{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9836065573770492,
  "eval_steps": 1000,
  "global_step": 18000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00546448087431694,
      "grad_norm": 0.33557644486427307,
      "learning_rate": 0.00014918852459016392,
      "loss": 3.2997506713867186,
      "step": 100
    },
    {
      "epoch": 0.01092896174863388,
      "grad_norm": 0.3691432774066925,
      "learning_rate": 0.00014836885245901637,
      "loss": 3.297074279785156,
      "step": 200
    },
    {
      "epoch": 0.01639344262295082,
      "grad_norm": 0.33957722783088684,
      "learning_rate": 0.00014754918032786884,
      "loss": 3.284024353027344,
      "step": 300
    },
    {
      "epoch": 0.02185792349726776,
      "grad_norm": 0.324457049369812,
      "learning_rate": 0.0001467295081967213,
      "loss": 3.2819815063476563,
      "step": 400
    },
    {
      "epoch": 0.0273224043715847,
      "grad_norm": 0.3473037779331207,
      "learning_rate": 0.00014590983606557376,
      "loss": 3.273143005371094,
      "step": 500
    },
    {
      "epoch": 0.03278688524590164,
      "grad_norm": 0.33360302448272705,
      "learning_rate": 0.0001450901639344262,
      "loss": 3.2673272705078125,
      "step": 600
    },
    {
      "epoch": 0.03825136612021858,
      "grad_norm": 0.35968998074531555,
      "learning_rate": 0.00014427049180327868,
      "loss": 3.2718386840820313,
      "step": 700
    },
    {
      "epoch": 0.04371584699453552,
      "grad_norm": 0.34233754873275757,
      "learning_rate": 0.00014345081967213113,
      "loss": 3.2761669921875,
      "step": 800
    },
    {
      "epoch": 0.04918032786885246,
      "grad_norm": 0.34049418568611145,
      "learning_rate": 0.00014263114754098358,
      "loss": 3.2359161376953125,
      "step": 900
    },
    {
      "epoch": 0.0546448087431694,
      "grad_norm": 0.3437342047691345,
      "learning_rate": 0.00014181147540983605,
      "loss": 3.2309716796875,
      "step": 1000
    },
    {
      "epoch": 0.0546448087431694,
      "eval_loss": 3.141683578491211,
      "eval_runtime": 167.5376,
      "eval_samples_per_second": 51.851,
      "eval_steps_per_second": 1.624,
      "step": 1000
    },
    {
      "epoch": 0.060109289617486336,
      "grad_norm": 0.3368486166000366,
      "learning_rate": 0.00014099180327868853,
      "loss": 3.2491357421875,
      "step": 1100
    },
    {
      "epoch": 0.06557377049180328,
      "grad_norm": 0.3406209945678711,
      "learning_rate": 0.00014017213114754097,
      "loss": 3.241882629394531,
      "step": 1200
    },
    {
      "epoch": 0.07103825136612021,
      "grad_norm": 0.36890581250190735,
      "learning_rate": 0.00013935245901639345,
      "loss": 3.21847412109375,
      "step": 1300
    },
    {
      "epoch": 0.07650273224043716,
      "grad_norm": 0.32925161719322205,
      "learning_rate": 0.0001385327868852459,
      "loss": 3.247527770996094,
      "step": 1400
    },
    {
      "epoch": 0.08196721311475409,
      "grad_norm": 0.3606502115726471,
      "learning_rate": 0.00013771311475409834,
      "loss": 3.2279705810546875,
      "step": 1500
    },
    {
      "epoch": 0.08743169398907104,
      "grad_norm": 0.35976818203926086,
      "learning_rate": 0.00013689344262295082,
      "loss": 3.2205975341796873,
      "step": 1600
    },
    {
      "epoch": 0.09289617486338798,
      "grad_norm": 0.35467588901519775,
      "learning_rate": 0.00013607377049180326,
      "loss": 3.23948486328125,
      "step": 1700
    },
    {
      "epoch": 0.09836065573770492,
      "grad_norm": 0.33707618713378906,
      "learning_rate": 0.00013525409836065574,
      "loss": 3.22716796875,
      "step": 1800
    },
    {
      "epoch": 0.10382513661202186,
      "grad_norm": 0.34896889328956604,
      "learning_rate": 0.00013443442622950818,
      "loss": 3.254466247558594,
      "step": 1900
    },
    {
      "epoch": 0.1092896174863388,
      "grad_norm": 0.33358848094940186,
      "learning_rate": 0.00013361475409836063,
      "loss": 3.235157470703125,
      "step": 2000
    },
    {
      "epoch": 0.1092896174863388,
      "eval_loss": 3.129152297973633,
      "eval_runtime": 166.3667,
      "eval_samples_per_second": 52.216,
      "eval_steps_per_second": 1.635,
      "step": 2000
    },
    {
      "epoch": 0.11475409836065574,
      "grad_norm": 0.36108165979385376,
      "learning_rate": 0.0001327950819672131,
      "loss": 3.2411184692382813,
      "step": 2100
    },
    {
      "epoch": 0.12021857923497267,
      "grad_norm": 0.33732327818870544,
      "learning_rate": 0.00013197540983606555,
      "loss": 3.2485076904296877,
      "step": 2200
    },
    {
      "epoch": 0.12568306010928962,
