{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1395,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.021528525296017224,
      "grad_norm": 0.0001240880519617349,
      "learning_rate": 2.9996919063517692e-05,
      "loss": 0.6177,
      "num_input_tokens_seen": 37152,
      "step": 10,
      "train_runtime": 20.7249,
      "train_tokens_per_second": 1792.63
    },
    {
      "epoch": 0.04305705059203445,
      "grad_norm": 0.00010908672265941277,
      "learning_rate": 2.9986270537572827e-05,
      "loss": 0.1472,
      "num_input_tokens_seen": 73856,
      "step": 20,
      "train_runtime": 32.8295,
      "train_tokens_per_second": 2249.682
    },
    {
      "epoch": 0.06458557588805167,
      "grad_norm": 2.9509448722819798e-05,
      "learning_rate": 2.9968021784983436e-05,
      "loss": 0.1295,
      "num_input_tokens_seen": 111232,
      "step": 30,
      "train_runtime": 45.5682,
      "train_tokens_per_second": 2441.002
    },
    {
      "epoch": 0.0861141011840689,
      "grad_norm": 3.8006528484402224e-05,
      "learning_rate": 2.994218206053077e-05,
      "loss": 0.1432,
      "num_input_tokens_seen": 148992,
      "step": 40,
      "train_runtime": 57.8544,
      "train_tokens_per_second": 2575.294
    },
    {
      "epoch": 0.10764262648008611,
      "grad_norm": 3.5578465030994266e-05,
      "learning_rate": 2.9908764468726846e-05,
      "loss": 0.1252,
      "num_input_tokens_seen": 186176,
      "step": 50,
      "train_runtime": 70.0018,
      "train_tokens_per_second": 2659.588
    },
    {
      "epoch": 0.12917115177610333,
      "grad_norm": 5.3873140132054687e-05,
      "learning_rate": 2.986778595716853e-05,
      "loss": 0.129,
      "num_input_tokens_seen": 223840,
      "step": 60,
      "train_runtime": 82.2467,
      "train_tokens_per_second": 2721.569
    },
    {
      "epoch": 0.15069967707212056,
      "grad_norm": 4.8572950618108734e-05,
      "learning_rate": 2.981926730794264e-05,
      "loss": 0.1346,
      "num_input_tokens_seen": 260960,
      "step": 70,
      "train_runtime": 94.4188,
      "train_tokens_per_second": 2763.857
    },
    {
      "epoch": 0.1722282023681378,
      "grad_norm": 3.341471528983675e-05,
      "learning_rate": 2.9763233127086394e-05,
      "loss": 0.1078,
      "num_input_tokens_seen": 298048,
      "step": 80,
      "train_runtime": 106.5982,
      "train_tokens_per_second": 2795.994
    },
    {
      "epoch": 0.193756727664155,
      "grad_norm": 3.282168836449273e-05,
      "learning_rate": 2.9699711832108555e-05,
      "loss": 0.118,
      "num_input_tokens_seen": 334496,
      "step": 90,
      "train_runtime": 118.5321,
      "train_tokens_per_second": 2821.987
    },
    {
      "epoch": 0.21528525296017223,
      "grad_norm": 3.009026295330841e-05,
      "learning_rate": 2.9628735637577612e-05,
      "loss": 0.1273,
      "num_input_tokens_seen": 371712,
      "step": 100,
      "train_runtime": 130.7227,
      "train_tokens_per_second": 2843.516
    },
    {
      "epoch": 0.23681377825618946,
      "grad_norm": 6.089722228352912e-05,
      "learning_rate": 2.955034053878428e-05,
      "loss": 0.1077,
      "num_input_tokens_seen": 408512,
      "step": 110,
      "train_runtime": 142.7614,
      "train_tokens_per_second": 2861.503
    },
    {
      "epoch": 0.25834230355220666,
      "grad_norm": 4.01161887566559e-05,
      "learning_rate": 2.9464566293486653e-05,
      "loss": 0.1102,
      "num_input_tokens_seen": 446048,
      "step": 120,
      "train_runtime": 155.002,
      "train_tokens_per_second": 2877.691
    },
    {
      "epoch": 0.2798708288482239,
      "grad_norm": 3.839691271423362e-05,
      "learning_rate": 2.9371456401747193e-05,
      "loss": 0.1049,
      "num_input_tokens_seen": 483616,
      "step": 130,
      "train_runtime": 167.2469,
      "train_tokens_per_second": 2891.629
    },
    {
      "epoch": 0.3013993541442411,
      "grad_norm": 7.051438296912238e-05,
      "learning_rate": 2.9271058083871864e-05,
      "loss": 0.1224,
      "num_input_tokens_seen": 520768,
      "step": 140,
      "train_runtime": 179.3786,
      "train_tokens_per_second": 2903.178
    },
    {
      "epoch": 0.32292787944025836,
      "grad_norm": 2.7564614356379025e-05,
      "learning_rate": 2.916342225646254e-05,
      "loss": 0.0955,
      "num_input_tokens_seen": 558240,
      "step": 150,
      "train_runtime": 191.7012,
      "train_tokens_per_second": 2912.031
    },
    {
      "epoch": 0.3444564047362756,
      "grad_norm": 4.490939318202436e-05,
      "learning_rate": 2.904860350659487e-05,
      "loss": 0.0988,
      "num_input_tokens_seen": 595648,
      "step": 160,
      "train_runtime": 203.9607,
      "train_tokens_per_second": 2920.405
    },
    {
      "epoch": 0.36598493003229277,
      "grad_norm": 5.966637036181055e-05,
      "learning_rate": 2.8926660064134637e-05,
      "loss": 0.1308,
      "num_input_tokens_seen": 632352,
      "step": 170,
      "train_runtime": 216.0429,
      "train_tokens_per_second": 2926.974
    },
    {
      "epoch": 0.38751345532831,
      "grad_norm": 2.6004610845120624e-05,
      "learning_rate": 2.8797653772206726e-05,
