
      "grad_norm": 0.28628188371658325,
      "learning_rate": 8.369234116623151e-05,
      "loss": 0.0266,
      "step": 1500
    },
    {
      "epoch": 0.16862030514835868,
      "grad_norm": 0.3383677005767822,
      "learning_rate": 8.314838990426458e-05,
      "loss": 0.0279,
      "step": 1550
    },
    {
      "epoch": 0.17405966983056378,
      "grad_norm": 0.1695743054151535,
      "learning_rate": 8.260443864229765e-05,
      "loss": 0.0273,
      "step": 1600
    },
    {
      "epoch": 0.1794990345127689,
      "grad_norm": 0.08721613138914108,
      "learning_rate": 8.206048738033073e-05,
      "loss": 0.0281,
      "step": 1650
    },
    {
      "epoch": 0.18493839919497404,
      "grad_norm": 0.28628501296043396,
      "learning_rate": 8.15165361183638e-05,
      "loss": 0.0286,
      "step": 1700
    },
    {
      "epoch": 0.19037776387717914,
      "grad_norm": 0.11635430157184601,
      "learning_rate": 8.097258485639687e-05,
      "loss": 0.0295,
      "step": 1750
    },
    {
      "epoch": 0.19581712855938427,
      "grad_norm": 0.17274004220962524,
      "learning_rate": 8.042863359442994e-05,
      "loss": 0.0276,
      "step": 1800
    },
    {
      "epoch": 0.2012564932415894,
      "grad_norm": 0.3431914150714874,
      "learning_rate": 7.988468233246302e-05,
      "loss": 0.0236,
      "step": 1850
    },
    {
      "epoch": 0.2066958579237945,
      "grad_norm": 0.29347464442253113,
      "learning_rate": 7.934073107049609e-05,
      "loss": 0.0259,
      "step": 1900
    },
    {
      "epoch": 0.21213522260599962,
      "grad_norm": 0.2548673450946808,
      "learning_rate": 7.879677980852916e-05,
      "loss": 0.0247,
      "step": 1950
    },
    {
      "epoch": 0.21757458728820472,
      "grad_norm": 0.1950218230485916,
      "learning_rate": 7.825282854656223e-05,
      "loss": 0.0245,
      "step": 2000
    },
    {
      "epoch": 0.21757458728820472,
      "eval_loss": 0.024380268529057503,
      "eval_runtime": 612.7819,
      "eval_samples_per_second": 26.669,
      "eval_steps_per_second": 3.334,
      "step": 2000
    },
    {
      "epoch": 0.22301395197040985,
      "grad_norm": 0.14043785631656647,
      "learning_rate": 7.77088772845953e-05,
      "loss": 0.0258,
      "step": 2050
    },
    {
      "epoch": 0.22845331665261498,
      "grad_norm": 0.2565617263317108,
      "learning_rate": 7.716492602262838e-05,
      "loss": 0.026,
      "step": 2100
    },
    {
      "epoch": 0.23389268133482008,
      "grad_norm": 0.19237670302391052,
      "learning_rate": 7.662097476066145e-05,
      "loss": 0.0219,
      "step": 2150
    },
    {
      "epoch": 0.2393320460170252,
      "grad_norm": 0.17301031947135925,
      "learning_rate": 7.607702349869452e-05,
      "loss": 0.0273,
      "step": 2200
    },
    {
      "epoch": 0.24477141069923034,
      "grad_norm": 0.4380134344100952,
      "learning_rate": 7.55330722367276e-05,
      "loss": 0.0206,
      "step": 2250
    },
    {
      "epoch": 0.25021077538143544,
      "grad_norm": 0.5898825526237488,
      "learning_rate": 7.498912097476067e-05,
      "loss": 0.0221,
      "step": 2300
    },
    {
      "epoch": 0.25565014006364056,
      "grad_norm": 0.38086146116256714,
      "learning_rate": 7.444516971279374e-05,
      "loss": 0.0219,
      "step": 2350
    },
    {
      "epoch": 0.2610895047458457,
      "grad_norm": 0.4513295888900757,
      "learning_rate": 7.390121845082681e-05,
      "loss": 0.0232,
      "step": 2400
    },
    {
      "epoch": 0.2665288694280508,
      "grad_norm": 0.09053909033536911,
