{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.2350383427739144,
      "epoch": 0.021458076283461188,
      "grad_norm": 1.4921875,
      "learning_rate": 6.282051282051282e-06,
      "loss": 2.3469,
      "mean_token_accuracy": 0.5447674424946308,
      "num_tokens": 13027304.0,
      "step": 50
    },
    {
      "entropy": 2.0760803684592246,
      "epoch": 0.042916152566922376,
      "grad_norm": 0.94140625,
      "learning_rate": 1.2692307692307693e-05,
      "loss": 2.0684,
      "mean_token_accuracy": 0.5785961505770684,
      "num_tokens": 26060152.0,
      "step": 100
    },
    {
      "entropy": 1.8538544979691505,
      "epoch": 0.06437422885038356,
      "grad_norm": 0.97265625,
      "learning_rate": 1.9102564102564103e-05,
      "loss": 1.8192,
      "mean_token_accuracy": 0.6143639494478702,
      "num_tokens": 39093818.0,
      "step": 150
    },
    {
      "entropy": 1.673665246963501,
      "epoch": 0.08583230513384475,
      "grad_norm": 0.85546875,
      "learning_rate": 2.551282051282051e-05,
      "loss": 1.6482,
      "mean_token_accuracy": 0.642510991692543,
      "num_tokens": 52123695.0,
      "step": 200
    },
    {
      "entropy": 1.5300365367531776,
      "epoch": 0.10729038141730594,
      "grad_norm": 0.87109375,
      "learning_rate": 2.9996212711496212e-05,
      "loss": 1.5103,
      "mean_token_accuracy": 0.6674637199938297,
      "num_tokens": 65156301.0,
      "step": 250
    },
    {
      "entropy": 1.4350718268752098,
      "epoch": 0.12874845770076712,
      "grad_norm": 0.90234375,
      "learning_rate": 2.992893632784019e-05,
      "loss": 1.4207,
      "mean_token_accuracy": 0.6824429173767567,
      "num_tokens": 78189155.0,
      "step": 300
    },
    {
      "entropy": 1.3636138886213303,
      "epoch": 0.1502065339842283,
      "grad_norm": 0.94921875,
      "learning_rate": 2.9777932337106776e-05,
      "loss": 1.3515,
      "mean_token_accuracy": 0.6942937538027764,
      "num_tokens": 91225387.0,
      "step": 350
    },
    {
      "entropy": 1.3275644582509996,
      "epoch": 0.1716646102676895,
      "grad_norm": 0.88671875,
      "learning_rate": 2.95440476317018e-05,
      "loss": 1.3165,
      "mean_token_accuracy": 0.7003772820532322,
      "num_tokens": 104264395.0,
      "step": 400
    },
    {
      "entropy": 1.2944868895411492,
      "epoch": 0.1931226865511507,
      "grad_norm": 0.8203125,
      "learning_rate": 2.9228593933122738e-05,
      "loss": 1.284,
      "mean_token_accuracy": 0.7061465001106262,
      "num_tokens": 117304006.0,
      "step": 450
    },
    {
      "entropy": 1.2659800139069557,
      "epoch": 0.2145807628346119,
      "grad_norm": 0.80859375,
      "learning_rate": 2.8833340435285732e-05,
      "loss": 1.2563,
      "mean_token_accuracy": 0.710605901479721,
      "num_tokens": 130330369.0,
      "step": 500
    },
    {
      "entropy": 1.2360862845182419,
      "epoch": 0.23603883911807305,
      "grad_norm": 0.81640625,
      "learning_rate": 2.8360503882159413e-05,
      "loss": 1.2268,
      "mean_token_accuracy": 0.7159372013807297,
      "num_tokens": 143357759.0,
      "step": 550
    },
    {
      "entropy": 1.21781676992774,
      "epoch": 0.25749691540153424,
      "grad_norm": 0.81640625,
      "learning_rate": 2.7812736135354192e-05,