      "loss": 0.0988,
      "num_input_tokens_seen": 669440,
      "step": 180,
      "train_runtime": 228.1674,
      "train_tokens_per_second": 2933.986
    },
    {
      "epoch": 0.40904198062432723,
      "grad_norm": 3.3529446227476e-05,
      "learning_rate": 2.8661650055831645e-05,
      "loss": 0.1057,
      "num_input_tokens_seen": 706784,
      "step": 190,
      "train_runtime": 240.3668,
      "train_tokens_per_second": 2940.439
    },
    {
      "epoch": 0.43057050592034446,
      "grad_norm": 6.642789230681956e-05,
      "learning_rate": 2.8518717888745486e-05,
      "loss": 0.0889,
      "num_input_tokens_seen": 744480,
      "step": 200,
      "train_runtime": 252.6752,
      "train_tokens_per_second": 2946.391
    },
    {
      "epoch": 0.4520990312163617,
      "grad_norm": 2.6275289201294072e-05,
      "learning_rate": 2.836892975842015e-05,
      "loss": 0.0923,
      "num_input_tokens_seen": 781632,
      "step": 210,
      "train_runtime": 264.825,
      "train_tokens_per_second": 2951.504
    },
    {
      "epoch": 0.4736275565123789,
      "grad_norm": 5.466080619953573e-05,
      "learning_rate": 2.8212361629301655e-05,
      "loss": 0.1053,
      "num_input_tokens_seen": 819264,
      "step": 220,
      "train_runtime": 277.0866,
      "train_tokens_per_second": 2956.707
    },
    {
      "epoch": 0.4951560818083961,
      "grad_norm": 2.5802628442761488e-05,
      "learning_rate": 2.8049092904285033e-05,
      "loss": 0.0993,
      "num_input_tokens_seen": 855936,
      "step": 230,
      "train_runtime": 289.1761,
      "train_tokens_per_second": 2959.913
    },
    {
      "epoch": 0.5166846071044133,
      "grad_norm": 3.3568783692317083e-05,
      "learning_rate": 2.7879206384445526e-05,
      "loss": 0.0998,
      "num_input_tokens_seen": 893536,
      "step": 240,
      "train_runtime": 301.3716,
      "train_tokens_per_second": 2964.898
    },
    {
      "epoch": 0.5382131324004306,
      "grad_norm": 1.9844068447127938e-05,
      "learning_rate": 2.7702788227046286e-05,
      "loss": 0.0949,
      "num_input_tokens_seen": 930272,
      "step": 250,
      "train_runtime": 313.4265,
      "train_tokens_per_second": 2968.071
    },
    {
      "epoch": 0.5597416576964478,
      "grad_norm": 2.8381900847307406e-05,
      "learning_rate": 2.751992790184413e-05,
      "loss": 0.0934,
      "num_input_tokens_seen": 968768,
      "step": 260,
      "train_runtime": 325.8814,
      "train_tokens_per_second": 2972.762
    },
    {
      "epoch": 0.581270182992465,
      "grad_norm": 4.446608363650739e-05,
      "learning_rate": 2.733071814571527e-05,
      "loss": 0.1071,
      "num_input_tokens_seen": 1005888,
      "step": 270,
      "train_runtime": 338.0393,
      "train_tokens_per_second": 2975.654
    },
    {
      "epoch": 0.6027987082884823,
      "grad_norm": 2.2868445739732124e-05,
      "learning_rate": 2.7135254915624213e-05,
      "loss": 0.0846,
      "num_input_tokens_seen": 1043744,
      "step": 280,
      "train_runtime": 350.3703,
      "train_tokens_per_second": 2978.974
    },
    {
      "epoch": 0.6243272335844995,
      "grad_norm": 5.735125523642637e-05,
      "learning_rate": 2.693363733995954e-05,
      "loss": 0.1123,
      "num_input_tokens_seen": 1081920,
      "step": 290,
      "train_runtime": 362.7636,
      "train_tokens_per_second": 2982.438
    },
    {
      "epoch": 0.6458557588805167,
      "grad_norm": 6.513342668768018e-05,
      "learning_rate": 2.672596766826138e-05,
      "loss": 0.085,
      "num_input_tokens_seen": 1118784,
      "step": 300,
      "train_runtime": 374.9594,
      "train_tokens_per_second": 2983.747
    },
    {
      "epoch": 0.667384284176534,
      "grad_norm": 4.052193617098965e-05,
      "learning_rate": 2.6512351219365927e-05,
      "loss": 0.0894,
      "num_input_tokens_seen": 1155392,
      "step": 310,
      "train_runtime": 387.0264,
      "train_tokens_per_second": 2985.306
    },
    {
      "epoch": 0.6889128094725512,
      "grad_norm": 1.752378921082709e-05,
      "learning_rate": 2.6292896327993467e-05,
      "loss": 0.1046,
      "num_input_tokens_seen": 1192768,
      "step": 320,
      "train_runtime": 399.3125,
      "train_tokens_per_second": 2987.054
    },
    {
      "epoch": 0.7104413347685683,
      "grad_norm": 3.3892840292537585e-05,
      "learning_rate": 2.606771428980681e-05,
      "loss": 0.0896,
      "num_input_tokens_seen": 1230464,
      "step": 330,
      "train_runtime": 411.6218,
      "train_tokens_per_second": 2989.307
    },
    {
      "epoch": 0.7319698600645855,
      "grad_norm": 2.5454823116888292e-05,
      "learning_rate": 2.5836919304968184e-05,
      "loss": 0.0956,
      "num_input_tokens_seen": 1267520,
      "step": 340,
      "train_runtime": 423.8081,
      "train_tokens_per_second": 2990.787
    },
    {
      "epoch": 0.7534983853606028,
      "grad_norm": 3.811945134657435e-05,
      "learning_rate": 2.5600628420223057e-05,
      "loss": 0.0936,
      "num_input_tokens_seen": 1306592,
      "step": 350,
      "train_runtime": 436.408,
      "train_tokens_per_second": 2993.969
    },
    {
