{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.444444444444445,
  "eval_steps": 10,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08888888888888889,
      "grad_norm": 6.322239398956299,
      "learning_rate": 4e-05,
      "loss": 2.1156,
      "step": 10
    },
    {
      "epoch": 0.08888888888888889,
      "eval_loss": 1.5894473791122437,
      "eval_runtime": 8.0398,
      "eval_samples_per_second": 12.438,
      "eval_steps_per_second": 12.438,
      "step": 10
    },
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 4.342599391937256,
      "learning_rate": 8e-05,
      "loss": 1.1893,
      "step": 20
    },
    {
      "epoch": 0.17777777777777778,
      "eval_loss": 0.6867849826812744,
      "eval_runtime": 8.0305,
      "eval_samples_per_second": 12.453,
      "eval_steps_per_second": 12.453,
      "step": 20
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 2.0583019256591797,
      "learning_rate": 0.00012,
      "loss": 0.5218,
      "step": 30
    },
    {
      "epoch": 0.26666666666666666,
      "eval_loss": 0.45551854372024536,
      "eval_runtime": 8.0298,
      "eval_samples_per_second": 12.454,
      "eval_steps_per_second": 12.454,
      "step": 30
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 1.83012056350708,
      "learning_rate": 0.00016,
      "loss": 0.5292,
      "step": 40
    },
    {
      "epoch": 0.35555555555555557,
      "eval_loss": 0.3795148730278015,
      "eval_runtime": 8.0289,
      "eval_samples_per_second": 12.455,
      "eval_steps_per_second": 12.455,
      "step": 40
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 2.1445064544677734,
      "learning_rate": 0.0002,
      "loss": 0.3866,
      "step": 50
    },
    {
      "epoch": 0.4444444444444444,
      "eval_loss": 0.30648669600486755,
      "eval_runtime": 8.0213,
      "eval_samples_per_second": 12.467,
      "eval_steps_per_second": 12.467,
      "step": 50
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 1.3674068450927734,
      "learning_rate": 0.00019975640502598244,
      "loss": 0.3232,
      "step": 60
    },
    {
      "epoch": 0.5333333333333333,
      "eval_loss": 0.20737296342849731,
      "eval_runtime": 8.0166,
      "eval_samples_per_second": 12.474,
      "eval_steps_per_second": 12.474,
      "step": 60
    },
    {
      "epoch": 0.6222222222222222,
      "grad_norm": 1.0751285552978516,
      "learning_rate": 0.00019902680687415705,
      "loss": 0.1802,
      "step": 70
    },
    {
      "epoch": 0.6222222222222222,
      "eval_loss": 0.15315091609954834,
      "eval_runtime": 8.025,
      "eval_samples_per_second": 12.461,
      "eval_steps_per_second": 12.461,
      "step": 70
    },
    {
      "epoch": 0.7111111111111111,
      "grad_norm": 0.568341076374054,
      "learning_rate": 0.00019781476007338058,
      "loss": 0.21,
      "step": 80
    },
    {
      "epoch": 0.7111111111111111,
      "eval_loss": 0.13480396568775177,
      "eval_runtime": 8.0237,
      "eval_samples_per_second": 12.463,
      "eval_steps_per_second": 12.463,
      "step": 80
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.9985604286193848,
      "learning_rate": 0.0001961261695938319,
      "loss": 0.158,
      "step": 90
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.13721750676631927,
      "eval_runtime": 8.0237,
      "eval_samples_per_second": 12.463,
      "eval_steps_per_second": 12.463,
      "step": 90
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 0.8096336722373962,
      "learning_rate": 0.00019396926207859084,
      "loss": 0.1629,
      "step": 100
    },
    {
      "epoch": 0.8888888888888888,
      "eval_loss": 0.12762245535850525,
      "eval_runtime": 8.0161,
      "eval_samples_per_second": 12.475,
      "eval_steps_per_second": 12.475,
      "step": 100
    },
    {
      "epoch": 0.9777777777777777,
      "grad_norm": 1.2173279523849487,