      "loss": 1.2093,
      "mean_token_accuracy": 0.7191242362558842,
      "num_tokens": 156390933.0,
      "step": 600
    },
    {
      "entropy": 1.2027850279211998,
      "epoch": 0.27895499168499543,
      "grad_norm": 0.82421875,
      "learning_rate": 2.7193109301393063e-05,
      "loss": 1.1937,
      "mean_token_accuracy": 0.7216536393761634,
      "num_tokens": 169421819.0,
      "step": 650
    },
    {
      "entropy": 1.1848885342478752,
      "epoch": 0.3004130679684566,
      "grad_norm": 0.76171875,
      "learning_rate": 2.6505098502076353e-05,
      "loss": 1.1769,
      "mean_token_accuracy": 0.7247118213772774,
      "num_tokens": 182458556.0,
      "step": 700
    },
    {
      "entropy": 1.1768927517533303,
      "epoch": 0.3218711442519178,
      "grad_norm": 0.80859375,
      "learning_rate": 2.575256238457122e-05,
      "loss": 1.1685,
      "mean_token_accuracy": 0.7262762024998665,
      "num_tokens": 195498872.0,
      "step": 750
    },
    {
      "entropy": 1.1722314822673798,
      "epoch": 0.343329220535379,
      "grad_norm": 0.7734375,
      "learning_rate": 2.4939721480533394e-05,
      "loss": 1.1639,
      "mean_token_accuracy": 0.7267990556359291,
      "num_tokens": 208529244.0,
      "step": 800
    },
    {
      "entropy": 1.1594374303519726,
      "epoch": 0.3647872968188402,
      "grad_norm": 0.75,
      "learning_rate": 2.4071134535632042e-05,
      "loss": 1.1518,
      "mean_token_accuracy": 0.7289327536523342,
      "num_tokens": 221559622.0,
      "step": 850
    },
    {
      "entropy": 1.1491477774083614,
      "epoch": 0.3862453731023014,
      "grad_norm": 0.7578125,
      "learning_rate": 2.3151672942231506e-05,
      "loss": 1.1412,
      "mean_token_accuracy": 0.7309597064554691,
      "num_tokens": 234596401.0,
      "step": 900
    },
    {
      "entropy": 1.1420708937942983,
      "epoch": 0.4077034493857626,
      "grad_norm": 0.82421875,
      "learning_rate": 2.218649341862198e-05,
      "loss": 1.1342,
      "mean_token_accuracy": 0.7319549879431725,
      "num_tokens": 247619938.0,
      "step": 950
    },
    {
      "entropy": 1.1424958753585814,
      "epoch": 0.4291615256692238,
      "grad_norm": 0.80859375,
      "learning_rate": 2.1181009088025206e-05,
      "loss": 1.1349,
      "mean_token_accuracy": 0.7321111057698727,
      "num_tokens": 260654172.0,
      "step": 1000
    },
    {
      "entropy": 1.1404405610263348,
      "epoch": 0.45061960195268497,
      "grad_norm": 0.7890625,
      "learning_rate": 2.014085911957592e-05,
      "loss": 1.1327,
      "mean_token_accuracy": 0.7325018960237503,
      "num_tokens": 273684574.0,
      "step": 1050
    },
    {
      "entropy": 1.1398129032552242,
      "epoch": 0.4720776782361461,
      "grad_norm": 0.72265625,
      "learning_rate": 1.907187710154509e-05,
      "loss": 1.1316,
      "mean_token_accuracy": 0.7324371525645256,
      "num_tokens": 286722109.0,
      "step": 1100
    },
    {
      "entropy": 1.1286205492913723,
      "epoch": 0.4935357545196073,
      "grad_norm": 0.71875,
      "learning_rate": 1.7980058324180654e-05,
      "loss": 1.1208,
      "mean_token_accuracy": 0.7347702208161354,
      "num_tokens": 299759049.0,
      "step": 1150
    },
    {
      "entropy": 1.127575832605362,