      "grad_norm": 0.35349759459495544,
      "learning_rate": 0.00013115573770491803,
      "loss": 3.2446453857421873,
      "step": 2300
    },
    {
      "epoch": 0.13114754098360656,
      "grad_norm": 0.33498039841651917,
      "learning_rate": 0.00013033606557377047,
      "loss": 3.2644442749023437,
      "step": 2400
    },
    {
      "epoch": 0.1366120218579235,
      "grad_norm": 0.3694196939468384,
      "learning_rate": 0.00012951639344262292,
      "loss": 3.235172119140625,
      "step": 2500
    },
    {
      "epoch": 0.14207650273224043,
      "grad_norm": 0.34429556131362915,
      "learning_rate": 0.0001286967213114754,
      "loss": 3.2346224975585938,
      "step": 2600
    },
    {
      "epoch": 0.14754098360655737,
      "grad_norm": 0.3450092673301697,
      "learning_rate": 0.00012787704918032784,
      "loss": 3.2645361328125,
      "step": 2700
    },
    {
      "epoch": 0.15300546448087432,
      "grad_norm": 0.3353455662727356,
      "learning_rate": 0.00012705737704918032,
      "loss": 3.259942626953125,
      "step": 2800
    },
    {
      "epoch": 0.15846994535519127,
      "grad_norm": 0.3381529450416565,
      "learning_rate": 0.0001262377049180328,
      "loss": 3.219759826660156,
      "step": 2900
    },
    {
      "epoch": 0.16393442622950818,
      "grad_norm": 0.3392621576786041,
      "learning_rate": 0.00012541803278688524,
      "loss": 3.2325253295898437,
      "step": 3000
    },
    {
      "epoch": 0.16393442622950818,
      "eval_loss": 3.116848945617676,
      "eval_runtime": 167.4485,
      "eval_samples_per_second": 51.879,
      "eval_steps_per_second": 1.624,
      "step": 3000
    },
    {
      "epoch": 0.16939890710382513,
      "grad_norm": 0.3316171169281006,
      "learning_rate": 0.0001245983606557377,
      "loss": 3.2304959106445312,
      "step": 3100
    },
    {
      "epoch": 0.17486338797814208,
      "grad_norm": 0.3558633625507355,
      "learning_rate": 0.00012377868852459016,
      "loss": 3.233113098144531,
      "step": 3200
    },
    {
      "epoch": 0.18032786885245902,
      "grad_norm": 0.3265064060688019,
      "learning_rate": 0.0001229590163934426,
      "loss": 3.2334609985351563,
      "step": 3300
    },
    {
      "epoch": 0.18579234972677597,
      "grad_norm": 0.3422014117240906,
      "learning_rate": 0.00012213934426229508,
      "loss": 3.2320993041992185,
      "step": 3400
    },
    {
      "epoch": 0.1912568306010929,
      "grad_norm": 0.3564997911453247,
      "learning_rate": 0.00012131967213114753,
      "loss": 3.24385009765625,
      "step": 3500
    },
    {
      "epoch": 0.19672131147540983,
      "grad_norm": 0.350437730550766,
      "learning_rate": 0.00012049999999999999,
      "loss": 3.2268707275390627,
      "step": 3600
    },
    {
      "epoch": 0.20218579234972678,
      "grad_norm": 0.3819025158882141,
      "learning_rate": 0.00011968032786885245,
      "loss": 3.2548171997070314,
      "step": 3700
    },
    {
      "epoch": 0.20765027322404372,
      "grad_norm": 0.3577027916908264,
      "learning_rate": 0.00011886065573770491,
      "loss": 3.242948303222656,
      "step": 3800
    },
    {
      "epoch": 0.21311475409836064,
      "grad_norm": 0.34992942214012146,
      "learning_rate": 0.00011804098360655737,
      "loss": 3.247112121582031,
      "step": 3900
    },
    {
      "epoch": 0.2185792349726776,
      "grad_norm": 0.33957329392433167,
      "learning_rate": 0.00011722131147540982,
      "loss": 3.2855084228515623,
      "step": 4000
    },
    {
      "epoch": 0.2185792349726776,
      "eval_loss": 3.1081910133361816,
      "eval_runtime": 166.2296,
      "eval_samples_per_second": 52.259,
      "eval_steps_per_second": 1.636,
      "step": 4000
    },
    {
      "epoch": 0.22404371584699453,
      "grad_norm": 0.39298462867736816,
      "learning_rate": 0.00011640163934426228,
      "loss": 3.2725079345703123,
      "step": 4100
    },
    {
      "epoch": 0.22950819672131148,
      "grad_norm": 0.3378917872905731,
      "learning_rate": 0.00011558196721311474,
      "loss": 3.2375408935546877,
      "step": 4200
    },
    {
      "epoch": 0.23497267759562843,
      "grad_norm": 0.34373924136161804,
      "learning_rate": 0.0001147622950819672,
      "loss": 3.264992980957031,
      "step": 4300
    },
    {
      "epoch": 0.24043715846994534,
      "grad_norm": 0.3387899398803711,
      "learning_rate": 0.00011394262295081966,
      "loss": 3.25844482421875,
      "step": 4400
    },
    {
      "epoch": 0.2459016393442623,
      "grad_norm": 0.35700681805610657,
      "learning_rate": 0.00011312295081967212,
      "loss": 3.262210693359375,
      "step": 4500
    },
    {