      "learning_rate": 7.335726718885987e-05,
      "loss": 0.0221,
      "step": 2450
    },
    {
      "epoch": 0.27196823411025595,
      "grad_norm": 0.20339776575565338,
      "learning_rate": 7.281331592689296e-05,
      "loss": 0.0214,
      "step": 2500
    },
    {
      "epoch": 0.277407598792461,
      "grad_norm": 0.09539163112640381,
      "learning_rate": 7.226936466492603e-05,
      "loss": 0.0186,
      "step": 2550
    },
    {
      "epoch": 0.28284696347466615,
      "grad_norm": 0.8186866044998169,
      "learning_rate": 7.172541340295909e-05,
      "loss": 0.0216,
      "step": 2600
    },
    {
      "epoch": 0.2882863281568713,
      "grad_norm": 0.2770097553730011,
      "learning_rate": 7.118146214099218e-05,
      "loss": 0.0203,
      "step": 2650
    },
    {
      "epoch": 0.2937256928390764,
      "grad_norm": 0.2176240086555481,
      "learning_rate": 7.063751087902525e-05,
      "loss": 0.0284,
      "step": 2700
    },
    {
      "epoch": 0.29916505752128153,
      "grad_norm": 0.3172820210456848,
      "learning_rate": 7.009355961705831e-05,
      "loss": 0.0196,
      "step": 2750
    },
    {
      "epoch": 0.30460442220348666,
      "grad_norm": 0.5219048261642456,
      "learning_rate": 6.95496083550914e-05,
      "loss": 0.0209,
      "step": 2800
    },
    {
      "epoch": 0.31004378688569173,
      "grad_norm": 0.16620703041553497,
      "learning_rate": 6.900565709312445e-05,
      "loss": 0.0253,
      "step": 2850
    },
    {
      "epoch": 0.31548315156789686,
      "grad_norm": 0.33126798272132874,
      "learning_rate": 6.846170583115753e-05,
      "loss": 0.0217,
      "step": 2900
    },
    {
      "epoch": 0.320922516250102,
      "grad_norm": 0.07635556906461716,
      "learning_rate": 6.791775456919061e-05,
      "loss": 0.0232,
      "step": 2950
    },
    {
      "epoch": 0.3263618809323071,
      "grad_norm": 0.3889711797237396,
      "learning_rate": 6.737380330722367e-05,
      "loss": 0.0212,
      "step": 3000
    },
    {
      "epoch": 0.3263618809323071,
      "eval_loss": 0.021349932998418808,
      "eval_runtime": 613.1791,
      "eval_samples_per_second": 26.651,
      "eval_steps_per_second": 3.332,
      "step": 3000
    },
    {
      "epoch": 0.33180124561451224,
      "grad_norm": 0.3016066551208496,
      "learning_rate": 6.682985204525674e-05,
      "loss": 0.0211,
      "step": 3050
    },
    {
      "epoch": 0.33724061029671737,
      "grad_norm": 0.07768701761960983,
      "learning_rate": 6.628590078328982e-05,
      "loss": 0.0178,
      "step": 3100
    },
    {
      "epoch": 0.34267997497892244,
      "grad_norm": 0.39988207817077637,
      "learning_rate": 6.574194952132289e-05,
      "loss": 0.0183,
      "step": 3150
    },
    {
      "epoch": 0.34811933966112757,
      "grad_norm": 0.17848381400108337,
      "learning_rate": 6.519799825935596e-05,
      "loss": 0.0177,
      "step": 3200
    },
    {
      "epoch": 0.3535587043433327,
      "grad_norm": 0.3475474417209625,
      "learning_rate": 6.465404699738903e-05,
      "loss": 0.0202,
      "step": 3250
    },
    {
      "epoch": 0.3589980690255378,
      "grad_norm": 0.13056176900863647,
      "learning_rate": 6.411009573542211e-05,
      "loss": 0.0191,
      "step": 3300
    },
    {
      "epoch": 0.36443743370774295,
      "grad_norm": 0.1627858579158783,
      "learning_rate": 6.356614447345518e-05,
      "loss": 0.0189,
      "step": 3350
    },
    {
      "epoch": 0.3698767983899481,
      "grad_norm": 0.3630661070346832,
      "learning_rate": 6.302219321148825e-05,
      "loss": 0.0223,
      "step": 3400
