{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 351,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0861141011840689,
      "grad_norm": 333.3435363769531,
      "learning_rate": 9e-07,
      "loss": 9.803564453125,
      "num_input_tokens_seen": 90208,
      "step": 10,
      "train_runtime": 96.6186,
      "train_tokens_per_second": 933.651
    },
    {
      "epoch": 0.1722282023681378,
      "grad_norm": 63.6615104675293,
      "learning_rate": 1.9e-06,
      "loss": 7.278943634033203,
      "num_input_tokens_seen": 180128,
      "step": 20,
      "train_runtime": 161.1293,
      "train_tokens_per_second": 1117.909
    },
    {
      "epoch": 0.25834230355220666,
      "grad_norm": 49.02418899536133,
      "learning_rate": 2.9e-06,
      "loss": 4.277183914184571,
      "num_input_tokens_seen": 270720,
      "step": 30,
      "train_runtime": 225.3607,
      "train_tokens_per_second": 1201.274
    },
    {
      "epoch": 0.3444564047362756,
      "grad_norm": 17.75627326965332,
      "learning_rate": 3.9e-06,
      "loss": 1.295894241333008,
      "num_input_tokens_seen": 361952,
      "step": 40,
      "train_runtime": 289.3937,
      "train_tokens_per_second": 1250.725
    },
    {
      "epoch": 0.43057050592034446,
      "grad_norm": 20.77004051208496,
      "learning_rate": 4.9000000000000005e-06,
      "loss": 0.3324465751647949,
      "num_input_tokens_seen": 452352,
      "step": 50,
      "train_runtime": 353.7775,
      "train_tokens_per_second": 1278.634
    },
    {
      "epoch": 0.5166846071044133,
      "grad_norm": 11.49393081665039,
      "learning_rate": 5.9e-06,
      "loss": 0.257261061668396,
      "num_input_tokens_seen": 542400,
      "step": 60,
      "train_runtime": 417.9525,
      "train_tokens_per_second": 1297.755
    },
    {
      "epoch": 0.6027987082884823,
      "grad_norm": 7.95831298828125,
      "learning_rate": 6.900000000000001e-06,
      "loss": 0.22105672359466552,
      "num_input_tokens_seen": 633504,
      "step": 70,
      "train_runtime": 481.94,
      "train_tokens_per_second": 1314.487
    },
    {
      "epoch": 0.6889128094725512,
      "grad_norm": 11.869803428649902,
      "learning_rate": 7.9e-06,
      "loss": 0.21478445529937745,
      "num_input_tokens_seen": 723808,
      "step": 80,
      "train_runtime": 545.4453,
      "train_tokens_per_second": 1327.004
    },
    {
      "epoch": 0.77502691065662,
      "grad_norm": 5.659975051879883,
      "learning_rate": 8.9e-06,
      "loss": 0.18611286878585814,
      "num_input_tokens_seen": 815392,
      "step": 90,
      "train_runtime": 609.3714,
      "train_tokens_per_second": 1338.087
    },
    {
      "epoch": 0.8611410118406889,
      "grad_norm": 4.791771411895752,
      "learning_rate": 9.900000000000002e-06,
      "loss": 0.1616648554801941,
      "num_input_tokens_seen": 905760,
      "step": 100,
      "train_runtime": 673.6631,
      "train_tokens_per_second": 1344.53
    },
    {
      "epoch": 0.9472551130247578,
      "grad_norm": 5.305026054382324,
      "learning_rate": 1.09e-05,
      "loss": 0.16810319423675538,
      "num_input_tokens_seen": 995392,
      "step": 110,
      "train_runtime": 738.0244,
      "train_tokens_per_second": 1348.725
    },
    {
      "epoch": 1.0258342303552206,
      "grad_norm": 10.043594360351562,
      "learning_rate": 1.19e-05,
      "loss": 0.15800862312316893,
      "num_input_tokens_seen": 1077536,
      "step": 120,
      "train_runtime": 796.8721,
      "train_tokens_per_second": 1352.207
    },
    {
      "epoch": 1.1119483315392895,
      "grad_norm": 3.663066864013672,