      "epoch": 0.77502691065662,
      "grad_norm": 3.069249578402378e-05,
      "learning_rate": 2.535896146954037e-05,
      "loss": 0.1003,
      "num_input_tokens_seen": 1343840,
      "step": 360,
      "train_runtime": 448.6377,
      "train_tokens_per_second": 2995.379
    },
    {
      "epoch": 0.7965554359526372,
      "grad_norm": 2.3971600967342965e-05,
      "learning_rate": 2.5112041013339212e-05,
      "loss": 0.0935,
      "num_input_tokens_seen": 1382816,
      "step": 370,
      "train_runtime": 461.2283,
      "train_tokens_per_second": 2998.116
    },
    {
      "epoch": 0.8180839612486545,
      "grad_norm": 2.4271197617053986e-05,
      "learning_rate": 2.48599922763328e-05,
      "loss": 0.0842,
      "num_input_tokens_seen": 1420352,
      "step": 380,
      "train_runtime": 473.4256,
      "train_tokens_per_second": 3000.159
    },
    {
      "epoch": 0.8396124865446717,
      "grad_norm": 1.868541949079372e-05,
      "learning_rate": 2.4602943084021286e-05,
      "loss": 0.1066,
      "num_input_tokens_seen": 1457280,
      "step": 390,
      "train_runtime": 485.5904,
      "train_tokens_per_second": 3001.048
    },
    {
      "epoch": 0.8611410118406889,
      "grad_norm": 3.90959539799951e-05,
      "learning_rate": 2.434102379786553e-05,
      "loss": 0.0703,
      "num_input_tokens_seen": 1493984,
      "step": 400,
      "train_runtime": 497.6537,
      "train_tokens_per_second": 3002.056
    },
    {
      "epoch": 0.8826695371367062,
      "grad_norm": 3.146451854263432e-05,
      "learning_rate": 2.4074367249174864e-05,
      "loss": 0.0734,
      "num_input_tokens_seen": 1531392,
      "step": 410,
      "train_runtime": 509.8552,
      "train_tokens_per_second": 3003.582
    },
    {
      "epoch": 0.9041980624327234,
      "grad_norm": 4.228637044434436e-05,
      "learning_rate": 2.380310867174219e-05,
      "loss": 0.0802,
      "num_input_tokens_seen": 1568992,
      "step": 420,
      "train_runtime": 522.1162,
      "train_tokens_per_second": 3005.063
    },
    {
      "epoch": 0.9257265877287406,
      "grad_norm": 2.047202542598825e-05,
      "learning_rate": 2.3527385633260724e-05,
      "loss": 0.0864,
      "num_input_tokens_seen": 1606144,
      "step": 430,
      "train_runtime": 534.1654,
      "train_tokens_per_second": 3006.829
    },
    {
      "epoch": 0.9472551130247578,
      "grad_norm": 3.2742798794060946e-05,
      "learning_rate": 2.3247337965557097e-05,
      "loss": 0.0751,
      "num_input_tokens_seen": 1641696,
      "step": 440,
      "train_runtime": 545.9356,
      "train_tokens_per_second": 3007.124
    },
    {
      "epoch": 0.9687836383207751,
      "grad_norm": 1.9717903342097998e-05,
      "learning_rate": 2.2963107693676192e-05,
      "loss": 0.0804,
      "num_input_tokens_seen": 1678816,
      "step": 450,
      "train_runtime": 558.0857,
      "train_tokens_per_second": 3008.169
    },
    {
      "epoch": 0.9903121636167922,
      "grad_norm": 3.3491553040221334e-05,
      "learning_rate": 2.2674838963853738e-05,
      "loss": 0.0781,
      "num_input_tokens_seen": 1715808,
      "step": 460,
      "train_runtime": 570.1739,
      "train_tokens_per_second": 3009.271
    },
    {
      "epoch": 1.0107642626480087,
      "grad_norm": 7.698799890931696e-05,
      "learning_rate": 2.2382677970413097e-05,
      "loss": 0.0699,
      "num_input_tokens_seen": 1751552,
      "step": 470,
      "train_runtime": 581.7157,
      "train_tokens_per_second": 3011.01
    },
    {
      "epoch": 1.0322927879440258,
      "grad_norm": 2.397145544819068e-05,
      "learning_rate": 2.2086772881623455e-05,
      "loss": 0.0794,
      "num_input_tokens_seen": 1788544,
      "step": 480,
      "train_runtime": 593.7982,
      "train_tokens_per_second": 3012.04
    },
    {
      "epoch": 1.0538213132400431,
      "grad_norm": 3.573806679924019e-05,
      "learning_rate": 2.1787273764556827e-05,
      "loss": 0.0573,
      "num_input_tokens_seen": 1826496,
      "step": 490,
      "train_runtime": 606.1316,
      "train_tokens_per_second": 3013.365
    },
    {
      "epoch": 1.0753498385360603,
      "grad_norm": 3.507183646433987e-05,
      "learning_rate": 2.1484332508982167e-05,
      "loss": 0.0726,
      "num_input_tokens_seen": 1864544,
      "step": 500,
      "train_runtime": 618.4797,
      "train_tokens_per_second": 3014.721
    },
    {
      "epoch": 1.0968783638320776,
      "grad_norm": 2.8500042390078306e-05,
      "learning_rate": 2.11781027503351e-05,
      "loss": 0.0869,
      "num_input_tokens_seen": 1901856,
      "step": 510,
      "train_runtime": 630.7014,
      "train_tokens_per_second": 3015.462
    },
    {
      "epoch": 1.1184068891280947,
      "grad_norm": 3.343845310155302e-05,
      "learning_rate": 2.0868739791802312e-05,
      "loss": 0.081,
      "num_input_tokens_seen": 1939264,
      "step": 520,
      "train_runtime": 642.8909,
      "train_tokens_per_second": 3016.474
    },
    {
      "epoch": 1.1399354144241118,
      "grad_norm": 5.641104144160636e-05,
      "learning_rate": 2.0556400525560168e-05,
      "loss": 0.0662,
      "num_input_tokens_seen": 1975872,
