{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.807017543859649,
  "eval_steps": 30,
  "global_step": 800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03508771929824561,
      "grad_norm": 4.878787040710449,
      "learning_rate": 2.0930232558139536e-05,
      "loss": 2.4934,
      "step": 10
    },
    {
      "epoch": 0.07017543859649122,
      "grad_norm": 0.776142418384552,
      "learning_rate": 4.418604651162791e-05,
      "loss": 0.9685,
      "step": 20
    },
    {
      "epoch": 0.10526315789473684,
      "grad_norm": 0.7527797818183899,
      "learning_rate": 6.744186046511628e-05,
      "loss": 0.5823,
      "step": 30
    },
    {
      "epoch": 0.10526315789473684,
      "eval_loss": 0.45968249440193176,
      "eval_runtime": 25.548,
      "eval_samples_per_second": 4.697,
      "eval_steps_per_second": 4.697,
      "step": 30
    },
    {
      "epoch": 0.14035087719298245,
      "grad_norm": 0.5810465812683105,
      "learning_rate": 9.069767441860465e-05,
      "loss": 0.4143,
      "step": 40
    },
    {
      "epoch": 0.17543859649122806,
      "grad_norm": 0.6364761590957642,
      "learning_rate": 0.00011395348837209304,
      "loss": 0.3,
      "step": 50
    },
    {
      "epoch": 0.21052631578947367,
      "grad_norm": 0.7183641791343689,
      "learning_rate": 0.0001372093023255814,
      "loss": 0.2551,
      "step": 60
    },
    {
      "epoch": 0.21052631578947367,
      "eval_loss": 0.21707303822040558,
      "eval_runtime": 24.8756,
      "eval_samples_per_second": 4.824,
      "eval_steps_per_second": 4.824,
      "step": 60
    },
    {
      "epoch": 0.24561403508771928,
      "grad_norm": 0.4606807231903076,
      "learning_rate": 0.00016046511627906978,
      "loss": 0.223,
      "step": 70
    },
    {
      "epoch": 0.2807017543859649,
      "grad_norm": 0.5203229784965515,
      "learning_rate": 0.00018372093023255815,
      "loss": 0.1933,
      "step": 80
    },
    {
      "epoch": 0.3157894736842105,
      "grad_norm": 0.44623491168022156,
      "learning_rate": 0.0001999924897543333,
      "loss": 0.1902,
      "step": 90
    },
    {
      "epoch": 0.3157894736842105,
      "eval_loss": 0.1671793907880783,
      "eval_runtime": 25.9688,
      "eval_samples_per_second": 4.621,
      "eval_steps_per_second": 4.621,
      "step": 90
    },
    {
      "epoch": 0.3508771929824561,
      "grad_norm": 0.5045297145843506,
      "learning_rate": 0.000199859005655354,
      "loss": 0.1724,
      "step": 100
    },
    {
      "epoch": 0.38596491228070173,
      "grad_norm": 0.2889767289161682,
      "learning_rate": 0.00019955888361169273,
      "loss": 0.1666,
      "step": 110
    },
    {
      "epoch": 0.42105263157894735,
      "grad_norm": 0.4197916090488434,
      "learning_rate": 0.00019909262444668715,
      "loss": 0.1611,
      "step": 120
    },
    {
      "epoch": 0.42105263157894735,
      "eval_loss": 0.13614678382873535,
      "eval_runtime": 25.0008,
      "eval_samples_per_second": 4.8,
      "eval_steps_per_second": 4.8,
      "step": 120
    },
    {
      "epoch": 0.45614035087719296,
      "grad_norm": 0.3050219714641571,
      "learning_rate": 0.00019846100622204974,
      "loss": 0.1593,
      "step": 130
    },
    {
      "epoch": 0.49122807017543857,
      "grad_norm": 0.24451592564582825,
      "learning_rate": 0.00019766508293949108,
      "loss": 0.1594,
      "step": 140
    },
    {
      "epoch": 0.5263157894736842,
      "grad_norm": 0.29427045583724976,
      "learning_rate": 0.00019670618278187318,
      "loss": 0.1424,
      "step": 150
    },
    {
      "epoch": 0.5263157894736842,
      "eval_loss": 0.13314871490001678,
      "eval_runtime": 25.9586,
      "eval_samples_per_second": 4.623,
      "eval_steps_per_second": 4.623,
      "step": 150
    },
    {
      "epoch": 0.5614035087719298,
      "grad_norm": 0.36817243695259094,
      "learning_rate": 0.00019558590589682795,