    },
    {
      "epoch": 0.37531616307215315,
      "grad_norm": 0.18658381700515747,
      "learning_rate": 6.247824194952132e-05,
      "loss": 0.0208,
      "step": 3450
    },
    {
      "epoch": 0.3807555277543583,
      "grad_norm": 0.255504310131073,
      "learning_rate": 6.19342906875544e-05,
      "loss": 0.0177,
      "step": 3500
    },
    {
      "epoch": 0.3861948924365634,
      "grad_norm": 0.28671759366989136,
      "learning_rate": 6.139033942558747e-05,
      "loss": 0.0204,
      "step": 3550
    },
    {
      "epoch": 0.39163425711876854,
      "grad_norm": 0.10439465939998627,
      "learning_rate": 6.0846388163620536e-05,
      "loss": 0.0148,
      "step": 3600
    },
    {
      "epoch": 0.39707362180097366,
      "grad_norm": 0.1532258689403534,
      "learning_rate": 6.0302436901653615e-05,
      "loss": 0.0173,
      "step": 3650
    },
    {
      "epoch": 0.4025129864831788,
      "grad_norm": 0.45614758133888245,
      "learning_rate": 5.975848563968669e-05,
      "loss": 0.015,
      "step": 3700
    },
    {
      "epoch": 0.40795235116538386,
      "grad_norm": 0.19495117664337158,
      "learning_rate": 5.9214534377719754e-05,
      "loss": 0.0186,
      "step": 3750
    },
    {
      "epoch": 0.413391715847589,
      "grad_norm": 0.1256154626607895,
      "learning_rate": 5.867058311575283e-05,
      "loss": 0.0156,
      "step": 3800
    },
    {
      "epoch": 0.4188310805297941,
      "grad_norm": 0.10156747698783875,
      "learning_rate": 5.8126631853785905e-05,
      "loss": 0.0188,
      "step": 3850
    },
    {
      "epoch": 0.42427044521199925,
      "grad_norm": 0.2215280681848526,
      "learning_rate": 5.758268059181897e-05,
      "loss": 0.0191,
      "step": 3900
    },
    {
      "epoch": 0.4297098098942044,
      "grad_norm": 0.5749198198318481,
      "learning_rate": 5.703872932985205e-05,
      "loss": 0.0161,
      "step": 3950
    },
    {
      "epoch": 0.43514917457640945,
      "grad_norm": 0.2301347702741623,
      "learning_rate": 5.649477806788512e-05,
      "loss": 0.018,
      "step": 4000
    },
    {
      "epoch": 0.43514917457640945,
      "eval_loss": 0.017653847113251686,
      "eval_runtime": 611.9222,
      "eval_samples_per_second": 26.706,
      "eval_steps_per_second": 3.339,
      "step": 4000
    },
    {
      "epoch": 0.4405885392586146,
      "grad_norm": 0.18594112992286682,
      "learning_rate": 5.595082680591819e-05,
      "loss": 0.0177,
      "step": 4050
    },
    {
      "epoch": 0.4460279039408197,
      "grad_norm": 0.18527820706367493,
      "learning_rate": 5.540687554395126e-05,
      "loss": 0.0167,
      "step": 4100
    },
    {
      "epoch": 0.45146726862302483,
      "grad_norm": 0.6579691171646118,
      "learning_rate": 5.486292428198434e-05,
      "loss": 0.0162,
      "step": 4150
    },
    {
      "epoch": 0.45690663330522996,
      "grad_norm": 0.03042120486497879,
      "learning_rate": 5.431897302001741e-05,
      "loss": 0.0144,
      "step": 4200
    },
    {
      "epoch": 0.4623459979874351,
      "grad_norm": 0.2600632309913635,
      "learning_rate": 5.377502175805048e-05,
      "loss": 0.0154,
      "step": 4250
    },
    {
      "epoch": 0.46778536266964016,
      "grad_norm": 0.20093311369419098,
      "learning_rate": 5.323107049608356e-05,
      "loss": 0.0168,
      "step": 4300
    },
    {
      "epoch": 0.4732247273518453,
      "grad_norm": 0.307822048664093,
      "learning_rate": 5.2687119234116625e-05,
      "loss": 0.0148,
      "step": 4350
    },
    {
      "epoch": 0.4786640920340504,