      "step": 530,
      "train_runtime": 654.9175,
      "train_tokens_per_second": 3016.978
    },
    {
      "epoch": 1.1614639397201292,
      "grad_norm": 3.0715815228177235e-05,
      "learning_rate": 2.0241243353207484e-05,
      "loss": 0.0669,
      "num_input_tokens_seen": 2012928,
      "step": 540,
      "train_runtime": 667.0485,
      "train_tokens_per_second": 3017.664
    },
    {
      "epoch": 1.1829924650161463,
      "grad_norm": 2.5172572350129485e-05,
      "learning_rate": 1.9923428105432796e-05,
      "loss": 0.0685,
      "num_input_tokens_seen": 2051168,
      "step": 550,
      "train_runtime": 679.4071,
      "train_tokens_per_second": 3019.056
    },
    {
      "epoch": 1.2045209903121636,
      "grad_norm": 5.135889659868553e-05,
      "learning_rate": 1.9603115960956852e-05,
      "loss": 0.0765,
      "num_input_tokens_seen": 2088576,
      "step": 560,
      "train_runtime": 691.6021,
      "train_tokens_per_second": 3019.91
    },
    {
      "epoch": 1.2260495156081808,
      "grad_norm": 2.326056892343331e-05,
      "learning_rate": 1.9280469364791505e-05,
      "loss": 0.0703,
      "num_input_tokens_seen": 2125216,
      "step": 570,
      "train_runtime": 703.6101,
      "train_tokens_per_second": 3020.445
    },
    {
      "epoch": 1.247578040904198,
      "grad_norm": 5.269297616905533e-05,
      "learning_rate": 1.8955651945856354e-05,
      "loss": 0.0793,
      "num_input_tokens_seen": 2162816,
      "step": 580,
      "train_runtime": 715.8692,
      "train_tokens_per_second": 3021.245
    },
    {
      "epoch": 1.2691065662002152,
      "grad_norm": 1.9257287931395695e-05,
      "learning_rate": 1.8628828433995013e-05,
      "loss": 0.0555,
      "num_input_tokens_seen": 2199712,
      "step": 590,
      "train_runtime": 727.9662,
      "train_tokens_per_second": 3021.723
    },
    {
      "epoch": 1.2906350914962326,
      "grad_norm": 3.0456994863925502e-05,
      "learning_rate": 1.8300164576433035e-05,
      "loss": 0.0714,
      "num_input_tokens_seen": 2236416,
      "step": 600,
      "train_runtime": 740.0539,
      "train_tokens_per_second": 3021.963
    },
    {
      "epoch": 1.3121636167922497,
      "grad_norm": 3.705903873196803e-05,
      "learning_rate": 1.7969827053719862e-05,
      "loss": 0.077,
      "num_input_tokens_seen": 2274272,
      "step": 610,
      "train_runtime": 752.3772,
      "train_tokens_per_second": 3022.782
    },
    {
      "epoch": 1.333692142088267,
      "grad_norm": 3.318333619972691e-05,
      "learning_rate": 1.7637983395197465e-05,
      "loss": 0.0971,
      "num_input_tokens_seen": 2310688,
      "step": 620,
      "train_runtime": 764.3993,
      "train_tokens_per_second": 3022.881
    },
    {
      "epoch": 1.3552206673842842,
      "grad_norm": 5.1804101531160995e-05,
      "learning_rate": 1.7304801894038522e-05,
      "loss": 0.0805,
      "num_input_tokens_seen": 2348608,
      "step": 630,
      "train_runtime": 776.7349,
      "train_tokens_per_second": 3023.693
    },
    {
      "epoch": 1.3767491926803013,
      "grad_norm": 2.364752617722843e-05,
      "learning_rate": 1.6970451521897198e-05,
      "loss": 0.0644,
      "num_input_tokens_seen": 2386144,
      "step": 640,
      "train_runtime": 789.0008,
      "train_tokens_per_second": 3024.26
    },
    {
      "epoch": 1.3982777179763186,
      "grad_norm": 9.15842319955118e-05,
      "learning_rate": 1.663510184321587e-05,
      "loss": 0.0782,
      "num_input_tokens_seen": 2422912,
      "step": 650,
      "train_runtime": 801.1296,
      "train_tokens_per_second": 3024.37
    },
    {
      "epoch": 1.419806243272336,
      "grad_norm": 3.013072273461148e-05,
      "learning_rate": 1.6298922929231194e-05,
      "loss": 0.0726,
      "num_input_tokens_seen": 2459744,
      "step": 660,
      "train_runtime": 813.3302,
      "train_tokens_per_second": 3024.287
    },
    {
      "epoch": 1.441334768568353,
      "grad_norm": 5.607481216429733e-05,
      "learning_rate": 1.5962085271723158e-05,
      "loss": 0.0651,
      "num_input_tokens_seen": 2497184,
      "step": 670,
      "train_runtime": 825.5543,
      "train_tokens_per_second": 3024.857
    },
    {
      "epoch": 1.4628632938643702,
      "grad_norm": 5.578975105891004e-05,
      "learning_rate": 1.5624759696550874e-05,
      "loss": 0.0777,
      "num_input_tokens_seen": 2535456,
      "step": 680,
      "train_runtime": 838.0003,
      "train_tokens_per_second": 3025.603
    },
    {
      "epoch": 1.4843918191603875,
      "grad_norm": 2.8685000870609656e-05,
      "learning_rate": 1.528711727701893e-05,
      "loss": 0.055,
      "num_input_tokens_seen": 2572704,
      "step": 690,
      "train_runtime": 850.1695,
      "train_tokens_per_second": 3026.107
    },
    {
      "epoch": 1.5059203444564049,
      "grad_norm": 6.515823770314455e-05,
      "learning_rate": 1.4949329247118258e-05,
      "loss": 0.0784,
      "num_input_tokens_seen": 2609696,
      "step": 700,
      "train_runtime": 862.2522,
      "train_tokens_per_second": 3026.604
    },
    {
      "epoch": 1.527448869752422,
      "grad_norm": 4.3776995880762115e-05,