      "loss": 0.1383,
      "step": 160
    },
    {
      "epoch": 0.5964912280701754,
      "grad_norm": 0.16424335539340973,
      "learning_rate": 0.00019430612172653908,
      "loss": 0.1447,
      "step": 170
    },
    {
      "epoch": 0.631578947368421,
      "grad_norm": 0.17733046412467957,
      "learning_rate": 0.00019286896588814373,
      "loss": 0.1335,
      "step": 180
    },
    {
      "epoch": 0.631578947368421,
      "eval_loss": 0.13173671066761017,
      "eval_runtime": 25.9972,
      "eval_samples_per_second": 4.616,
      "eval_steps_per_second": 4.616,
      "step": 180
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.22004777193069458,
      "learning_rate": 0.00019127683660995914,
      "loss": 0.1457,
      "step": 190
    },
    {
      "epoch": 0.7017543859649122,
      "grad_norm": 0.2817850112915039,
      "learning_rate": 0.00018953239072948182,
      "loss": 0.1304,
      "step": 200
    },
    {
      "epoch": 0.7368421052631579,
      "grad_norm": 0.22587062418460846,
      "learning_rate": 0.00018763853925983693,
      "loss": 0.1426,
      "step": 210
    },
    {
      "epoch": 0.7368421052631579,
      "eval_loss": 0.12666551768779755,
      "eval_runtime": 25.8854,
      "eval_samples_per_second": 4.636,
      "eval_steps_per_second": 4.636,
      "step": 210
    },
    {
      "epoch": 0.7719298245614035,
      "grad_norm": 0.1741916984319687,
      "learning_rate": 0.0001855984425320769,
      "loss": 0.1404,
      "step": 220
    },
    {
      "epoch": 0.8070175438596491,
      "grad_norm": 0.14528629183769226,
      "learning_rate": 0.00018341550492143496,
      "loss": 0.1486,
      "step": 230
    },
    {
      "epoch": 0.8421052631578947,
      "grad_norm": 0.26741355657577515,
      "learning_rate": 0.00018109336916633426,
      "loss": 0.1281,
      "step": 240
    },
    {
      "epoch": 0.8421052631578947,
      "eval_loss": 0.12144554406404495,
      "eval_runtime": 24.7052,
      "eval_samples_per_second": 4.857,
      "eval_steps_per_second": 4.857,
      "step": 240
    },
    {
      "epoch": 0.8771929824561403,
      "grad_norm": 0.12268830090761185,
      "learning_rate": 0.000178635910289633,
      "loss": 0.1281,
      "step": 250
    },
    {
      "epoch": 0.9122807017543859,
      "grad_norm": 0.2916463613510132,
      "learning_rate": 0.00017604722913224842,
      "loss": 0.1348,
      "step": 260
    },
    {
      "epoch": 0.9473684210526315,
      "grad_norm": 0.13800819218158722,
      "learning_rate": 0.00017333164550995152,
      "loss": 0.1258,
      "step": 270
    },
    {
      "epoch": 0.9473684210526315,
      "eval_loss": 0.11921060085296631,
      "eval_runtime": 24.7901,
      "eval_samples_per_second": 4.841,
      "eval_steps_per_second": 4.841,
      "step": 270
    },
    {
      "epoch": 0.9824561403508771,
      "grad_norm": 0.11254072189331055,
      "learning_rate": 0.000170493691004751,
      "loss": 0.1304,
      "step": 280
    },
    {
      "epoch": 1.0175438596491229,
      "grad_norm": 0.14597080647945404,
      "learning_rate": 0.00016753810140289607,
      "loss": 0.1288,
      "step": 290
    },
    {
      "epoch": 1.0526315789473684,
      "grad_norm": 0.12981514632701874,
      "learning_rate": 0.0001644698087921173,
      "loss": 0.1307,
      "step": 300
    },
    {
      "epoch": 1.0526315789473684,
      "eval_loss": 0.1170278862118721,
      "eval_runtime": 24.9635,
      "eval_samples_per_second": 4.807,
      "eval_steps_per_second": 4.807,
      "step": 300
    },
    {
      "epoch": 1.087719298245614,
      "grad_norm": 0.10934378206729889,
      "learning_rate": 0.0001612939333312926,
      "loss": 0.1111,
      "step": 310
    },
    {
      "epoch": 1.1228070175438596,
      "grad_norm": 0.1015116274356842,
      "learning_rate": 0.00015801577470627286,
      "loss": 0.1288,
      "step": 320
    },
    {
      "epoch": 1.1578947368421053,