      "epoch": 0.25136612021857924,
      "grad_norm": 0.3377688229084015,
      "learning_rate": 0.00011230327868852457,
      "loss": 3.2448056030273436,
      "step": 4600
    },
    {
      "epoch": 0.2568306010928962,
      "grad_norm": 0.3596217930316925,
      "learning_rate": 0.00011148360655737704,
      "loss": 3.2498098754882814,
      "step": 4700
    },
    {
      "epoch": 0.26229508196721313,
      "grad_norm": 0.33321240544319153,
      "learning_rate": 0.0001106639344262295,
      "loss": 3.23698486328125,
      "step": 4800
    },
    {
      "epoch": 0.2677595628415301,
      "grad_norm": 0.3356533944606781,
      "learning_rate": 0.00010984426229508196,
      "loss": 3.2359185791015626,
      "step": 4900
    },
    {
      "epoch": 0.273224043715847,
      "grad_norm": 0.3382013142108917,
      "learning_rate": 0.00010902459016393442,
      "loss": 3.17927978515625,
      "step": 5000
    },
    {
      "epoch": 0.273224043715847,
      "eval_loss": 3.099688768386841,
      "eval_runtime": 167.6483,
      "eval_samples_per_second": 51.817,
      "eval_steps_per_second": 1.622,
      "step": 5000
    },
    {
      "epoch": 0.2786885245901639,
      "grad_norm": 0.3593096435070038,
      "learning_rate": 0.00010820491803278688,
      "loss": 3.2057940673828127,
      "step": 5100
    },
    {
      "epoch": 0.28415300546448086,
      "grad_norm": 0.3411197364330292,
      "learning_rate": 0.00010738524590163934,
      "loss": 3.1890634155273436,
      "step": 5200
    },
    {
      "epoch": 0.2896174863387978,
      "grad_norm": 0.3614797592163086,
      "learning_rate": 0.00010656557377049179,
      "loss": 3.1959136962890624,
      "step": 5300
    },
    {
      "epoch": 0.29508196721311475,
      "grad_norm": 0.3614165782928467,
      "learning_rate": 0.00010574590163934425,
      "loss": 3.190245361328125,
      "step": 5400
    },
    {
      "epoch": 0.3005464480874317,
      "grad_norm": 0.34652748703956604,
      "learning_rate": 0.00010492622950819671,
      "loss": 3.170889892578125,
      "step": 5500
    },
    {
      "epoch": 0.30601092896174864,
      "grad_norm": 0.3489515781402588,
      "learning_rate": 0.00010410655737704917,
      "loss": 3.202828063964844,
      "step": 5600
    },
    {
      "epoch": 0.3114754098360656,
      "grad_norm": 0.33419397473335266,
      "learning_rate": 0.00010328688524590163,
      "loss": 3.1874429321289064,
      "step": 5700
    },
    {
      "epoch": 0.31693989071038253,
      "grad_norm": 0.33504703640937805,
      "learning_rate": 0.00010246721311475408,
      "loss": 3.180716857910156,
      "step": 5800
    },
    {
      "epoch": 0.3224043715846995,
      "grad_norm": 0.3674669861793518,
      "learning_rate": 0.00010164754098360654,
      "loss": 3.1808792114257813,
      "step": 5900
    },
    {
      "epoch": 0.32786885245901637,
      "grad_norm": 0.3543224036693573,
      "learning_rate": 0.000100827868852459,
      "loss": 3.2052008056640626,
      "step": 6000
    },
    {
      "epoch": 0.32786885245901637,
      "eval_loss": 3.0916807651519775,
      "eval_runtime": 167.0778,
      "eval_samples_per_second": 51.994,
      "eval_steps_per_second": 1.628,
      "step": 6000
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.34399476647377014,
      "learning_rate": 0.00010000819672131146,
      "loss": 3.2100375366210936,
      "step": 6100
    },
    {
      "epoch": 0.33879781420765026,
      "grad_norm": 0.33845406770706177,
      "learning_rate": 9.918852459016392e-05,
      "loss": 3.2021624755859377,
      "step": 6200
    },
    {
      "epoch": 0.3442622950819672,
      "grad_norm": 0.35047417879104614,
      "learning_rate": 9.836885245901638e-05,
      "loss": 3.1676751708984376,
      "step": 6300
    },
    {
      "epoch": 0.34972677595628415,
      "grad_norm": 0.3295944929122925,
      "learning_rate": 9.754918032786883e-05,
      "loss": 3.191239929199219,
      "step": 6400
    },
    {
      "epoch": 0.3551912568306011,
      "grad_norm": 0.35513830184936523,
      "learning_rate": 9.672950819672129e-05,
      "loss": 3.2161325073242186,
      "step": 6500
    },
    {
      "epoch": 0.36065573770491804,
      "grad_norm": 0.34763020277023315,
      "learning_rate": 9.590983606557377e-05,
      "loss": 3.1975909423828126,
      "step": 6600
    },
    {
      "epoch": 0.366120218579235,
      "grad_norm": 0.35204017162323,
      "learning_rate": 9.509016393442623e-05,
      "loss": 3.2129571533203123,
      "step": 6700
    },
    {
      "epoch": 0.37158469945355194,
      "grad_norm": 0.3598620295524597,
      "learning_rate": 9.427049180327869e-05,
      "loss": 3.233637390136719,
      "step": 6800
    },
    {
      "epoch": 0.3770491803278688,