      "learning_rate": 1.4611566914685489e-05,
      "loss": 0.0693,
      "num_input_tokens_seen": 2646976,
      "step": 710,
      "train_runtime": 874.4727,
      "train_tokens_per_second": 3026.94
    },
    {
      "epoch": 1.5489773950484391,
      "grad_norm": 3.7121964851394296e-05,
      "learning_rate": 1.4274001574524894e-05,
      "loss": 0.0696,
      "num_input_tokens_seen": 2684576,
      "step": 720,
      "train_runtime": 886.7361,
      "train_tokens_per_second": 3027.48
    },
    {
      "epoch": 1.5705059203444565,
      "grad_norm": 4.2310501157771796e-05,
      "learning_rate": 1.3936804421536906e-05,
      "loss": 0.0527,
      "num_input_tokens_seen": 2720736,
      "step": 730,
      "train_runtime": 899.1348,
      "train_tokens_per_second": 3025.949
    },
    {
      "epoch": 1.5920344456404736,
      "grad_norm": 3.037115493498277e-05,
      "learning_rate": 1.3600146463897327e-05,
      "loss": 0.0637,
      "num_input_tokens_seen": 2757728,
      "step": 740,
      "train_runtime": 911.269,
      "train_tokens_per_second": 3026.25
    },
    {
      "epoch": 1.6135629709364907,
      "grad_norm": 6.425149331334978e-05,
      "learning_rate": 1.3264198436331193e-05,
      "loss": 0.0876,
      "num_input_tokens_seen": 2795168,
      "step": 750,
      "train_runtime": 923.4281,
      "train_tokens_per_second": 3026.947
    },
    {
      "epoch": 1.635091496232508,
      "grad_norm": 2.6526600777287968e-05,
      "learning_rate": 1.2929130713525396e-05,
      "loss": 0.069,
      "num_input_tokens_seen": 2832000,
      "step": 760,
      "train_runtime": 935.4717,
      "train_tokens_per_second": 3027.35
    },
    {
      "epoch": 1.6566200215285254,
      "grad_norm": 2.431479333608877e-05,
      "learning_rate": 1.2595113223723788e-05,
      "loss": 0.0818,
      "num_input_tokens_seen": 2869696,
      "step": 770,
      "train_runtime": 947.7383,
      "train_tokens_per_second": 3027.941
    },
    {
      "epoch": 1.6781485468245425,
      "grad_norm": 4.668053952627815e-05,
      "learning_rate": 1.2262315362548825e-05,
      "loss": 0.0566,
      "num_input_tokens_seen": 2907840,
      "step": 780,
      "train_runtime": 960.1454,
      "train_tokens_per_second": 3028.541
    },
    {
      "epoch": 1.6996770721205596,
      "grad_norm": 7.140763045754284e-05,
      "learning_rate": 1.1930905907093196e-05,
      "loss": 0.0682,
      "num_input_tokens_seen": 2943232,
      "step": 790,
      "train_runtime": 971.949,
      "train_tokens_per_second": 3028.175
    },
    {
      "epoch": 1.721205597416577,
      "grad_norm": 9.523890184937045e-05,
      "learning_rate": 1.1601052930325236e-05,
      "loss": 0.067,
      "num_input_tokens_seen": 2980608,
      "step": 800,
      "train_runtime": 984.1834,
      "train_tokens_per_second": 3028.509
    },
    {
      "epoch": 1.7427341227125943,
      "grad_norm": 3.8205707824090496e-05,
      "learning_rate": 1.127292371585141e-05,
      "loss": 0.0654,
      "num_input_tokens_seen": 3018080,
      "step": 810,
      "train_runtime": 996.4307,
      "train_tokens_per_second": 3028.891
    },
    {
      "epoch": 1.7642626480086114,
      "grad_norm": 7.051431020954624e-05,
      "learning_rate": 1.0946684673079074e-05,
      "loss": 0.0811,
      "num_input_tokens_seen": 3055520,
      "step": 820,
      "train_runtime": 1008.6552,
      "train_tokens_per_second": 3029.301
    },
    {
      "epoch": 1.7857911733046286,
      "grad_norm": 2.1617921447614208e-05,
      "learning_rate": 1.0622501252822628e-05,
      "loss": 0.0545,
      "num_input_tokens_seen": 3092800,
      "step": 830,
      "train_runtime": 1020.8276,
      "train_tokens_per_second": 3029.699
    },
    {
      "epoch": 1.807319698600646,
      "grad_norm": 4.000938861281611e-05,
      "learning_rate": 1.0300537863395823e-05,
      "loss": 0.0716,
      "num_input_tokens_seen": 3130656,
      "step": 840,
      "train_runtime": 1033.0891,
      "train_tokens_per_second": 3030.383
    },
    {
      "epoch": 1.8288482238966632,
      "grad_norm": 2.290739575983025e-05,
      "learning_rate": 9.98095778723271e-06,
      "loss": 0.0635,
      "num_input_tokens_seen": 3167744,
      "step": 850,
      "train_runtime": 1045.1665,
      "train_tokens_per_second": 3030.851
    },
    {
      "epoch": 1.8503767491926801,
      "grad_norm": 3.616542380768806e-05,
      "learning_rate": 9.663923098079656e-06,
      "loss": 0.0667,
      "num_input_tokens_seen": 3205376,
      "step": 860,
      "train_runtime": 1057.3713,
      "train_tokens_per_second": 3031.457
    },
    {
      "epoch": 1.8719052744886975,
      "grad_norm": 2.1878349798498675e-05,
      "learning_rate": 9.349594578800259e-06,
      "loss": 0.0762,
      "num_input_tokens_seen": 3242976,
      "step": 870,
      "train_runtime": 1069.6685,
      "train_tokens_per_second": 3031.758
    },
    {
      "epoch": 1.8934337997847148,
      "grad_norm": 4.789059312315658e-05,
      "learning_rate": 9.038131639834977e-06,
      "loss": 0.0939,
      "num_input_tokens_seen": 3280096,
      "step": 880,
      "train_runtime": 1081.805,
      "train_tokens_per_second": 3032.059