      "grad_norm": 0.23011909425258636,
      "learning_rate": 5.21431679721497e-05,
      "loss": 0.0153,
      "step": 4400
    },
    {
      "epoch": 0.48410345671625554,
      "grad_norm": 0.10574093461036682,
      "learning_rate": 5.1599216710182777e-05,
      "loss": 0.0171,
      "step": 4450
    },
    {
      "epoch": 0.48954282139846067,
      "grad_norm": 0.2588224709033966,
      "learning_rate": 5.1055265448215836e-05,
      "loss": 0.0181,
      "step": 4500
    },
    {
      "epoch": 0.4949821860806658,
      "grad_norm": 0.10799703747034073,
      "learning_rate": 5.0511314186248915e-05,
      "loss": 0.0169,
      "step": 4550
    },
    {
      "epoch": 0.5004215507628709,
      "grad_norm": 0.10647980868816376,
      "learning_rate": 4.996736292428199e-05,
      "loss": 0.0134,
      "step": 4600
    },
    {
      "epoch": 0.505860915445076,
      "grad_norm": 0.026128219440579414,
      "learning_rate": 4.942341166231506e-05,
      "loss": 0.0164,
      "step": 4650
    },
    {
      "epoch": 0.5113002801272811,
      "grad_norm": 0.22313055396080017,
      "learning_rate": 4.887946040034813e-05,
      "loss": 0.0139,
      "step": 4700
    },
    {
      "epoch": 0.5167396448094862,
      "grad_norm": 0.41442468762397766,
      "learning_rate": 4.8335509138381205e-05,
      "loss": 0.0135,
      "step": 4750
    },
    {
      "epoch": 0.5221790094916914,
      "grad_norm": 0.13141965866088867,
      "learning_rate": 4.779155787641428e-05,
      "loss": 0.0138,
      "step": 4800
    },
    {
      "epoch": 0.5276183741738965,
      "grad_norm": 0.23391123116016388,
      "learning_rate": 4.724760661444735e-05,
      "loss": 0.0148,
      "step": 4850
    },
    {
      "epoch": 0.5330577388561016,
      "grad_norm": 0.2290557622909546,
      "learning_rate": 4.6703655352480416e-05,
      "loss": 0.014,
      "step": 4900
    },
    {
      "epoch": 0.5384971035383067,
      "grad_norm": 0.263883113861084,
      "learning_rate": 4.6159704090513496e-05,
      "loss": 0.0129,
      "step": 4950
    },
    {
      "epoch": 0.5439364682205119,
      "grad_norm": 0.4367043673992157,
      "learning_rate": 4.561575282854657e-05,
      "loss": 0.0138,
      "step": 5000
    },
    {
      "epoch": 0.5439364682205119,
      "eval_loss": 0.015788141638040543,
      "eval_runtime": 612.4211,
      "eval_samples_per_second": 26.684,
      "eval_steps_per_second": 3.336,
      "step": 5000
    },
    {
      "epoch": 0.549375832902717,
      "grad_norm": 0.17054913938045502,
      "learning_rate": 4.5071801566579634e-05,
      "loss": 0.0122,
      "step": 5050
    },
    {
      "epoch": 0.554815197584922,
      "grad_norm": 0.47613638639450073,
      "learning_rate": 4.4527850304612713e-05,
      "loss": 0.0148,
      "step": 5100
    },
    {
      "epoch": 0.5602545622671272,
      "grad_norm": 0.1685837209224701,
      "learning_rate": 4.398389904264578e-05,
      "loss": 0.0166,
      "step": 5150
    },
    {
      "epoch": 0.5656939269493323,
      "grad_norm": 0.23199672996997833,
      "learning_rate": 4.343994778067885e-05,
      "loss": 0.0139,
      "step": 5200
    },
    {
      "epoch": 0.5711332916315375,
      "grad_norm": 0.5401563048362732,
      "learning_rate": 4.2895996518711924e-05,
      "loss": 0.0134,
      "step": 5250
    },
    {
      "epoch": 0.5765726563137425,
      "grad_norm": 0.1670864224433899,
      "learning_rate": 4.2352045256745e-05,
      "loss": 0.0143,
      "step": 5300
    },
    {
      "epoch": 0.5820120209959476,
      "grad_norm": 0.08487512916326523,
      "learning_rate": 4.180809399477807e-05,