      "grad_norm": 0.3516027331352234,
      "learning_rate": 9.345081967213115e-05,
      "loss": 3.236307373046875,
      "step": 6900
    },
    {
      "epoch": 0.3825136612021858,
      "grad_norm": 0.3794242739677429,
      "learning_rate": 9.26311475409836e-05,
      "loss": 3.2149234008789063,
      "step": 7000
    },
    {
      "epoch": 0.3825136612021858,
      "eval_loss": 3.0852153301239014,
      "eval_runtime": 166.2243,
      "eval_samples_per_second": 52.261,
      "eval_steps_per_second": 1.636,
      "step": 7000
    },
    {
      "epoch": 0.3879781420765027,
      "grad_norm": 0.355341374874115,
      "learning_rate": 9.181147540983606e-05,
      "loss": 3.2003558349609373,
      "step": 7100
    },
    {
      "epoch": 0.39344262295081966,
      "grad_norm": 0.3544386029243469,
      "learning_rate": 9.099180327868852e-05,
      "loss": 3.183940734863281,
      "step": 7200
    },
    {
      "epoch": 0.3989071038251366,
      "grad_norm": 0.35201796889305115,
      "learning_rate": 9.017213114754098e-05,
      "loss": 3.2111468505859375,
      "step": 7300
    },
    {
      "epoch": 0.40437158469945356,
      "grad_norm": 0.3590731918811798,
      "learning_rate": 8.935245901639344e-05,
      "loss": 3.2194284057617186,
      "step": 7400
    },
    {
      "epoch": 0.4098360655737705,
      "grad_norm": 0.34875085949897766,
      "learning_rate": 8.85327868852459e-05,
      "loss": 3.2164602661132813,
      "step": 7500
    },
    {
      "epoch": 0.41530054644808745,
      "grad_norm": 0.36371952295303345,
      "learning_rate": 8.771311475409835e-05,
      "loss": 3.1858474731445314,
      "step": 7600
    },
    {
      "epoch": 0.4207650273224044,
      "grad_norm": 0.3518809974193573,
      "learning_rate": 8.68934426229508e-05,
      "loss": 3.15969970703125,
      "step": 7700
    },
    {
      "epoch": 0.4262295081967213,
      "grad_norm": 0.3700133264064789,
      "learning_rate": 8.607377049180327e-05,
      "loss": 3.1636505126953125,
      "step": 7800
    },
    {
      "epoch": 0.43169398907103823,
      "grad_norm": 0.34737807512283325,
      "learning_rate": 8.525409836065573e-05,
      "loss": 3.15463623046875,
      "step": 7900
    },
    {
      "epoch": 0.4371584699453552,
      "grad_norm": 0.36737293004989624,
      "learning_rate": 8.443442622950819e-05,
      "loss": 3.138258056640625,
      "step": 8000
    },
    {
      "epoch": 0.4371584699453552,
      "eval_loss": 3.0774519443511963,
      "eval_runtime": 166.4678,
      "eval_samples_per_second": 52.184,
      "eval_steps_per_second": 1.634,
      "step": 8000
    },
    {
      "epoch": 0.4426229508196721,
      "grad_norm": 0.34968650341033936,
      "learning_rate": 8.361475409836064e-05,
      "loss": 3.1346026611328126,
      "step": 8100
    },
    {
      "epoch": 0.44808743169398907,
      "grad_norm": 0.347873330116272,
      "learning_rate": 8.27950819672131e-05,
      "loss": 3.1654531860351565,
      "step": 8200
    },
    {
      "epoch": 0.453551912568306,
      "grad_norm": 0.3486824929714203,
      "learning_rate": 8.197540983606556e-05,
      "loss": 3.158487548828125,
      "step": 8300
    },
    {
      "epoch": 0.45901639344262296,
      "grad_norm": 0.34259742498397827,
      "learning_rate": 8.115573770491803e-05,
      "loss": 3.1289462280273437,
      "step": 8400
    },
    {
      "epoch": 0.4644808743169399,
      "grad_norm": 0.34342944622039795,
      "learning_rate": 8.033606557377049e-05,
      "loss": 3.1434490966796873,
      "step": 8500
    },
    {
      "epoch": 0.46994535519125685,
      "grad_norm": 0.35807445645332336,
      "learning_rate": 7.951639344262295e-05,
      "loss": 3.1603570556640626,
      "step": 8600
    },
    {
      "epoch": 0.47540983606557374,
      "grad_norm": 0.3534564971923828,
      "learning_rate": 7.869672131147541e-05,
      "loss": 3.176658020019531,
      "step": 8700
    },
    {
      "epoch": 0.4808743169398907,
      "grad_norm": 0.36163410544395447,
      "learning_rate": 7.787704918032786e-05,
      "loss": 3.1730413818359375,
      "step": 8800
    },
    {
      "epoch": 0.48633879781420764,
      "grad_norm": 0.3280462920665741,
      "learning_rate": 7.705737704918032e-05,
      "loss": 3.1541357421875,
      "step": 8900
    },
    {
      "epoch": 0.4918032786885246,
      "grad_norm": 0.34871619939804077,
      "learning_rate": 7.623770491803278e-05,
      "loss": 3.1691409301757814,
      "step": 9000
    },
    {
      "epoch": 0.4918032786885246,
      "eval_loss": 3.0718374252319336,
      "eval_runtime": 165.8669,
      "eval_samples_per_second": 52.373,
      "eval_steps_per_second": 1.64,
      "step": 9000
    },
    {