      "learning_rate": 0.0001913545457642601,
      "loss": 0.0966,
      "step": 110
    },
    {
      "epoch": 0.9777777777777777,
      "eval_loss": 0.10031093657016754,
      "eval_runtime": 8.0285,
      "eval_samples_per_second": 12.456,
      "eval_steps_per_second": 12.456,
      "step": 110
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 0.7404722571372986,
      "learning_rate": 0.00018829475928589271,
      "loss": 0.0643,
      "step": 120
    },
    {
      "epoch": 1.0666666666666667,
      "eval_loss": 0.08794313669204712,
      "eval_runtime": 8.0399,
      "eval_samples_per_second": 12.438,
      "eval_steps_per_second": 12.438,
      "step": 120
    },
    {
      "epoch": 1.1555555555555554,
      "grad_norm": 0.9548615217208862,
      "learning_rate": 0.0001848048096156426,
      "loss": 0.0726,
      "step": 130
    },
    {
      "epoch": 1.1555555555555554,
      "eval_loss": 0.08720792084932327,
      "eval_runtime": 8.2778,
      "eval_samples_per_second": 12.08,
      "eval_steps_per_second": 12.08,
      "step": 130
    },
    {
      "epoch": 1.2444444444444445,
      "grad_norm": 1.0252933502197266,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0493,
      "step": 140
    },
    {
      "epoch": 1.2444444444444445,
      "eval_loss": 0.09057007730007172,
      "eval_runtime": 8.2874,
      "eval_samples_per_second": 12.067,
      "eval_steps_per_second": 12.067,
      "step": 140
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.5175765156745911,
      "learning_rate": 0.0001766044443118978,
      "loss": 0.0746,
      "step": 150
    },
    {
      "epoch": 1.3333333333333333,
      "eval_loss": 0.058685798197984695,
      "eval_runtime": 8.2384,
      "eval_samples_per_second": 12.138,
      "eval_steps_per_second": 12.138,
      "step": 150
    },
    {
      "epoch": 1.4222222222222223,
      "grad_norm": 0.33382534980773926,
      "learning_rate": 0.0001719339800338651,
      "loss": 0.0473,
      "step": 160
    },
    {
      "epoch": 1.4222222222222223,
      "eval_loss": 0.0560651533305645,
      "eval_runtime": 8.1982,
      "eval_samples_per_second": 12.198,
      "eval_steps_per_second": 12.198,
      "step": 160
    },
    {
      "epoch": 1.511111111111111,
      "grad_norm": 1.0913978815078735,
      "learning_rate": 0.00016691306063588583,
      "loss": 0.0644,
      "step": 170
    },
    {
      "epoch": 1.511111111111111,
      "eval_loss": 0.05025744438171387,
      "eval_runtime": 8.239,
      "eval_samples_per_second": 12.137,
      "eval_steps_per_second": 12.137,
      "step": 170
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.8013315796852112,
      "learning_rate": 0.0001615661475325658,
      "loss": 0.0366,
      "step": 180
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.030684156343340874,
      "eval_runtime": 8.2772,
      "eval_samples_per_second": 12.081,
      "eval_steps_per_second": 12.081,
      "step": 180
    },
    {
      "epoch": 1.6888888888888889,
      "grad_norm": 0.2898242175579071,
      "learning_rate": 0.0001559192903470747,
      "loss": 0.0247,
      "step": 190
    },
    {
      "epoch": 1.6888888888888889,
      "eval_loss": 0.023328043520450592,
      "eval_runtime": 8.2724,
      "eval_samples_per_second": 12.088,
      "eval_steps_per_second": 12.088,
      "step": 190
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 0.0667782798409462,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.01,
      "step": 200
    },
    {
      "epoch": 1.7777777777777777,
      "eval_loss": 0.021545417606830597,
      "eval_runtime": 8.2753,
      "eval_samples_per_second": 12.084,
      "eval_steps_per_second": 12.084,
      "step": 200
    },
    {
      "epoch": 1.8666666666666667,
      "grad_norm": 1.8411855697631836,
      "learning_rate": 0.00014383711467890774,
      "loss": 0.0393,
      "step": 210
    },
    {
      "epoch": 1.8666666666666667,