      "loss": 0.0152,
      "step": 5350
    },
    {
      "epoch": 0.5874513856781528,
      "grad_norm": 0.07271425426006317,
      "learning_rate": 4.126414273281114e-05,
      "loss": 0.0166,
      "step": 5400
    },
    {
      "epoch": 0.5928907503603579,
      "grad_norm": 0.27096590399742126,
      "learning_rate": 4.0720191470844215e-05,
      "loss": 0.0134,
      "step": 5450
    },
    {
      "epoch": 0.5983301150425631,
      "grad_norm": 0.1035747304558754,
      "learning_rate": 4.017624020887729e-05,
      "loss": 0.0144,
      "step": 5500
    },
    {
      "epoch": 0.6037694797247681,
      "grad_norm": 0.34544578194618225,
      "learning_rate": 3.963228894691035e-05,
      "loss": 0.0138,
      "step": 5550
    },
    {
      "epoch": 0.6092088444069733,
      "grad_norm": 0.4660258889198303,
      "learning_rate": 3.908833768494343e-05,
      "loss": 0.0166,
      "step": 5600
    },
    {
      "epoch": 0.6146482090891784,
      "grad_norm": 0.12394747883081436,
      "learning_rate": 3.8544386422976505e-05,
      "loss": 0.0124,
      "step": 5650
    },
    {
      "epoch": 0.6200875737713835,
      "grad_norm": 0.13275495171546936,
      "learning_rate": 3.800043516100957e-05,
      "loss": 0.0144,
      "step": 5700
    },
    {
      "epoch": 0.6255269384535886,
      "grad_norm": 0.23382355272769928,
      "learning_rate": 3.745648389904265e-05,
      "loss": 0.0163,
      "step": 5750
    },
    {
      "epoch": 0.6309663031357937,
      "grad_norm": 0.10067889094352722,
      "learning_rate": 3.691253263707572e-05,
      "loss": 0.0166,
      "step": 5800
    },
    {
      "epoch": 0.6364056678179989,
      "grad_norm": 0.09269619733095169,
      "learning_rate": 3.636858137510879e-05,
      "loss": 0.0155,
      "step": 5850
    },
    {
      "epoch": 0.641845032500204,
      "grad_norm": 0.22883062064647675,
      "learning_rate": 3.582463011314187e-05,
      "loss": 0.0119,
      "step": 5900
    },
    {
      "epoch": 0.647284397182409,
      "grad_norm": 0.3405645787715912,
      "learning_rate": 3.5280678851174934e-05,
      "loss": 0.0134,
      "step": 5950
    },
    {
      "epoch": 0.6527237618646142,
      "grad_norm": 0.10464873909950256,
      "learning_rate": 3.4736727589208007e-05,
      "loss": 0.0115,
      "step": 6000
    },
    {
      "epoch": 0.6527237618646142,
      "eval_loss": 0.013636166229844093,
      "eval_runtime": 612.7881,
      "eval_samples_per_second": 26.668,
      "eval_steps_per_second": 3.334,
      "step": 6000
    },
    {
      "epoch": 0.6581631265468193,
      "grad_norm": 0.15743671357631683,
      "learning_rate": 3.4192776327241086e-05,
      "loss": 0.0127,
      "step": 6050
    },
    {
      "epoch": 0.6636024912290245,
      "grad_norm": 0.28343307971954346,
      "learning_rate": 3.364882506527415e-05,
      "loss": 0.013,
      "step": 6100
    },
    {
      "epoch": 0.6690418559112296,
      "grad_norm": 0.0802890881896019,
      "learning_rate": 3.3104873803307224e-05,
      "loss": 0.0156,
      "step": 6150
    },
    {
      "epoch": 0.6744812205934347,
      "grad_norm": 0.2691664397716522,
      "learning_rate": 3.25609225413403e-05,
      "loss": 0.0131,
      "step": 6200
    },
    {
      "epoch": 0.6799205852756398,
      "grad_norm": 0.08483448624610901,
      "learning_rate": 3.201697127937337e-05,
      "loss": 0.0127,
      "step": 6250
    },
    {
      "epoch": 0.6853599499578449,
      "grad_norm": 0.2310999631881714,
      "learning_rate": 3.147302001740644e-05,
      "loss": 0.01,
      "step": 6300
    },