      "grad_norm": 0.1372503787279129,
      "learning_rate": 0.0001546408032861252,
      "loss": 0.116,
      "step": 330
    },
    {
      "epoch": 1.1578947368421053,
      "eval_loss": 0.11487225443124771,
      "eval_runtime": 24.9694,
      "eval_samples_per_second": 4.806,
      "eval_steps_per_second": 4.806,
      "step": 330
    },
    {
      "epoch": 1.1929824561403508,
      "grad_norm": 0.2865680456161499,
      "learning_rate": 0.00015117465099455174,
      "loss": 0.1237,
      "step": 340
    },
    {
      "epoch": 1.2280701754385965,
      "grad_norm": 0.12083390355110168,
      "learning_rate": 0.00014762310191171655,
      "loss": 0.1305,
      "step": 350
    },
    {
      "epoch": 1.263157894736842,
      "grad_norm": 0.09869180619716644,
      "learning_rate": 0.00014399208262216474,
      "loss": 0.1167,
      "step": 360
    },
    {
      "epoch": 1.263157894736842,
      "eval_loss": 0.11612139642238617,
      "eval_runtime": 24.9809,
      "eval_samples_per_second": 4.804,
      "eval_steps_per_second": 4.804,
      "step": 360
    },
    {
      "epoch": 1.2982456140350878,
      "grad_norm": 0.10779603570699692,
      "learning_rate": 0.0001402876523249394,
      "loss": 0.1325,
      "step": 370
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.20813943445682526,
      "learning_rate": 0.00013651599272240076,
      "loss": 0.1203,
      "step": 380
    },
    {
      "epoch": 1.368421052631579,
      "grad_norm": 0.10286935418844223,
      "learning_rate": 0.0001326833977046199,
      "loss": 0.1198,
      "step": 390
    },
    {
      "epoch": 1.368421052631579,
      "eval_loss": 0.1140877828001976,
      "eval_runtime": 25.1216,
      "eval_samples_per_second": 4.777,
      "eval_steps_per_second": 4.777,
      "step": 390
    },
    {
      "epoch": 1.4035087719298245,
      "grad_norm": 0.10826771706342697,
      "learning_rate": 0.00012879626284656142,
      "loss": 0.1215,
      "step": 400
    },
    {
      "epoch": 1.4385964912280702,
      "grad_norm": 0.09107205271720886,
      "learning_rate": 0.00012486107473558118,
      "loss": 0.1258,
      "step": 410
    },
    {
      "epoch": 1.4736842105263157,
      "grad_norm": 0.23670166730880737,
      "learning_rate": 0.00012088440014704858,
      "loss": 0.1194,
      "step": 420
    },
    {
      "epoch": 1.4736842105263157,
      "eval_loss": 0.11560888588428497,
      "eval_runtime": 25.1077,
      "eval_samples_per_second": 4.779,
      "eval_steps_per_second": 4.779,
      "step": 420
    },
    {
      "epoch": 1.5087719298245614,
      "grad_norm": 0.09810595214366913,
      "learning_rate": 0.0001168728750861567,
      "loss": 0.1282,
      "step": 430
    },
    {
      "epoch": 1.543859649122807,
      "grad_norm": 0.11235892027616501,
      "learning_rate": 0.0001128331937142062,
      "loss": 0.1234,
      "step": 440
    },
    {
      "epoch": 1.5789473684210527,
      "grad_norm": 0.09267847239971161,
      "learning_rate": 0.0001087720971778426,
      "loss": 0.1292,
      "step": 450
    },
    {
      "epoch": 1.5789473684210527,
      "eval_loss": 0.11425888538360596,
      "eval_runtime": 25.0231,
      "eval_samples_per_second": 4.796,
      "eval_steps_per_second": 4.796,
      "step": 450
    },
    {
      "epoch": 1.6140350877192984,
      "grad_norm": 0.12776316702365875,
      "learning_rate": 0.0001046963623598871,
      "loss": 0.12,
      "step": 460
    },
    {
      "epoch": 1.6491228070175439,
      "grad_norm": 0.12537536025047302,
      "learning_rate": 0.00010061279057053386,
      "loss": 0.1231,
      "step": 470
    },
    {
      "epoch": 1.6842105263157894,
      "grad_norm": 0.11115119606256485,
      "learning_rate": 9.652819619778386e-05,
      "loss": 0.1262,
      "step": 480
    },
    {
      "epoch": 1.6842105263157894,
      "eval_loss": 0.11288858950138092,
      "eval_runtime": 24.9273,
      "eval_samples_per_second": 4.814,