      "epoch": 0.4972677595628415,
      "grad_norm": 0.37653326988220215,
      "learning_rate": 7.541803278688524e-05,
      "loss": 3.173928527832031,
      "step": 9100
    },
    {
      "epoch": 0.5027322404371585,
      "grad_norm": 0.3302043676376343,
      "learning_rate": 7.45983606557377e-05,
      "loss": 3.185494689941406,
      "step": 9200
    },
    {
      "epoch": 0.5081967213114754,
      "grad_norm": 0.35126709938049316,
      "learning_rate": 7.377868852459016e-05,
      "loss": 3.16490234375,
      "step": 9300
    },
    {
      "epoch": 0.5136612021857924,
      "grad_norm": 0.38984546065330505,
      "learning_rate": 7.295901639344261e-05,
      "loss": 3.20061279296875,
      "step": 9400
    },
    {
      "epoch": 0.5191256830601093,
      "grad_norm": 0.3627556264400482,
      "learning_rate": 7.213934426229507e-05,
      "loss": 3.2119021606445313,
      "step": 9500
    },
    {
      "epoch": 0.5245901639344263,
      "grad_norm": 0.34836238622665405,
      "learning_rate": 7.131967213114753e-05,
      "loss": 3.192005920410156,
      "step": 9600
    },
    {
      "epoch": 0.5300546448087432,
      "grad_norm": 0.362745463848114,
      "learning_rate": 7.049999999999999e-05,
      "loss": 3.2035980224609375,
      "step": 9700
    },
    {
      "epoch": 0.5355191256830601,
      "grad_norm": 0.35114485025405884,
      "learning_rate": 6.968032786885245e-05,
      "loss": 3.200631103515625,
      "step": 9800
    },
    {
      "epoch": 0.5409836065573771,
      "grad_norm": 0.45677778124809265,
      "learning_rate": 6.886065573770491e-05,
      "loss": 3.2338800048828125,
      "step": 9900
    },
    {
      "epoch": 0.546448087431694,
      "grad_norm": 0.38240864872932434,
      "learning_rate": 6.804098360655737e-05,
      "loss": 3.2340545654296875,
      "step": 10000
    },
    {
      "epoch": 0.546448087431694,
      "eval_loss": 3.064678907394409,
      "eval_runtime": 166.0237,
      "eval_samples_per_second": 52.324,
      "eval_steps_per_second": 1.638,
      "step": 10000
    },
    {
      "epoch": 0.5519125683060109,
      "grad_norm": 0.36993658542633057,
      "learning_rate": 6.722131147540984e-05,
      "loss": 3.217573547363281,
      "step": 10100
    },
    {
      "epoch": 0.5573770491803278,
      "grad_norm": 0.34514835476875305,
      "learning_rate": 6.64016393442623e-05,
      "loss": 3.228822937011719,
      "step": 10200
    },
    {
      "epoch": 0.5628415300546448,
      "grad_norm": 0.3509940803050995,
      "learning_rate": 6.558196721311474e-05,
      "loss": 3.224437561035156,
      "step": 10300
    },
    {
      "epoch": 0.5683060109289617,
      "grad_norm": 0.34477314352989197,
      "learning_rate": 6.47622950819672e-05,
      "loss": 3.226252136230469,
      "step": 10400
    },
    {
      "epoch": 0.5737704918032787,
      "grad_norm": 0.3412553071975708,
      "learning_rate": 6.394262295081966e-05,
      "loss": 3.2269625854492188,
      "step": 10500
    },
    {
      "epoch": 0.5792349726775956,
      "grad_norm": 0.3485320508480072,
      "learning_rate": 6.312295081967212e-05,
      "loss": 3.2041766357421877,
      "step": 10600
    },
    {
      "epoch": 0.5846994535519126,
      "grad_norm": 0.33190447092056274,
      "learning_rate": 6.230327868852459e-05,
      "loss": 3.1985888671875,
      "step": 10700
    },
    {
      "epoch": 0.5901639344262295,
      "grad_norm": 0.36425039172172546,
      "learning_rate": 6.148360655737705e-05,
      "loss": 3.208878479003906,
      "step": 10800
    },
    {
      "epoch": 0.5956284153005464,
      "grad_norm": 0.3473658859729767,
      "learning_rate": 6.066393442622951e-05,
      "loss": 3.2074591064453126,
      "step": 10900
    },
    {
      "epoch": 0.6010928961748634,
      "grad_norm": 0.43133220076560974,
      "learning_rate": 5.984426229508196e-05,
      "loss": 3.2229376220703125,
      "step": 11000
    },
    {
      "epoch": 0.6010928961748634,
      "eval_loss": 3.0597853660583496,
      "eval_runtime": 165.5808,
      "eval_samples_per_second": 52.464,
      "eval_steps_per_second": 1.643,
      "step": 11000
    },
    {
      "epoch": 0.6065573770491803,
      "grad_norm": 0.37588736414909363,
      "learning_rate": 5.902459016393442e-05,
      "loss": 3.1838412475585938,
      "step": 11100
    },
    {
      "epoch": 0.6120218579234973,
      "grad_norm": 0.34238529205322266,
      "learning_rate": 5.820491803278688e-05,
      "loss": 3.1839044189453123,
      "step": 11200
    },
    {
      "epoch": 0.6174863387978142,
      "grad_norm": 0.3399714231491089,
      "learning_rate": 5.7385245901639336e-05,
      "loss": 3.2120074462890624,
      "step": 11300
    },
    {