    },
    {
      "epoch": 1.914962325080732,
      "grad_norm": 2.750419662334025e-05,
      "learning_rate": 8.7296922383568e-06,
      "loss": 0.0653,
      "num_input_tokens_seen": 3316960,
      "step": 890,
      "train_runtime": 1093.8603,
      "train_tokens_per_second": 3032.343
    },
    {
      "epoch": 1.936490850376749,
      "grad_norm": 3.527307853801176e-05,
      "learning_rate": 8.424432798163838e-06,
      "loss": 0.0666,
      "num_input_tokens_seen": 3353600,
      "step": 900,
      "train_runtime": 1105.8625,
      "train_tokens_per_second": 3032.565
    },
    {
      "epoch": 1.9580193756727664,
      "grad_norm": 3.1842311727814376e-05,
      "learning_rate": 8.122508130349711e-06,
      "loss": 0.0528,
      "num_input_tokens_seen": 3390592,
      "step": 910,
      "train_runtime": 1117.9098,
      "train_tokens_per_second": 3032.975
    },
    {
      "epoch": 1.9795479009687837,
      "grad_norm": 3.884573743562214e-05,
      "learning_rate": 7.82407135479164e-06,
      "loss": 0.073,
      "num_input_tokens_seen": 3427840,
      "step": 920,
      "train_runtime": 1130.0845,
      "train_tokens_per_second": 3033.26
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.225112651416566e-06,
      "learning_rate": 7.529273822496379e-06,
      "loss": 0.0465,
      "num_input_tokens_seen": 3464032,
      "step": 930,
      "train_runtime": 1141.7615,
      "train_tokens_per_second": 3033.937
    },
    {
      "epoch": 2.0215285252960173,
      "grad_norm": 6.878191925352439e-05,
      "learning_rate": 7.238265038843143e-06,
      "loss": 0.0644,
      "num_input_tokens_seen": 3501056,
      "step": 940,
      "train_runtime": 1153.8746,
      "train_tokens_per_second": 3034.174
    },
    {
      "epoch": 2.0430570505920342,
      "grad_norm": 3.4607230190886185e-05,
      "learning_rate": 6.951192587762555e-06,
      "loss": 0.0383,
      "num_input_tokens_seen": 3538464,
      "step": 950,
      "train_runtime": 1166.1028,
      "train_tokens_per_second": 3034.436
    },
    {
      "epoch": 2.0645855758880516,
      "grad_norm": 2.9962731787236407e-05,
      "learning_rate": 6.668202056890052e-06,
      "loss": 0.046,
      "num_input_tokens_seen": 3575552,
      "step": 960,
      "train_runtime": 1178.2163,
      "train_tokens_per_second": 3034.716
    },
    {
      "epoch": 2.086114101184069,
      "grad_norm": 6.91282912157476e-05,
      "learning_rate": 6.389436963731707e-06,
      "loss": 0.0534,
      "num_input_tokens_seen": 3612736,
      "step": 970,
      "train_runtime": 1190.3951,
      "train_tokens_per_second": 3034.905
    },
    {
      "epoch": 2.1076426264800863,
      "grad_norm": 4.139909287914634e-05,
      "learning_rate": 6.115038682879904e-06,
      "loss": 0.0661,
      "num_input_tokens_seen": 3650464,
      "step": 980,
      "train_runtime": 1202.6943,
      "train_tokens_per_second": 3035.238
    },
    {
      "epoch": 2.129171151776103,
      "grad_norm": 2.161462907679379e-05,
      "learning_rate": 5.845146374315781e-06,
      "loss": 0.052,
      "num_input_tokens_seen": 3687392,
      "step": 990,
      "train_runtime": 1214.8688,
      "train_tokens_per_second": 3035.218
    },
    {
      "epoch": 2.1506996770721205,
      "grad_norm": 4.022341818199493e-05,
      "learning_rate": 5.579896912834807e-06,
      "loss": 0.0735,
      "num_input_tokens_seen": 3725152,
      "step": 1000,
      "train_runtime": 1227.1475,
      "train_tokens_per_second": 3035.619
    },
    {
      "epoch": 2.172228202368138,
      "grad_norm": 6.646889232797548e-05,
      "learning_rate": 5.319424818631293e-06,
      "loss": 0.074,
      "num_input_tokens_seen": 3762752,
      "step": 1010,
      "train_runtime": 1240.4659,
      "train_tokens_per_second": 3033.338
    },
    {
      "epoch": 2.193756727664155,
      "grad_norm": 4.291176446713507e-05,
      "learning_rate": 5.063862189076988e-06,
      "loss": 0.0575,
      "num_input_tokens_seen": 3800736,
      "step": 1020,
      "train_runtime": 1252.7902,
      "train_tokens_per_second": 3033.817
    },
    {
      "epoch": 2.215285252960172,
      "grad_norm": 4.701681973529048e-05,
      "learning_rate": 4.813338631728471e-06,
      "loss": 0.0621,
      "num_input_tokens_seen": 3838432,
      "step": 1030,
      "train_runtime": 1265.0445,
      "train_tokens_per_second": 3034.227
    },
    {
      "epoch": 2.2368137782561894,
      "grad_norm": 6.257462518988177e-05,
      "learning_rate": 4.567981198597168e-06,
      "loss": 0.0649,
      "num_input_tokens_seen": 3876224,
      "step": 1040,
      "train_runtime": 1277.3555,
      "train_tokens_per_second": 3034.569
    },
    {
      "epoch": 2.2583423035522068,
      "grad_norm": 3.18081074510701e-05,
      "learning_rate": 4.327914321715459e-06,
      "loss": 0.0453,
      "num_input_tokens_seen": 3913216,
      "step": 1050,
      "train_runtime": 1289.4599,
      "train_tokens_per_second": 3034.771
    },
    {
      "epoch": 2.2798708288482237,
      "grad_norm": 4.293344682082534e-05,
      "learning_rate": 4.09325975003146e-06,
      "loss": 0.0513,
      "num_input_tokens_seen": 3950208,