      "eval_loss": 0.012232878245413303,
      "eval_runtime": 8.275,
      "eval_samples_per_second": 12.085,
      "eval_steps_per_second": 12.085,
      "step": 210
    },
    {
      "epoch": 1.9555555555555557,
      "grad_norm": 2.9970459938049316,
      "learning_rate": 0.00013746065934159123,
      "loss": 0.0299,
      "step": 220
    },
    {
      "epoch": 1.9555555555555557,
      "eval_loss": 0.01798514649271965,
      "eval_runtime": 8.2746,
      "eval_samples_per_second": 12.085,
      "eval_steps_per_second": 12.085,
      "step": 220
    },
    {
      "epoch": 2.0444444444444443,
      "grad_norm": 1.0432640314102173,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0166,
      "step": 230
    },
    {
      "epoch": 2.0444444444444443,
      "eval_loss": 0.008207106962800026,
      "eval_runtime": 8.2792,
      "eval_samples_per_second": 12.078,
      "eval_steps_per_second": 12.078,
      "step": 230
    },
    {
      "epoch": 2.1333333333333333,
      "grad_norm": 1.9301716089248657,
      "learning_rate": 0.00012419218955996676,
      "loss": 0.0319,
      "step": 240
    },
    {
      "epoch": 2.1333333333333333,
      "eval_loss": 0.008276881650090218,
      "eval_runtime": 8.0782,
      "eval_samples_per_second": 12.379,
      "eval_steps_per_second": 12.379,
      "step": 240
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.07122544944286346,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.0077,
      "step": 250
    },
    {
      "epoch": 2.2222222222222223,
      "eval_loss": 0.007150276098400354,
      "eval_runtime": 8.0319,
      "eval_samples_per_second": 12.45,
      "eval_steps_per_second": 12.45,
      "step": 250
    },
    {
      "epoch": 2.311111111111111,
      "grad_norm": 0.16882538795471191,
      "learning_rate": 0.00011045284632676536,
      "loss": 0.0141,
      "step": 260
    },
    {
      "epoch": 2.311111111111111,
      "eval_loss": 0.003109171986579895,
      "eval_runtime": 8.0309,
      "eval_samples_per_second": 12.452,
      "eval_steps_per_second": 12.452,
      "step": 260
    },
    {
      "epoch": 2.4,
      "grad_norm": 0.037282537668943405,
      "learning_rate": 0.00010348994967025012,
      "loss": 0.0017,
      "step": 270
    },
    {
      "epoch": 2.4,
      "eval_loss": 0.012033730745315552,
      "eval_runtime": 8.0265,
      "eval_samples_per_second": 12.459,
      "eval_steps_per_second": 12.459,
      "step": 270
    },
    {
      "epoch": 2.488888888888889,
      "grad_norm": 0.022591086104512215,
      "learning_rate": 9.651005032974994e-05,
      "loss": 0.0015,
      "step": 280
    },
    {
      "epoch": 2.488888888888889,
      "eval_loss": 0.015280201099812984,
      "eval_runtime": 8.0184,
      "eval_samples_per_second": 12.471,
      "eval_steps_per_second": 12.471,
      "step": 280
    },
    {
      "epoch": 2.5777777777777775,
      "grad_norm": 1.2939496040344238,
      "learning_rate": 8.954715367323468e-05,
      "loss": 0.0126,
      "step": 290
    },
    {
      "epoch": 2.5777777777777775,
      "eval_loss": 0.01406156551092863,
      "eval_runtime": 8.0171,
      "eval_samples_per_second": 12.473,
      "eval_steps_per_second": 12.473,
      "step": 290
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.009937470778822899,
      "learning_rate": 8.263518223330697e-05,
      "loss": 0.0043,
      "step": 300
    },
    {
      "epoch": 2.6666666666666665,
      "eval_loss": 0.0021963752806186676,
      "eval_runtime": 8.0193,
      "eval_samples_per_second": 12.47,
      "eval_steps_per_second": 12.47,
      "step": 300
    },
    {
      "epoch": 2.7555555555555555,
      "grad_norm": 0.022694729268550873,
      "learning_rate": 7.580781044003324e-05,
      "loss": 0.0068,
      "step": 310
    },
    {
      "epoch": 2.7555555555555555,
      "eval_loss": 0.001871286309324205,
      "eval_runtime": 8.0241,