      "epoch": 0.6229508196721312,
      "grad_norm": 0.3523642420768738,
      "learning_rate": 5.6565573770491796e-05,
      "loss": 3.1816641235351564,
      "step": 11400
    },
    {
      "epoch": 0.6284153005464481,
      "grad_norm": 0.3372739255428314,
      "learning_rate": 5.574590163934426e-05,
      "loss": 3.194277038574219,
      "step": 11500
    },
    {
      "epoch": 0.6338797814207651,
      "grad_norm": 0.36970576643943787,
      "learning_rate": 5.492622950819672e-05,
      "loss": 3.183469543457031,
      "step": 11600
    },
    {
      "epoch": 0.639344262295082,
      "grad_norm": 0.3468629717826843,
      "learning_rate": 5.410655737704918e-05,
      "loss": 3.1655535888671875,
      "step": 11700
    },
    {
      "epoch": 0.644808743169399,
      "grad_norm": 0.3488897979259491,
      "learning_rate": 5.328688524590164e-05,
      "loss": 3.1525448608398436,
      "step": 11800
    },
    {
      "epoch": 0.6502732240437158,
      "grad_norm": 0.3539671301841736,
      "learning_rate": 5.246721311475409e-05,
      "loss": 3.1822393798828124,
      "step": 11900
    },
    {
      "epoch": 0.6557377049180327,
      "grad_norm": 0.3764471709728241,
      "learning_rate": 5.164754098360655e-05,
      "loss": 3.1826312255859377,
      "step": 12000
    },
    {
      "epoch": 0.6557377049180327,
      "eval_loss": 3.053892135620117,
      "eval_runtime": 169.2137,
      "eval_samples_per_second": 51.337,
      "eval_steps_per_second": 1.607,
      "step": 12000
    },
    {
      "epoch": 0.6612021857923497,
      "grad_norm": 0.36165982484817505,
      "learning_rate": 5.0827868852459014e-05,
      "loss": 3.1621029663085936,
      "step": 12100
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.3601846694946289,
      "learning_rate": 5.000819672131147e-05,
      "loss": 3.136105651855469,
      "step": 12200
    },
    {
      "epoch": 0.6721311475409836,
      "grad_norm": 0.3468482196331024,
      "learning_rate": 4.918852459016393e-05,
      "loss": 3.1172000122070314,
      "step": 12300
    },
    {
      "epoch": 0.6775956284153005,
      "grad_norm": 0.3401918411254883,
      "learning_rate": 4.836885245901638e-05,
      "loss": 3.1227383422851562,
      "step": 12400
    },
    {
      "epoch": 0.6830601092896175,
      "grad_norm": 0.3300206661224365,
      "learning_rate": 4.754918032786885e-05,
      "loss": 3.1588729858398437,
      "step": 12500
    },
    {
      "epoch": 0.6885245901639344,
      "grad_norm": 0.33165860176086426,
      "learning_rate": 4.672950819672131e-05,
      "loss": 3.163753662109375,
      "step": 12600
    },
    {
      "epoch": 0.6939890710382514,
      "grad_norm": 0.35165345668792725,
      "learning_rate": 4.590983606557377e-05,
      "loss": 3.1771914672851564,
      "step": 12700
    },
    {
      "epoch": 0.6994535519125683,
      "grad_norm": 0.37126895785331726,
      "learning_rate": 4.5090163934426225e-05,
      "loss": 3.182335205078125,
      "step": 12800
    },
    {
      "epoch": 0.7049180327868853,
      "grad_norm": 0.355465292930603,
      "learning_rate": 4.4270491803278686e-05,
      "loss": 3.1645904541015626,
      "step": 12900
    },
    {
      "epoch": 0.7103825136612022,
      "grad_norm": 0.34211593866348267,
      "learning_rate": 4.3450819672131146e-05,
      "loss": 3.160174560546875,
      "step": 13000
    },
    {
      "epoch": 0.7103825136612022,
      "eval_loss": 3.048466444015503,
      "eval_runtime": 170.5338,
      "eval_samples_per_second": 50.94,
      "eval_steps_per_second": 1.595,
      "step": 13000
    },
    {
      "epoch": 0.7158469945355191,
      "grad_norm": 0.351980984210968,
      "learning_rate": 4.26311475409836e-05,
      "loss": 3.180731201171875,
      "step": 13100
    },
    {
      "epoch": 0.7213114754098361,
      "grad_norm": 0.3479735851287842,
      "learning_rate": 4.181147540983606e-05,
      "loss": 3.199087829589844,
      "step": 13200
    },
    {
      "epoch": 0.726775956284153,
      "grad_norm": 0.34766390919685364,
      "learning_rate": 4.0991803278688515e-05,
      "loss": 3.2096636962890623,
      "step": 13300
    },
    {
      "epoch": 0.73224043715847,
      "grad_norm": 0.3618048429489136,
      "learning_rate": 4.017213114754098e-05,
      "loss": 3.214843444824219,
      "step": 13400
    },
    {
      "epoch": 0.7377049180327869,
      "grad_norm": 0.36265110969543457,
      "learning_rate": 3.935245901639344e-05,
      "loss": 3.1923486328125,
      "step": 13500
    },
    {
      "epoch": 0.7431693989071039,
      "grad_norm": 0.36705461144447327,
      "learning_rate": 3.85327868852459e-05,
      "loss": 3.206243591308594,
      "step": 13600
    },
    {
      "epoch": 0.7486338797814208,