    {
      "epoch": 0.6907993146400501,
      "grad_norm": 0.10749073326587677,
      "learning_rate": 3.0929068755439515e-05,
      "loss": 0.0128,
      "step": 6350
    },
    {
      "epoch": 0.6962386793222551,
      "grad_norm": 0.14852339029312134,
      "learning_rate": 3.0385117493472587e-05,
      "loss": 0.0097,
      "step": 6400
    },
    {
      "epoch": 0.7016780440044603,
      "grad_norm": 0.09708331525325775,
      "learning_rate": 2.9841166231505656e-05,
      "loss": 0.0137,
      "step": 6450
    },
    {
      "epoch": 0.7071174086866654,
      "grad_norm": 0.13305231928825378,
      "learning_rate": 2.9297214969538732e-05,
      "loss": 0.0106,
      "step": 6500
    },
    {
      "epoch": 0.7125567733688705,
      "grad_norm": 0.2321113497018814,
      "learning_rate": 2.8753263707571805e-05,
      "loss": 0.0161,
      "step": 6550
    },
    {
      "epoch": 0.7179961380510756,
      "grad_norm": 0.1432623565196991,
      "learning_rate": 2.8209312445604874e-05,
      "loss": 0.0096,
      "step": 6600
    },
    {
      "epoch": 0.7234355027332807,
      "grad_norm": 0.1964827924966812,
      "learning_rate": 2.766536118363795e-05,
      "loss": 0.0174,
      "step": 6650
    },
    {
      "epoch": 0.7288748674154859,
      "grad_norm": 0.21941740810871124,
      "learning_rate": 2.712140992167102e-05,
      "loss": 0.0125,
      "step": 6700
    },
    {
      "epoch": 0.734314232097691,
      "grad_norm": 0.06487424671649933,
      "learning_rate": 2.6577458659704092e-05,
      "loss": 0.0143,
      "step": 6750
    },
    {
      "epoch": 0.7397535967798962,
      "grad_norm": 0.08458438515663147,
      "learning_rate": 2.603350739773716e-05,
      "loss": 0.0122,
      "step": 6800
    },
    {
      "epoch": 0.7451929614621012,
      "grad_norm": 0.21706067025661469,
      "learning_rate": 2.5489556135770237e-05,
      "loss": 0.0136,
      "step": 6850
    },
    {
      "epoch": 0.7506323261443063,
      "grad_norm": 0.04576512426137924,
      "learning_rate": 2.4945604873803306e-05,
      "loss": 0.0127,
      "step": 6900
    },
    {
      "epoch": 0.7560716908265115,
      "grad_norm": 0.05280361324548721,
      "learning_rate": 2.4401653611836382e-05,
      "loss": 0.0107,
      "step": 6950
    },
    {
      "epoch": 0.7615110555087166,
      "grad_norm": 0.5558903217315674,
      "learning_rate": 2.3857702349869455e-05,
      "loss": 0.0088,
      "step": 7000
    },
    {
      "epoch": 0.7615110555087166,
      "eval_loss": 0.01208407897502184,
      "eval_runtime": 610.9461,
      "eval_samples_per_second": 26.749,
      "eval_steps_per_second": 3.344,
      "step": 7000
    },
    {
      "epoch": 0.7669504201909217,
      "grad_norm": 0.470074325799942,
      "learning_rate": 2.3313751087902524e-05,
      "loss": 0.0135,
      "step": 7050
    },
    {
      "epoch": 0.7723897848731268,
      "grad_norm": 0.2360675036907196,
      "learning_rate": 2.2769799825935597e-05,
      "loss": 0.0133,
      "step": 7100
    },
    {
      "epoch": 0.7778291495553319,
      "grad_norm": 0.11457215994596481,
      "learning_rate": 2.222584856396867e-05,
      "loss": 0.011,
      "step": 7150
    },
    {
      "epoch": 0.7832685142375371,
      "grad_norm": 0.010224751196801662,
      "learning_rate": 2.1681897302001742e-05,
      "loss": 0.0148,
      "step": 7200
    },
    {
      "epoch": 0.7887078789197421,
      "grad_norm": 0.2885963022708893,
      "learning_rate": 2.1137946040034815e-05,
      "loss": 0.0132,
      "step": 7250
    },
    {
      "epoch": 0.7941472436019473,