      "eval_samples_per_second": 12.463,
      "eval_steps_per_second": 12.463,
      "step": 310
    },
    {
      "epoch": 2.8444444444444446,
      "grad_norm": 0.11103329062461853,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.0018,
      "step": 320
    },
    {
      "epoch": 2.8444444444444446,
      "eval_loss": 0.002184124430641532,
      "eval_runtime": 8.0123,
      "eval_samples_per_second": 12.481,
      "eval_steps_per_second": 12.481,
      "step": 320
    },
    {
      "epoch": 2.9333333333333336,
      "grad_norm": 0.02097630314528942,
      "learning_rate": 6.25393406584088e-05,
      "loss": 0.0026,
      "step": 330
    },
    {
      "epoch": 2.9333333333333336,
      "eval_loss": 0.0034216546919196844,
      "eval_runtime": 8.0164,
      "eval_samples_per_second": 12.474,
      "eval_steps_per_second": 12.474,
      "step": 330
    },
    {
      "epoch": 3.022222222222222,
      "grad_norm": 0.0426625981926918,
      "learning_rate": 5.616288532109225e-05,
      "loss": 0.0017,
      "step": 340
    },
    {
      "epoch": 3.022222222222222,
      "eval_loss": 0.007565508596599102,
      "eval_runtime": 8.0255,
      "eval_samples_per_second": 12.46,
      "eval_steps_per_second": 12.46,
      "step": 340
    },
    {
      "epoch": 3.111111111111111,
      "grad_norm": 0.0036587081849575043,
      "learning_rate": 5.000000000000002e-05,
      "loss": 0.0002,
      "step": 350
    },
    {
      "epoch": 3.111111111111111,
      "eval_loss": 0.010203778743743896,
      "eval_runtime": 8.017,
      "eval_samples_per_second": 12.473,
      "eval_steps_per_second": 12.473,
      "step": 350
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.008132525719702244,
      "learning_rate": 4.4080709652925336e-05,
      "loss": 0.0004,
      "step": 360
    },
    {
      "epoch": 3.2,
      "eval_loss": 0.011154056526720524,
      "eval_runtime": 8.0288,
      "eval_samples_per_second": 12.455,
      "eval_steps_per_second": 12.455,
      "step": 360
    },
    {
      "epoch": 3.2888888888888888,
      "grad_norm": 0.00727389520034194,
      "learning_rate": 3.843385246743417e-05,
      "loss": 0.006,
      "step": 370
    },
    {
      "epoch": 3.2888888888888888,
      "eval_loss": 0.00937813799828291,
      "eval_runtime": 8.0273,
      "eval_samples_per_second": 12.458,
      "eval_steps_per_second": 12.458,
      "step": 370
    },
    {
      "epoch": 3.3777777777777778,
      "grad_norm": 0.004803340416401625,
      "learning_rate": 3.308693936411421e-05,
      "loss": 0.0003,
      "step": 380
    },
    {
      "epoch": 3.3777777777777778,
      "eval_loss": 0.007455301936715841,
      "eval_runtime": 8.0254,
      "eval_samples_per_second": 12.46,
      "eval_steps_per_second": 12.46,
      "step": 380
    },
    {
      "epoch": 3.466666666666667,
      "grad_norm": 0.005293034482747316,
      "learning_rate": 2.8066019966134904e-05,
      "loss": 0.0003,
      "step": 390
    },
    {
      "epoch": 3.466666666666667,
      "eval_loss": 0.006887929514050484,
      "eval_runtime": 8.0268,
      "eval_samples_per_second": 12.458,
      "eval_steps_per_second": 12.458,
      "step": 390
    },
    {
      "epoch": 3.5555555555555554,
      "grad_norm": 0.01216947752982378,
      "learning_rate": 2.339555568810221e-05,
      "loss": 0.0002,
      "step": 400
    },
    {
      "epoch": 3.5555555555555554,
      "eval_loss": 0.006745634134858847,
      "eval_runtime": 8.032,
      "eval_samples_per_second": 12.45,
      "eval_steps_per_second": 12.45,
      "step": 400
    },
    {
      "epoch": 3.6444444444444444,
      "grad_norm": 0.008572138845920563,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0005,
      "step": 410
    },
    {
      "epoch": 3.6444444444444444,
      "eval_loss": 0.006609635427594185,
      "eval_runtime": 8.0332,
      "eval_samples_per_second": 12.448,
      "eval_steps_per_second": 12.448,