      "grad_norm": 0.3736395835876465,
      "learning_rate": 3.771311475409836e-05,
      "loss": 3.1797048950195315,
      "step": 13700
    },
    {
      "epoch": 0.7540983606557377,
      "grad_norm": 0.353672593832016,
      "learning_rate": 3.689344262295082e-05,
      "loss": 3.199601745605469,
      "step": 13800
    },
    {
      "epoch": 0.7595628415300546,
      "grad_norm": 0.3430495858192444,
      "learning_rate": 3.607377049180327e-05,
      "loss": 3.2101876831054685,
      "step": 13900
    },
    {
      "epoch": 0.7650273224043715,
      "grad_norm": 0.36116746068000793,
      "learning_rate": 3.525409836065573e-05,
      "loss": 3.2108331298828126,
      "step": 14000
    },
    {
      "epoch": 0.7650273224043715,
      "eval_loss": 3.0441460609436035,
      "eval_runtime": 169.191,
      "eval_samples_per_second": 51.344,
      "eval_steps_per_second": 1.608,
      "step": 14000
    },
    {
      "epoch": 0.7704918032786885,
      "grad_norm": 0.37100476026535034,
      "learning_rate": 3.443442622950819e-05,
      "loss": 3.2165988159179686,
      "step": 14100
    },
    {
      "epoch": 0.7759562841530054,
      "grad_norm": 0.36540594696998596,
      "learning_rate": 3.3614754098360654e-05,
      "loss": 3.193106994628906,
      "step": 14200
    },
    {
      "epoch": 0.7814207650273224,
      "grad_norm": 0.340986043214798,
      "learning_rate": 3.2795081967213114e-05,
      "loss": 3.205763244628906,
      "step": 14300
    },
    {
      "epoch": 0.7868852459016393,
      "grad_norm": 0.35636964440345764,
      "learning_rate": 3.197540983606557e-05,
      "loss": 3.220958251953125,
      "step": 14400
    },
    {
      "epoch": 0.7923497267759563,
      "grad_norm": 0.35717952251434326,
      "learning_rate": 3.115573770491803e-05,
      "loss": 3.225035400390625,
      "step": 14500
    },
    {
      "epoch": 0.7978142076502732,
      "grad_norm": 0.3586474061012268,
      "learning_rate": 3.033606557377049e-05,
      "loss": 3.205286560058594,
      "step": 14600
    },
    {
      "epoch": 0.8032786885245902,
      "grad_norm": 0.3829101324081421,
      "learning_rate": 2.951639344262295e-05,
      "loss": 3.2143185424804686,
      "step": 14700
    },
    {
      "epoch": 0.8087431693989071,
      "grad_norm": 0.3761688768863678,
      "learning_rate": 2.8696721311475407e-05,
      "loss": 3.189816589355469,
      "step": 14800
    },
    {
      "epoch": 0.8142076502732241,
      "grad_norm": 0.3838573396205902,
      "learning_rate": 2.7877049180327865e-05,
      "loss": 3.1748077392578127,
      "step": 14900
    },
    {
      "epoch": 0.819672131147541,
      "grad_norm": 0.3493940532207489,
      "learning_rate": 2.705737704918033e-05,
      "loss": 3.18576171875,
      "step": 15000
    },
    {
      "epoch": 0.819672131147541,
      "eval_loss": 3.040698766708374,
      "eval_runtime": 168.2186,
      "eval_samples_per_second": 51.641,
      "eval_steps_per_second": 1.617,
      "step": 15000
    },
    {
      "epoch": 0.825136612021858,
      "grad_norm": 0.34466949105262756,
      "learning_rate": 2.6237704918032786e-05,
      "loss": 3.1684368896484374,
      "step": 15100
    },
    {
      "epoch": 0.8306010928961749,
      "grad_norm": 0.3771664798259735,
      "learning_rate": 2.5418032786885243e-05,
      "loss": 3.202015075683594,
      "step": 15200
    },
    {
      "epoch": 0.8360655737704918,
      "grad_norm": 0.37019965052604675,
      "learning_rate": 2.45983606557377e-05,
      "loss": 3.203460693359375,
      "step": 15300
    },
    {
      "epoch": 0.8415300546448088,
      "grad_norm": 0.3610929250717163,
      "learning_rate": 2.377868852459016e-05,
      "loss": 3.1997000122070314,
      "step": 15400
    },
    {
      "epoch": 0.8469945355191257,
      "grad_norm": 0.35692885518074036,
      "learning_rate": 2.2959016393442622e-05,
      "loss": 3.1686630249023438,
      "step": 15500
    },
    {
      "epoch": 0.8524590163934426,
      "grad_norm": 0.36669033765792847,
      "learning_rate": 2.213934426229508e-05,
      "loss": 3.165161437988281,
      "step": 15600
    },
    {
      "epoch": 0.8579234972677595,
      "grad_norm": 0.35931694507598877,
      "learning_rate": 2.131967213114754e-05,
      "loss": 3.1681549072265627,
      "step": 15700
    },
    {
      "epoch": 0.8633879781420765,
      "grad_norm": 0.36468565464019775,
      "learning_rate": 2.0499999999999997e-05,
      "loss": 3.150556945800781,
      "step": 15800
    },
    {
      "epoch": 0.8688524590163934,
      "grad_norm": 0.3460843861103058,
      "learning_rate": 1.968032786885246e-05,
      "loss": 3.1520755004882814,
      "step": 15900
    },
    {