      "step": 1060,
      "train_runtime": 1301.5277,
      "train_tokens_per_second": 3035.055
    },
    {
      "epoch": 2.301399354144241,
      "grad_norm": 4.381421967991628e-05,
      "learning_rate": 3.864136487664538e-06,
      "loss": 0.0502,
      "num_input_tokens_seen": 3986752,
      "step": 1070,
      "train_runtime": 1313.5554,
      "train_tokens_per_second": 3035.085
    },
    {
      "epoch": 2.3229278794402584,
      "grad_norm": 2.7669038900057785e-05,
      "learning_rate": 3.640660733552863e-06,
      "loss": 0.0373,
      "num_input_tokens_seen": 4023360,
      "step": 1080,
      "train_runtime": 1325.5459,
      "train_tokens_per_second": 3035.248
    },
    {
      "epoch": 2.3444564047362757,
      "grad_norm": 4.423334030434489e-05,
      "learning_rate": 3.4229458225235696e-06,
      "loss": 0.0405,
      "num_input_tokens_seen": 4061856,
      "step": 1090,
      "train_runtime": 1337.9632,
      "train_tokens_per_second": 3035.85
    },
    {
      "epoch": 2.3659849300322926,
      "grad_norm": 3.791322887991555e-05,
      "learning_rate": 3.211102167815469e-06,
      "loss": 0.0558,
      "num_input_tokens_seen": 4098464,
      "step": 1100,
      "train_runtime": 1349.9275,
      "train_tokens_per_second": 3036.062
    },
    {
      "epoch": 2.38751345532831,
      "grad_norm": 9.097262227442116e-05,
      "learning_rate": 3.0052372050834025e-06,
      "loss": 0.0675,
      "num_input_tokens_seen": 4136576,
      "step": 1110,
      "train_runtime": 1362.2813,
      "train_tokens_per_second": 3036.506
    },
    {
      "epoch": 2.4090419806243273,
      "grad_norm": 6.769008177798241e-05,
      "learning_rate": 2.805455337912722e-06,
      "loss": 0.054,
      "num_input_tokens_seen": 4174528,
      "step": 1120,
      "train_runtime": 1374.5305,
      "train_tokens_per_second": 3037.057
    },
    {
      "epoch": 2.4305705059203446,
      "grad_norm": 2.151660737581551e-05,
      "learning_rate": 2.611857884871379e-06,
      "loss": 0.0512,
      "num_input_tokens_seen": 4211712,
      "step": 1130,
      "train_runtime": 1386.6676,
      "train_tokens_per_second": 3037.29
    },
    {
      "epoch": 2.4520990312163615,
      "grad_norm": 1.225809410243528e-05,
      "learning_rate": 2.424543028126661e-06,
      "loss": 0.0605,
      "num_input_tokens_seen": 4248800,
      "step": 1140,
      "train_runtime": 1398.8282,
      "train_tokens_per_second": 3037.4
    },
    {
      "epoch": 2.473627556512379,
      "grad_norm": 5.214564589550719e-05,
      "learning_rate": 2.243605763652489e-06,
      "loss": 0.0506,
      "num_input_tokens_seen": 4286496,
      "step": 1150,
      "train_runtime": 1411.0662,
      "train_tokens_per_second": 3037.771
    },
    {
      "epoch": 2.495156081808396,
      "grad_norm": 6.123852654127404e-05,
      "learning_rate": 2.0691378530525933e-06,
      "loss": 0.058,
      "num_input_tokens_seen": 4323328,
      "step": 1160,
      "train_runtime": 1423.1546,
      "train_tokens_per_second": 3037.848
    },
    {
      "epoch": 2.5166846071044136,
      "grad_norm": 7.809853559592739e-05,
      "learning_rate": 1.9012277770239829e-06,
      "loss": 0.0531,
      "num_input_tokens_seen": 4360384,
      "step": 1170,
      "train_runtime": 1435.2826,
      "train_tokens_per_second": 3037.997
    },
    {
      "epoch": 2.5382131324004304,
      "grad_norm": 3.4470616810722277e-05,
      "learning_rate": 1.7399606904843224e-06,
      "loss": 0.0402,
      "num_input_tokens_seen": 4397440,
      "step": 1180,
      "train_runtime": 1447.4068,
      "train_tokens_per_second": 3038.151
    },
    {
      "epoch": 2.559741657696448,
      "grad_norm": 4.962022285326384e-05,
      "learning_rate": 1.5854183793859577e-06,
      "loss": 0.0482,
      "num_input_tokens_seen": 4435584,
      "step": 1190,
      "train_runtime": 1459.8108,
      "train_tokens_per_second": 3038.465
    },
    {
      "epoch": 2.581270182992465,
      "grad_norm": 2.7074293029727414e-05,
      "learning_rate": 1.437679219238498e-06,
      "loss": 0.0475,
      "num_input_tokens_seen": 4472800,
      "step": 1200,
      "train_runtime": 1471.9296,
      "train_tokens_per_second": 3038.732
    },
    {
      "epoch": 2.602798708288482,
      "grad_norm": 4.890407217317261e-05,
      "learning_rate": 1.2968181353609854e-06,
      "loss": 0.0505,
      "num_input_tokens_seen": 4510464,
      "step": 1210,
      "train_runtime": 1484.1248,
      "train_tokens_per_second": 3039.141
    },
    {
      "epoch": 2.6243272335844994,
      "grad_norm": 3.1680385291110724e-05,
      "learning_rate": 1.1629065648838495e-06,
      "loss": 0.0624,
      "num_input_tokens_seen": 4546592,
      "step": 1220,
      "train_runtime": 1496.1043,
      "train_tokens_per_second": 3038.954
    },
    {
      "epoch": 2.6458557588805167,
      "grad_norm": 1.584286474098917e-05,
      "learning_rate": 1.0360124205198102e-06,
      "loss": 0.0477,
      "num_input_tokens_seen": 4583680,
      "step": 1230,
      "train_runtime": 1508.1942,
      "train_tokens_per_second": 3039.184
    },
    {
      "epoch": 2.667384284176534,