      "eval_steps_per_second": 4.814,
      "step": 480
    },
    {
      "epoch": 1.719298245614035,
      "grad_norm": 0.09119213372468948,
      "learning_rate": 9.244939533605619e-05,
      "loss": 0.1233,
      "step": 490
    },
    {
      "epoch": 1.7543859649122808,
      "grad_norm": 0.12485823780298233,
      "learning_rate": 8.838319441195105e-05,
      "loss": 0.1212,
      "step": 500
    },
    {
      "epoch": 1.7894736842105263,
      "grad_norm": 0.10320759564638138,
      "learning_rate": 8.433637882614625e-05,
      "loss": 0.1298,
      "step": 510
    },
    {
      "epoch": 1.7894736842105263,
      "eval_loss": 0.11233663558959961,
      "eval_runtime": 24.9513,
      "eval_samples_per_second": 4.809,
      "eval_steps_per_second": 4.809,
      "step": 510
    },
    {
      "epoch": 1.8245614035087718,
      "grad_norm": 0.115440234541893,
      "learning_rate": 8.031570163038005e-05,
      "loss": 0.1298,
      "step": 520
    },
    {
      "epoch": 1.8596491228070176,
      "grad_norm": 0.09397278726100922,
      "learning_rate": 7.632787225841593e-05,
      "loss": 0.1258,
      "step": 530
    },
    {
      "epoch": 1.8947368421052633,
      "grad_norm": 0.08502210676670074,
      "learning_rate": 7.237954532979401e-05,
      "loss": 0.119,
      "step": 540
    },
    {
      "epoch": 1.8947368421052633,
      "eval_loss": 0.11294586211442947,
      "eval_runtime": 25.0566,
      "eval_samples_per_second": 4.789,
      "eval_steps_per_second": 4.789,
      "step": 540
    },
    {
      "epoch": 1.9298245614035088,
      "grad_norm": 0.09859590232372284,
      "learning_rate": 6.847730954505261e-05,
      "loss": 0.1222,
      "step": 550
    },
    {
      "epoch": 1.9649122807017543,
      "grad_norm": 0.07687846571207047,
      "learning_rate": 6.462767669095109e-05,
      "loss": 0.1143,
      "step": 560
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.09160079807043076,
      "learning_rate": 6.0837070774041284e-05,
      "loss": 0.1227,
      "step": 570
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.11345366388559341,
      "eval_runtime": 25.1502,
      "eval_samples_per_second": 4.771,
      "eval_steps_per_second": 4.771,
      "step": 570
    },
    {
      "epoch": 2.0350877192982457,
      "grad_norm": 0.08900679647922516,
      "learning_rate": 5.711181730072044e-05,
      "loss": 0.1068,
      "step": 580
    },
    {
      "epoch": 2.0701754385964914,
      "grad_norm": 0.11684390157461166,
      "learning_rate": 5.3458132721654564e-05,
      "loss": 0.122,
      "step": 590
    },
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 0.11595682054758072,
      "learning_rate": 4.9882114058186616e-05,
      "loss": 0.1493,
      "step": 600
    },
    {
      "epoch": 2.1052631578947367,
      "eval_loss": 0.11257700622081757,
      "eval_runtime": 24.9265,
      "eval_samples_per_second": 4.814,
      "eval_steps_per_second": 4.814,
      "step": 600
    },
    {
      "epoch": 2.1403508771929824,
      "grad_norm": 0.10183674842119217,
      "learning_rate": 4.638972872804038e-05,
      "loss": 0.1168,
      "step": 610
    },
    {
      "epoch": 2.175438596491228,
      "grad_norm": 0.10958362370729446,
      "learning_rate": 4.298680458729792e-05,
      "loss": 0.1206,
      "step": 620
    },
    {
      "epoch": 2.2105263157894735,
      "grad_norm": 0.09745143353939056,
      "learning_rate": 3.967902020526797e-05,
      "loss": 0.1272,
      "step": 630
    },
    {
      "epoch": 2.2105263157894735,
      "eval_loss": 0.11299394816160202,
      "eval_runtime": 25.0837,
      "eval_samples_per_second": 4.784,
      "eval_steps_per_second": 4.784,
      "step": 630
    },
    {
      "epoch": 2.245614035087719,
      "grad_norm": 0.08707338571548462,
      "learning_rate": 3.647189538847432e-05,
      "loss": 0.1212,
      "step": 640
    },
    {
      "epoch": 2.280701754385965,
      "grad_norm": 0.09840196371078491,