      "epoch": 0.5149938308030685,
      "grad_norm": 0.7421875,
      "learning_rate": 1.687152615565724e-05,
      "loss": 1.119,
      "mean_token_accuracy": 0.7347547617554665,
      "num_tokens": 312785989.0,
      "step": 1200
    },
    {
      "entropy": 1.120200854986906,
      "epoch": 0.5364519070865297,
      "grad_norm": 0.72265625,
      "learning_rate": 1.5752497699712207e-05,
      "loss": 1.1124,
      "mean_token_accuracy": 0.736199368685484,
      "num_tokens": 325818163.0,
      "step": 1250
    },
    {
      "entropy": 1.1245441977679729,
      "epoch": 0.5579099833699909,
      "grad_norm": 0.76171875,
      "learning_rate": 1.4629248927574217e-05,
      "loss": 1.1161,
      "mean_token_accuracy": 0.7352614414691925,
      "num_tokens": 338855461.0,
      "step": 1300
    },
    {
      "entropy": 1.1184861905872823,
      "epoch": 0.5793680596534521,
      "grad_norm": 0.7109375,
      "learning_rate": 1.3508079479738632e-05,
      "loss": 1.1107,
      "mean_token_accuracy": 0.7365366728603839,
      "num_tokens": 351888193.0,
      "step": 1350
    },
    {
      "entropy": 1.12065178334713,
      "epoch": 0.6008261359369133,
      "grad_norm": 0.6796875,
      "learning_rate": 1.2395277334996045e-05,
      "loss": 1.1127,
      "mean_token_accuracy": 0.735972375869751,
      "num_tokens": 364920249.0,
      "step": 1400
    },
    {
      "entropy": 1.1175432482361793,
      "epoch": 0.6222842122203744,
      "grad_norm": 0.73046875,
      "learning_rate": 1.1297083544864386e-05,
      "loss": 1.1094,
      "mean_token_accuracy": 0.7365812124311923,
      "num_tokens": 377954848.0,
      "step": 1450
    },
    {
      "entropy": 1.1152825058996678,
      "epoch": 0.6437422885038356,
      "grad_norm": 0.7265625,
      "learning_rate": 1.0219657231208455e-05,
      "loss": 1.1073,
      "mean_token_accuracy": 0.7370956964790821,
      "num_tokens": 390982547.0,
      "step": 1500
    },
    {
      "entropy": 1.1113113643229007,
      "epoch": 0.6652003647872968,
      "grad_norm": 0.70703125,
      "learning_rate": 9.1690410433546e-06,
      "loss": 1.1031,
      "mean_token_accuracy": 0.7381323325634003,
      "num_tokens": 404012732.0,
      "step": 1550
    },
    {
      "entropy": 1.117246091067791,
      "epoch": 0.686658441070758,
      "grad_norm": 0.671875,
      "learning_rate": 8.151127268431263e-06,
      "loss": 1.1095,
      "mean_token_accuracy": 0.7364383706450462,
      "num_tokens": 417050367.0,
      "step": 1600
    },
    {
      "entropy": 1.1141446343064307,
      "epoch": 0.7081165173542192,
      "grad_norm": 0.6796875,
      "learning_rate": 7.171624785002444e-06,
      "loss": 1.1054,
      "mean_token_accuracy": 0.7371199345588684,
      "num_tokens": 430085128.0,
      "step": 1650
    },
    {
      "entropy": 1.1189921294152736,
      "epoch": 0.7295745936376804,
      "grad_norm": 0.67578125,
      "learning_rate": 6.236027045331792e-06,
      "loss": 1.1109,
      "mean_token_accuracy": 0.7362688648700714,
      "num_tokens": 443116587.0,
      "step": 1700
    },
    {
      "entropy": 1.1170032300055026,
      "epoch": 0.7510326699211416,
      "grad_norm": 0.671875,
      "learning_rate": 5.349581265845804e-06,
      "loss": 1.1084,
      "mean_token_accuracy": 0.7368242286145688,
      "num_tokens": 456150190.0,