      "grad_norm": 5.2668325224658474e-05,
      "learning_rate": 9.162000561222672e-07,
      "loss": 0.0589,
      "num_input_tokens_seen": 4622080,
      "step": 1240,
      "train_runtime": 1520.6529,
      "train_tokens_per_second": 3039.537
    },
    {
      "epoch": 2.6889128094725514,
      "grad_norm": 3.029003164556343e-05,
      "learning_rate": 8.035302340484646e-07,
      "loss": 0.072,
      "num_input_tokens_seen": 4658848,
      "step": 1250,
      "train_runtime": 1532.7125,
      "train_tokens_per_second": 3039.61
    },
    {
      "epoch": 2.7104413347685683,
      "grad_norm": 4.495487883104943e-05,
      "learning_rate": 6.980600943441273e-07,
      "loss": 0.0484,
      "num_input_tokens_seen": 4695968,
      "step": 1260,
      "train_runtime": 1544.8493,
      "train_tokens_per_second": 3039.758
    },
    {
      "epoch": 2.7319698600645856,
      "grad_norm": 3.930521052097902e-05,
      "learning_rate": 5.998431257651132e-07,
      "loss": 0.0592,
      "num_input_tokens_seen": 4733344,
      "step": 1270,
      "train_runtime": 1557.0831,
      "train_tokens_per_second": 3039.879
    },
    {
      "epoch": 2.7534983853606025,
      "grad_norm": 3.577708048396744e-05,
      "learning_rate": 5.089291386508055e-07,
      "loss": 0.0499,
      "num_input_tokens_seen": 4770496,
      "step": 1280,
      "train_runtime": 1569.2037,
      "train_tokens_per_second": 3040.074
    },
    {
      "epoch": 2.77502691065662,
      "grad_norm": 3.729393210960552e-05,
      "learning_rate": 4.2536423966299256e-07,
      "loss": 0.045,
      "num_input_tokens_seen": 4807456,
      "step": 1290,
      "train_runtime": 1581.3251,
      "train_tokens_per_second": 3040.144
    },
    {
      "epoch": 2.7965554359526372,
      "grad_norm": 6.718758959323168e-05,
      "learning_rate": 3.491908084030776e-07,
      "loss": 0.0624,
      "num_input_tokens_seen": 4845440,
      "step": 1300,
      "train_runtime": 1593.7105,
      "train_tokens_per_second": 3040.351
    },
    {
      "epoch": 2.8180839612486546,
      "grad_norm": 3.5270619264338166e-05,
      "learning_rate": 2.804474759194176e-07,
      "loss": 0.0492,
      "num_input_tokens_seen": 4883392,
      "step": 1310,
      "train_runtime": 1606.0326,
      "train_tokens_per_second": 3040.656
    },
    {
      "epoch": 2.839612486544672,
      "grad_norm": 3.882671080646105e-05,
      "learning_rate": 2.1916910511575627e-07,
      "loss": 0.06,
      "num_input_tokens_seen": 4920384,
      "step": 1320,
      "train_runtime": 1618.1084,
      "train_tokens_per_second": 3040.825
    },
    {
      "epoch": 2.861141011840689,
      "grad_norm": 4.175250433036126e-05,
      "learning_rate": 1.6538677307062933e-07,
      "loss": 0.0498,
      "num_input_tokens_seen": 4957888,
      "step": 1330,
      "train_runtime": 1630.3658,
      "train_tokens_per_second": 3040.967
    },
    {
      "epoch": 2.882669537136706,
      "grad_norm": 6.56303673167713e-05,
      "learning_rate": 1.1912775527676278e-07,
      "loss": 0.0436,
      "num_input_tokens_seen": 4994944,
      "step": 1340,
      "train_runtime": 1642.4487,
      "train_tokens_per_second": 3041.157
    },
    {
      "epoch": 2.9041980624327235,
      "grad_norm": 4.389798777992837e-05,
      "learning_rate": 8.04155118084099e-08,
      "loss": 0.0697,
      "num_input_tokens_seen": 5031584,
      "step": 1350,
      "train_runtime": 1654.4352,
      "train_tokens_per_second": 3041.27
    },
    {
      "epoch": 2.9257265877287404,
      "grad_norm": 4.761307354783639e-05,
      "learning_rate": 4.9269675423671556e-08,
      "loss": 0.0583,
      "num_input_tokens_seen": 5068928,
      "step": 1360,
      "train_runtime": 1666.6414,
      "train_tokens_per_second": 3041.403
    },
    {
      "epoch": 2.9472551130247577,
      "grad_norm": 4.6141562052071095e-05,
      "learning_rate": 2.570604160781753e-08,
      "loss": 0.0457,
      "num_input_tokens_seen": 5105984,
      "step": 1370,
      "train_runtime": 1678.7701,
      "train_tokens_per_second": 3041.503
    },
    {
      "epoch": 2.968783638320775,
      "grad_norm": 4.6007688069948927e-05,
      "learning_rate": 9.736560562675912e-09,
      "loss": 0.0447,
      "num_input_tokens_seen": 5143296,
      "step": 1380,
      "train_runtime": 1690.9638,
      "train_tokens_per_second": 3041.636
    },
    {
      "epoch": 2.9903121636167924,
      "grad_norm": 2.3644568500458263e-05,
      "learning_rate": 1.3693311461326908e-09,
      "loss": 0.0571,
      "num_input_tokens_seen": 5181056,
      "step": 1390,
      "train_runtime": 1703.2237,
      "train_tokens_per_second": 3041.912
    },
    {
      "epoch": 3.0,
      "num_input_tokens_seen": 5197824,
      "step": 1395,
      "total_flos": 2.2251190147979674e+17,
      "train_loss": 0.07927756027508807,
      "train_runtime": 1710.0553,
      "train_samples_per_second": 6.519,
      "train_steps_per_second": 0.816
    }
  ],
  "logging_steps": 10,
  "max_steps": 1395,
  "num_input_tokens_seen": 5197824,
  "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": 2.2251190147979674e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}