      "epoch": 0.8743169398907104,
      "grad_norm": 0.355943888425827,
      "learning_rate": 1.8860655737704918e-05,
      "loss": 3.151009521484375,
      "step": 16000
    },
    {
      "epoch": 0.8743169398907104,
      "eval_loss": 3.03753924369812,
      "eval_runtime": 169.1323,
      "eval_samples_per_second": 51.362,
      "eval_steps_per_second": 1.608,
      "step": 16000
    },
    {
      "epoch": 0.8797814207650273,
      "grad_norm": 0.3545354902744293,
      "learning_rate": 1.8040983606557375e-05,
      "loss": 3.127799072265625,
      "step": 16100
    },
    {
      "epoch": 0.8852459016393442,
      "grad_norm": 0.3566863536834717,
      "learning_rate": 1.7221311475409833e-05,
      "loss": 3.1025897216796876,
      "step": 16200
    },
    {
      "epoch": 0.8907103825136612,
      "grad_norm": 0.337934285402298,
      "learning_rate": 1.6401639344262293e-05,
      "loss": 3.1121600341796873,
      "step": 16300
    },
    {
      "epoch": 0.8961748633879781,
      "grad_norm": 0.34477803111076355,
      "learning_rate": 1.5581967213114754e-05,
      "loss": 3.130706787109375,
      "step": 16400
    },
    {
      "epoch": 0.9016393442622951,
      "grad_norm": 0.3452692925930023,
      "learning_rate": 1.4762295081967213e-05,
      "loss": 3.1331210327148438,
      "step": 16500
    },
    {
      "epoch": 0.907103825136612,
      "grad_norm": 0.348063200712204,
      "learning_rate": 1.394262295081967e-05,
      "loss": 3.1176373291015627,
      "step": 16600
    },
    {
      "epoch": 0.912568306010929,
      "grad_norm": 0.335582971572876,
      "learning_rate": 1.312295081967213e-05,
      "loss": 3.1302474975585937,
      "step": 16700
    },
    {
      "epoch": 0.9180327868852459,
      "grad_norm": 0.34352442622184753,
      "learning_rate": 1.230327868852459e-05,
      "loss": 3.119664306640625,
      "step": 16800
    },
    {
      "epoch": 0.9234972677595629,
      "grad_norm": 0.370101660490036,
      "learning_rate": 1.1483606557377047e-05,
      "loss": 3.11188720703125,
      "step": 16900
    },
    {
      "epoch": 0.9289617486338798,
      "grad_norm": 0.35088032484054565,
      "learning_rate": 1.0663934426229508e-05,
      "loss": 3.13305908203125,
      "step": 17000
    },
    {
      "epoch": 0.9289617486338798,
      "eval_loss": 3.0354855060577393,
      "eval_runtime": 166.8874,
      "eval_samples_per_second": 52.053,
      "eval_steps_per_second": 1.63,
      "step": 17000
    },
    {
      "epoch": 0.9344262295081968,
      "grad_norm": 0.35557523369789124,
      "learning_rate": 9.844262295081967e-06,
      "loss": 3.1132748413085936,
      "step": 17100
    },
    {
      "epoch": 0.9398907103825137,
      "grad_norm": 0.3398247957229614,
      "learning_rate": 9.024590163934426e-06,
      "loss": 3.141661376953125,
      "step": 17200
    },
    {
      "epoch": 0.9453551912568307,
      "grad_norm": 0.34634870290756226,
      "learning_rate": 8.204918032786884e-06,
      "loss": 3.1398541259765627,
      "step": 17300
    },
    {
      "epoch": 0.9508196721311475,
      "grad_norm": 0.3561513423919678,
      "learning_rate": 7.3852459016393434e-06,
      "loss": 3.1273211669921874,
      "step": 17400
    },
    {
      "epoch": 0.9562841530054644,
      "grad_norm": 0.35161858797073364,
      "learning_rate": 6.565573770491802e-06,
      "loss": 3.1056549072265627,
      "step": 17500
    },
    {
      "epoch": 0.9617486338797814,
      "grad_norm": 0.3520038425922394,
      "learning_rate": 5.745901639344262e-06,
      "loss": 3.1260693359375,
      "step": 17600
    },
    {
      "epoch": 0.9672131147540983,
      "grad_norm": 0.34525951743125916,
      "learning_rate": 4.926229508196721e-06,
      "loss": 3.1003762817382814,
      "step": 17700
    },
    {
      "epoch": 0.9726775956284153,
      "grad_norm": 0.35116466879844666,
      "learning_rate": 4.10655737704918e-06,
      "loss": 3.1081192016601564,
      "step": 17800
    },
    {
      "epoch": 0.9781420765027322,
      "grad_norm": 0.36404362320899963,
      "learning_rate": 3.286885245901639e-06,
      "loss": 3.123824462890625,
      "step": 17900
    },
    {
      "epoch": 0.9836065573770492,
      "grad_norm": 0.3350881338119507,
      "learning_rate": 2.4672131147540984e-06,
      "loss": 3.122076110839844,
      "step": 18000
    },
    {
      "epoch": 0.9836065573770492,
      "eval_loss": 3.033825397491455,
      "eval_runtime": 163.9316,
      "eval_samples_per_second": 52.992,
      "eval_steps_per_second": 1.659,
      "step": 18000
    }
  ],
  "logging_steps": 100,
  "max_steps": 18300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.01008420864e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}