      "step": 1750
    },
    {
      "entropy": 1.1152006842195987,
      "epoch": 0.7724907462046028,
      "grad_norm": 0.69921875,
      "learning_rate": 4.517258998588792e-06,
      "loss": 1.1067,
      "mean_token_accuracy": 0.7371417485177517,
      "num_tokens": 469187645.0,
      "step": 1800
    },
    {
      "entropy": 1.1084230598807334,
      "epoch": 0.793948822488064,
      "grad_norm": 0.671875,
      "learning_rate": 3.743728248716951e-06,
      "loss": 1.0996,
      "mean_token_accuracy": 0.7385361175239086,
      "num_tokens": 482216628.0,
      "step": 1850
    },
    {
      "entropy": 1.1149806365370751,
      "epoch": 0.8154068987715252,
      "grad_norm": 0.71875,
      "learning_rate": 3.033327294407988e-06,
      "loss": 1.1066,
      "mean_token_accuracy": 0.7370860169827939,
      "num_tokens": 495240948.0,
      "step": 1900
    },
    {
      "entropy": 1.1116354890167712,
      "epoch": 0.8368649750549864,
      "grad_norm": 0.6640625,
      "learning_rate": 2.390040356015151e-06,
      "loss": 1.1037,
      "mean_token_accuracy": 0.7381251747906208,
      "num_tokens": 508270012.0,
      "step": 1950
    },
    {
      "entropy": 1.1099542181193829,
      "epoch": 0.8583230513384476,
      "grad_norm": 0.6953125,
      "learning_rate": 1.817475250922957e-06,
      "loss": 1.1013,
      "mean_token_accuracy": 0.7383160504698754,
      "num_tokens": 521303590.0,
      "step": 2000
    },
    {
      "entropy": 1.1151345440745353,
      "epoch": 0.8797811276219087,
      "grad_norm": 0.6875,
      "learning_rate": 1.318843159425332e-06,
      "loss": 1.1069,
      "mean_token_accuracy": 0.7368960759043693,
      "num_tokens": 534336739.0,
      "step": 2050
    },
    {
      "entropy": 1.1179504051804543,
      "epoch": 0.9012392039053699,
      "grad_norm": 0.69921875,
      "learning_rate": 8.969406151075494e-07,
      "loss": 1.1109,
      "mean_token_accuracy": 0.7363792227208614,
      "num_tokens": 547357092.0,
      "step": 2100
    },
    {
      "entropy": 1.1139133739471436,
      "epoch": 0.9226972801888311,
      "grad_norm": 0.6796875,
      "learning_rate": 5.541338207371605e-07,
      "loss": 1.106,
      "mean_token_accuracy": 0.7373194192349911,
      "num_tokens": 560392664.0,
      "step": 2150
    },
    {
      "entropy": 1.1176364059746264,
      "epoch": 0.9441553564722922,
      "grad_norm": 0.671875,
      "learning_rate": 2.923453776269558e-07,
      "loss": 1.1091,
      "mean_token_accuracy": 0.7365114194154739,
      "num_tokens": 573423859.0,
      "step": 2200
    },
    {
      "entropy": 1.1216451613605023,
      "epoch": 0.9656134327557534,
      "grad_norm": 0.69921875,
      "learning_rate": 1.130435028970872e-07,
      "loss": 1.113,
      "mean_token_accuracy": 0.7358175349235535,
      "num_tokens": 586459673.0,
      "step": 2250
    },
    {
      "entropy": 1.1107782909274102,
      "epoch": 0.9870715090392146,
      "grad_norm": 0.68359375,
      "learning_rate": 1.7233795110367935e-08,
      "loss": 1.1021,
      "mean_token_accuracy": 0.7379112888872623,
      "num_tokens": 599492881.0,
      "step": 2300
    }
  ],
  "logging_steps": 50,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.135994653597762e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}