      "learning_rate": 1.29e-05,
      "loss": 0.1582618236541748,
      "num_input_tokens_seen": 1168640,
      "step": 130,
      "train_runtime": 861.0741,
      "train_tokens_per_second": 1357.189
    },
    {
      "epoch": 1.1980624327233584,
      "grad_norm": 2.456631898880005,
      "learning_rate": 1.3900000000000002e-05,
      "loss": 0.14071855545043946,
      "num_input_tokens_seen": 1259296,
      "step": 140,
      "train_runtime": 925.3123,
      "train_tokens_per_second": 1360.942
    },
    {
      "epoch": 1.2841765339074274,
      "grad_norm": 5.205492973327637,
      "learning_rate": 1.49e-05,
      "loss": 0.15321035385131837,
      "num_input_tokens_seen": 1348864,
      "step": 150,
      "train_runtime": 989.3708,
      "train_tokens_per_second": 1363.355
    },
    {
      "epoch": 1.3702906350914963,
      "grad_norm": 6.281779766082764,
      "learning_rate": 1.59e-05,
      "loss": 0.1474855661392212,
      "num_input_tokens_seen": 1438816,
      "step": 160,
      "train_runtime": 1053.5038,
      "train_tokens_per_second": 1365.744
    },
    {
      "epoch": 1.456404736275565,
      "grad_norm": 4.965813159942627,
      "learning_rate": 1.69e-05,
      "loss": 0.1475322961807251,
      "num_input_tokens_seen": 1529472,
      "step": 170,
      "train_runtime": 1117.898,
      "train_tokens_per_second": 1368.168
    },
    {
      "epoch": 1.542518837459634,
      "grad_norm": 2.566067695617676,
      "learning_rate": 1.79e-05,
      "loss": 0.13842158317565917,
      "num_input_tokens_seen": 1619680,
      "step": 180,
      "train_runtime": 1182.1227,
      "train_tokens_per_second": 1370.145
    },
    {
      "epoch": 1.6286329386437028,
      "grad_norm": 2.7559897899627686,
      "learning_rate": 1.8900000000000002e-05,
      "loss": 0.14058760404586793,
      "num_input_tokens_seen": 1708512,
      "step": 190,
      "train_runtime": 1245.9096,
      "train_tokens_per_second": 1371.297
    },
    {
      "epoch": 1.7147470398277718,
      "grad_norm": 2.7535946369171143,
      "learning_rate": 1.9900000000000003e-05,
      "loss": 0.14503936767578124,
      "num_input_tokens_seen": 1799360,
      "step": 200,
      "train_runtime": 1310.3896,
      "train_tokens_per_second": 1373.149
    },
    {
      "epoch": 1.8008611410118407,
      "grad_norm": 2.8212454319000244,
      "learning_rate": 2.09e-05,
      "loss": 0.14306368827819824,
      "num_input_tokens_seen": 1890528,
      "step": 210,
      "train_runtime": 1374.3589,
      "train_tokens_per_second": 1375.571
    },
    {
      "epoch": 1.8869752421959096,
      "grad_norm": 3.7724499702453613,
      "learning_rate": 2.19e-05,
      "loss": 0.1484434723854065,
      "num_input_tokens_seen": 1981440,
      "step": 220,
      "train_runtime": 1438.735,
      "train_tokens_per_second": 1377.21
    },
    {
      "epoch": 1.9730893433799785,
      "grad_norm": 4.947230815887451,
      "learning_rate": 2.29e-05,
      "loss": 0.1293976902961731,
      "num_input_tokens_seen": 2070560,
      "step": 230,
      "train_runtime": 1502.503,
      "train_tokens_per_second": 1378.074
    },
    {
      "epoch": 2.051668460710441,
      "grad_norm": 2.538516044616699,
      "learning_rate": 2.39e-05,
      "loss": 0.0999522089958191,
      "num_input_tokens_seen": 2153312,
      "step": 240,
      "train_runtime": 1561.7539,
      "train_tokens_per_second": 1378.778
    },
    {
      "epoch": 2.13778256189451,
      "grad_norm": 3.129701614379883,
      "learning_rate": 2.4900000000000002e-05,
      "loss": 0.08672967553138733,
      "num_input_tokens_seen": 2243680,