      "step": 410
    },
    {
      "epoch": 3.7333333333333334,
      "grad_norm": 0.004443774465471506,
      "learning_rate": 1.5195190384357404e-05,
      "loss": 0.0003,
      "step": 420
    },
    {
      "epoch": 3.7333333333333334,
      "eval_loss": 0.007157918065786362,
      "eval_runtime": 8.039,
      "eval_samples_per_second": 12.439,
      "eval_steps_per_second": 12.439,
      "step": 420
    },
    {
      "epoch": 3.822222222222222,
      "grad_norm": 0.0011405730620026588,
      "learning_rate": 1.1705240714107302e-05,
      "loss": 0.0037,
      "step": 430
    },
    {
      "epoch": 3.822222222222222,
      "eval_loss": 0.006302958354353905,
      "eval_runtime": 8.0294,
      "eval_samples_per_second": 12.454,
      "eval_steps_per_second": 12.454,
      "step": 430
    },
    {
      "epoch": 3.911111111111111,
      "grad_norm": 0.004764024633914232,
      "learning_rate": 8.645454235739903e-06,
      "loss": 0.004,
      "step": 440
    },
    {
      "epoch": 3.911111111111111,
      "eval_loss": 0.005341523326933384,
      "eval_runtime": 8.0207,
      "eval_samples_per_second": 12.468,
      "eval_steps_per_second": 12.468,
      "step": 440
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.004020992666482925,
      "learning_rate": 6.030737921409169e-06,
      "loss": 0.0003,
      "step": 450
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.005209792871028185,
      "eval_runtime": 8.031,
      "eval_samples_per_second": 12.452,
      "eval_steps_per_second": 12.452,
      "step": 450
    },
    {
      "epoch": 4.088888888888889,
      "grad_norm": 0.013617518357932568,
      "learning_rate": 3.873830406168111e-06,
      "loss": 0.0002,
      "step": 460
    },
    {
      "epoch": 4.088888888888889,
      "eval_loss": 0.005074501037597656,
      "eval_runtime": 8.0352,
      "eval_samples_per_second": 12.445,
      "eval_steps_per_second": 12.445,
      "step": 460
    },
    {
      "epoch": 4.177777777777778,
      "grad_norm": 0.003857893170788884,
      "learning_rate": 2.1852399266194314e-06,
      "loss": 0.0002,
      "step": 470
    },
    {
      "epoch": 4.177777777777778,
      "eval_loss": 0.004982742480933666,
      "eval_runtime": 8.0284,
      "eval_samples_per_second": 12.456,
      "eval_steps_per_second": 12.456,
      "step": 470
    },
    {
      "epoch": 4.266666666666667,
      "grad_norm": 0.006583555601537228,
      "learning_rate": 9.731931258429638e-07,
      "loss": 0.0006,
      "step": 480
    },
    {
      "epoch": 4.266666666666667,
      "eval_loss": 0.004874282516539097,
      "eval_runtime": 8.0132,
      "eval_samples_per_second": 12.479,
      "eval_steps_per_second": 12.479,
      "step": 480
    },
    {
      "epoch": 4.355555555555555,
      "grad_norm": 0.004742850083857775,
      "learning_rate": 2.4359497401758024e-07,
      "loss": 0.0005,
      "step": 490
    },
    {
      "epoch": 4.355555555555555,
      "eval_loss": 0.004786411300301552,
      "eval_runtime": 8.0341,
      "eval_samples_per_second": 12.447,
      "eval_steps_per_second": 12.447,
      "step": 490
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 0.010840805247426033,
      "learning_rate": 0.0,
      "loss": 0.0002,
      "step": 500
    },
    {
      "epoch": 4.444444444444445,
      "eval_loss": 0.005105508491396904,
      "eval_runtime": 8.0277,
      "eval_samples_per_second": 12.457,
      "eval_steps_per_second": 12.457,
      "step": 500
    },
    {
      "epoch": 4.444444444444445,
      "step": 500,
      "total_flos": 1.038706354200576e+16,
      "train_loss": 0.1301481350355316,
      "train_runtime": 1611.4177,
      "train_samples_per_second": 2.482,
      "train_steps_per_second": 0.31
    }
  ],
  "logging_steps": 10,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.038706354200576e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}