      "grad_norm": 0.3628193438053131,
      "learning_rate": 2.0593994778067884e-05,
      "loss": 0.0105,
      "step": 7300
    },
    {
      "epoch": 0.7995866082841524,
      "grad_norm": 0.07265301048755646,
      "learning_rate": 2.005004351610096e-05,
      "loss": 0.0104,
      "step": 7350
    },
    {
      "epoch": 0.8050259729663576,
      "grad_norm": 0.13889417052268982,
      "learning_rate": 1.9506092254134032e-05,
      "loss": 0.0129,
      "step": 7400
    },
    {
      "epoch": 0.8104653376485627,
      "grad_norm": 0.06533674895763397,
      "learning_rate": 1.89621409921671e-05,
      "loss": 0.0097,
      "step": 7450
    },
    {
      "epoch": 0.8159047023307677,
      "grad_norm": 0.23986752331256866,
      "learning_rate": 1.8418189730200174e-05,
      "loss": 0.0119,
      "step": 7500
    },
    {
      "epoch": 0.8213440670129729,
      "grad_norm": 0.09028589725494385,
      "learning_rate": 1.7874238468233247e-05,
      "loss": 0.0093,
      "step": 7550
    },
    {
      "epoch": 0.826783431695178,
      "grad_norm": 0.11503873020410538,
      "learning_rate": 1.733028720626632e-05,
      "loss": 0.0091,
      "step": 7600
    },
    {
      "epoch": 0.8322227963773832,
      "grad_norm": 0.5392634868621826,
      "learning_rate": 1.6786335944299392e-05,
      "loss": 0.0118,
      "step": 7650
    },
    {
      "epoch": 0.8376621610595882,
      "grad_norm": 0.14633004367351532,
      "learning_rate": 1.6242384682332464e-05,
      "loss": 0.0112,
      "step": 7700
    },
    {
      "epoch": 0.8431015257417933,
      "grad_norm": 0.03690435364842415,
      "learning_rate": 1.5698433420365534e-05,
      "loss": 0.0117,
      "step": 7750
    },
    {
      "epoch": 0.8485408904239985,
      "grad_norm": 0.1749696284532547,
      "learning_rate": 1.5154482158398608e-05,
      "loss": 0.0108,
      "step": 7800
    },
    {
      "epoch": 0.8539802551062036,
      "grad_norm": 0.1117832213640213,
      "learning_rate": 1.4610530896431682e-05,
      "loss": 0.0142,
      "step": 7850
    },
    {
      "epoch": 0.8594196197884088,
      "grad_norm": 0.06253615021705627,
      "learning_rate": 1.4066579634464751e-05,
      "loss": 0.0101,
      "step": 7900
    },
    {
      "epoch": 0.8648589844706138,
      "grad_norm": 0.11594853550195694,
      "learning_rate": 1.3522628372497826e-05,
      "loss": 0.0102,
      "step": 7950
    },
    {
      "epoch": 0.8702983491528189,
      "grad_norm": 0.2881285846233368,
      "learning_rate": 1.2978677110530895e-05,
      "loss": 0.0096,
      "step": 8000
    },
    {
      "epoch": 0.8702983491528189,
      "eval_loss": 0.010398774407804012,
      "eval_runtime": 612.7356,
      "eval_samples_per_second": 26.671,
      "eval_steps_per_second": 3.334,
      "step": 8000
    },
    {
      "epoch": 0.8757377138350241,
      "grad_norm": 0.10731537640094757,
      "learning_rate": 1.243472584856397e-05,
      "loss": 0.0102,
      "step": 8050
    },
    {
      "epoch": 0.8811770785172292,
      "grad_norm": 0.08623083680868149,
      "learning_rate": 1.1890774586597042e-05,
      "loss": 0.0095,
      "step": 8100
    },
    {
      "epoch": 0.8866164431994343,
      "grad_norm": 0.16180342435836792,
      "learning_rate": 1.1346823324630114e-05,
      "loss": 0.0094,
      "step": 8150
    },
    {
      "epoch": 0.8920558078816394,
      "grad_norm": 0.09304390847682953,
      "learning_rate": 1.0802872062663185e-05,
      "loss": 0.0105,
      "step": 8200
    },
    {
      "epoch": 0.8974951725638446,
      "grad_norm": 0.08745424449443817,
      "learning_rate": 1.0258920800696258e-05,