      "learning_rate": 3.337078196957647e-05,
      "loss": 0.1137,
      "step": 650
    },
    {
      "epoch": 2.3157894736842106,
      "grad_norm": 0.10066810250282288,
      "learning_rate": 3.0380854876593723e-05,
      "loss": 0.1147,
      "step": 660
    },
    {
      "epoch": 2.3157894736842106,
      "eval_loss": 0.11169523745775223,
      "eval_runtime": 24.9457,
      "eval_samples_per_second": 4.81,
      "eval_steps_per_second": 4.81,
      "step": 660
    },
    {
      "epoch": 2.3508771929824563,
      "grad_norm": 0.11805769056081772,
      "learning_rate": 2.7507103497336016e-05,
      "loss": 0.124,
      "step": 670
    },
    {
      "epoch": 2.3859649122807016,
      "grad_norm": 0.09834598004817963,
      "learning_rate": 2.475432335345128e-05,
      "loss": 0.1269,
      "step": 680
    },
    {
      "epoch": 2.4210526315789473,
      "grad_norm": 0.11241532862186432,
      "learning_rate": 2.212710809798393e-05,
      "loss": 0.1137,
      "step": 690
    },
    {
      "epoch": 2.4210526315789473,
      "eval_loss": 0.11132458597421646,
      "eval_runtime": 24.998,
      "eval_samples_per_second": 4.8,
      "eval_steps_per_second": 4.8,
      "step": 690
    },
    {
      "epoch": 2.456140350877193,
      "grad_norm": 0.10056956857442856,
      "learning_rate": 1.9629841849797736e-05,
      "loss": 0.1151,
      "step": 700
    },
    {
      "epoch": 2.4912280701754383,
      "grad_norm": 0.09482597559690475,
      "learning_rate": 1.7266691877655127e-05,
      "loss": 0.1091,
      "step": 710
    },
    {
      "epoch": 2.526315789473684,
      "grad_norm": 0.11021561920642853,
      "learning_rate": 1.5041601646161585e-05,
      "loss": 0.1225,
      "step": 720
    },
    {
      "epoch": 2.526315789473684,
      "eval_loss": 0.11140093952417374,
      "eval_runtime": 25.0288,
      "eval_samples_per_second": 4.794,
      "eval_steps_per_second": 4.794,
      "step": 720
    },
    {
      "epoch": 2.56140350877193,
      "grad_norm": 0.09573051333427429,
      "learning_rate": 1.295828423517874e-05,
      "loss": 0.1124,
      "step": 730
    },
    {
      "epoch": 2.5964912280701755,
      "grad_norm": 0.1159501001238823,
      "learning_rate": 1.1020216143688444e-05,
      "loss": 0.1181,
      "step": 740
    },
    {
      "epoch": 2.6315789473684212,
      "grad_norm": 0.09099686145782471,
      "learning_rate": 9.230631488446639e-06,
      "loss": 0.1185,
      "step": 750
    },
    {
      "epoch": 2.6315789473684212,
      "eval_loss": 0.11138518154621124,
      "eval_runtime": 24.9674,
      "eval_samples_per_second": 4.806,
      "eval_steps_per_second": 4.806,
      "step": 750
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.10761657357215881,
      "learning_rate": 7.592516607108324e-06,
      "loss": 0.1131,
      "step": 760
    },
    {
      "epoch": 2.7017543859649122,
      "grad_norm": 0.11243995279073715,
      "learning_rate": 6.108605074829709e-06,
      "loss": 0.1178,
      "step": 770
    },
    {
      "epoch": 2.736842105263158,
      "grad_norm": 0.07883230596780777,
      "learning_rate": 4.781373142663004e-06,
      "loss": 0.1095,
      "step": 780
    },
    {
      "epoch": 2.736842105263158,
      "eval_loss": 0.11120390146970749,
      "eval_runtime": 24.9973,
      "eval_samples_per_second": 4.801,
      "eval_steps_per_second": 4.801,
      "step": 780
    },
    {
      "epoch": 2.7719298245614032,
      "grad_norm": 0.11856161803007126,
      "learning_rate": 3.613035605356463e-06,
      "loss": 0.1148,
      "step": 790
    },
    {
      "epoch": 2.807017543859649,
      "grad_norm": 0.08602704107761383,
      "learning_rate": 2.6055421054549613e-06,
      "loss": 0.1095,
      "step": 800
    }
  ],
  "logging_steps": 10,
  "max_steps": 855,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.123736288598528e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}