      "step": 250,
      "train_runtime": 1626.1253,
      "train_tokens_per_second": 1379.771
    },
    {
      "epoch": 2.223896663078579,
      "grad_norm": 1.4860609769821167,
      "learning_rate": 2.5900000000000003e-05,
      "loss": 0.08354443907737732,
      "num_input_tokens_seen": 2335520,
      "step": 260,
      "train_runtime": 1690.2808,
      "train_tokens_per_second": 1381.735
    },
    {
      "epoch": 2.310010764262648,
      "grad_norm": 4.073873043060303,
      "learning_rate": 2.6900000000000003e-05,
      "loss": 0.09892970323562622,
      "num_input_tokens_seen": 2425280,
      "step": 270,
      "train_runtime": 1754.0051,
      "train_tokens_per_second": 1382.71
    },
    {
      "epoch": 2.396124865446717,
      "grad_norm": 1.3554434776306152,
      "learning_rate": 2.7900000000000004e-05,
      "loss": 0.11728144884109497,
      "num_input_tokens_seen": 2516192,
      "step": 280,
      "train_runtime": 1818.4362,
      "train_tokens_per_second": 1383.712
    },
    {
      "epoch": 2.482238966630786,
      "grad_norm": 2.801039218902588,
      "learning_rate": 2.8899999999999998e-05,
      "loss": 0.10752675533294678,
      "num_input_tokens_seen": 2606944,
      "step": 290,
      "train_runtime": 1882.8892,
      "train_tokens_per_second": 1384.545
    },
    {
      "epoch": 2.5683530678148547,
      "grad_norm": 3.3027260303497314,
      "learning_rate": 2.9900000000000002e-05,
      "loss": 0.09044331908226014,
      "num_input_tokens_seen": 2697472,
      "step": 300,
      "train_runtime": 1946.9576,
      "train_tokens_per_second": 1385.481
    },
    {
      "epoch": 2.6544671689989237,
      "grad_norm": 2.025606393814087,
      "learning_rate": 3.09e-05,
      "loss": 0.11651453971862794,
      "num_input_tokens_seen": 2788032,
      "step": 310,
      "train_runtime": 2011.3616,
      "train_tokens_per_second": 1386.142
    },
    {
      "epoch": 2.7405812701829926,
      "grad_norm": 1.6964479684829712,
      "learning_rate": 3.19e-05,
      "loss": 0.0845784842967987,
      "num_input_tokens_seen": 2878336,
      "step": 320,
      "train_runtime": 2075.4824,
      "train_tokens_per_second": 1386.827
    },
    {
      "epoch": 2.8266953713670615,
      "grad_norm": 1.75762140750885,
      "learning_rate": 3.29e-05,
      "loss": 0.094532310962677,
      "num_input_tokens_seen": 2968768,
      "step": 330,
      "train_runtime": 2139.8459,
      "train_tokens_per_second": 1387.375
    },
    {
      "epoch": 2.91280947255113,
      "grad_norm": 4.5564165115356445,
      "learning_rate": 3.3900000000000004e-05,
      "loss": 0.10398908853530883,
      "num_input_tokens_seen": 3058112,
      "step": 340,
      "train_runtime": 2203.6666,
      "train_tokens_per_second": 1387.738
    },
    {
      "epoch": 2.9989235737351994,
      "grad_norm": 1.972362756729126,
      "learning_rate": 3.49e-05,
      "loss": 0.0984138309955597,
      "num_input_tokens_seen": 3148448,
      "step": 350,
      "train_runtime": 2267.9059,
      "train_tokens_per_second": 1388.262
    },
    {
      "epoch": 3.0,
      "num_input_tokens_seen": 3149504,
      "step": 351,
      "total_flos": 1.677056662170501e+17,
      "train_loss": 0.7731951655782865,
      "train_runtime": 2270.5085,
      "train_samples_per_second": 4.91,
      "train_steps_per_second": 0.155
    }
  ],
  "logging_steps": 10,
  "max_steps": 351,
  "num_input_tokens_seen": 3149504,
  "num_train_epochs": 3,
  "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": 1.677056662170501e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}