      "loss": 0.0077,
      "step": 8250
    },
    {
      "epoch": 0.9029345372460497,
      "grad_norm": 0.18484659492969513,
      "learning_rate": 9.71496953872933e-06,
      "loss": 0.0116,
      "step": 8300
    },
    {
      "epoch": 0.9083739019282547,
      "grad_norm": 0.18615852296352386,
      "learning_rate": 9.171018276762403e-06,
      "loss": 0.0087,
      "step": 8350
    },
    {
      "epoch": 0.9138132666104599,
      "grad_norm": 0.038191672414541245,
      "learning_rate": 8.627067014795474e-06,
      "loss": 0.0091,
      "step": 8400
    },
    {
      "epoch": 0.919252631292665,
      "grad_norm": 0.049192268401384354,
      "learning_rate": 8.083115752828548e-06,
      "loss": 0.0113,
      "step": 8450
    },
    {
      "epoch": 0.9246919959748702,
      "grad_norm": 0.12032605707645416,
      "learning_rate": 7.539164490861619e-06,
      "loss": 0.0116,
      "step": 8500
    },
    {
      "epoch": 0.9301313606570752,
      "grad_norm": 0.2927582561969757,
      "learning_rate": 7.006092254134029e-06,
      "loss": 0.0114,
      "step": 8550
    },
    {
      "epoch": 0.9355707253392803,
      "grad_norm": 0.04057031497359276,
      "learning_rate": 6.462140992167103e-06,
      "loss": 0.0093,
      "step": 8600
    },
    {
      "epoch": 0.9410100900214855,
      "grad_norm": 0.23389148712158203,
      "learning_rate": 5.918189730200174e-06,
      "loss": 0.0123,
      "step": 8650
    },
    {
      "epoch": 0.9464494547036906,
      "grad_norm": 0.14039556682109833,
      "learning_rate": 5.374238468233247e-06,
      "loss": 0.0088,
      "step": 8700
    },
    {
      "epoch": 0.9518888193858958,
      "grad_norm": 0.10834332555532455,
      "learning_rate": 4.830287206266319e-06,
      "loss": 0.0126,
      "step": 8750
    },
    {
      "epoch": 0.9573281840681008,
      "grad_norm": 0.20089313387870789,
      "learning_rate": 4.286335944299391e-06,
      "loss": 0.007,
      "step": 8800
    },
    {
      "epoch": 0.962767548750306,
      "grad_norm": 0.10719793289899826,
      "learning_rate": 3.742384682332463e-06,
      "loss": 0.0073,
      "step": 8850
    },
    {
      "epoch": 0.9682069134325111,
      "grad_norm": 0.051126107573509216,
      "learning_rate": 3.1984334203655352e-06,
      "loss": 0.0098,
      "step": 8900
    },
    {
      "epoch": 0.9736462781147162,
      "grad_norm": 0.2430637627840042,
      "learning_rate": 2.6544821583986074e-06,
      "loss": 0.0078,
      "step": 8950
    },
    {
      "epoch": 0.9790856427969213,
      "grad_norm": 0.06034141406416893,
      "learning_rate": 2.1105308964316795e-06,
      "loss": 0.0087,
      "step": 9000
    },
    {
      "epoch": 0.9790856427969213,
      "eval_loss": 0.009567675180733204,
      "eval_runtime": 610.7757,
      "eval_samples_per_second": 26.756,
      "eval_steps_per_second": 3.345,
      "step": 9000
    },
    {
      "epoch": 0.9845250074791264,
      "grad_norm": 0.23722563683986664,
      "learning_rate": 1.566579634464752e-06,
      "loss": 0.0094,
      "step": 9050
    },
    {
      "epoch": 0.9899643721613316,
      "grad_norm": 0.1395604908466339,
      "learning_rate": 1.0226283724978243e-06,
      "loss": 0.01,
      "step": 9100
    },
    {
      "epoch": 0.9954037368435367,
      "grad_norm": 0.12211757898330688,
      "learning_rate": 4.786771105308965e-07,
      "loss": 0.0091,
      "step": 9150
    }
  ],
  "logging_steps": 50,
  "max_steps": 9192,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.365723691894702e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}