{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.717391304347826,
  "eval_steps": 50,
  "global_step": 2500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010869565217391304,
      "grad_norm": 2.4570870399475098,
      "learning_rate": 4.981884057971015e-05,
      "loss": 2.3612,
      "step": 10
    },
    {
      "epoch": 0.021739130434782608,
      "grad_norm": 1.2042421102523804,
      "learning_rate": 4.963768115942029e-05,
      "loss": 1.1915,
      "step": 20
    },
    {
      "epoch": 0.03260869565217391,
      "grad_norm": 1.2019174098968506,
      "learning_rate": 4.945652173913044e-05,
      "loss": 0.8992,
      "step": 30
    },
    {
      "epoch": 0.043478260869565216,
      "grad_norm": 1.0546120405197144,
      "learning_rate": 4.9275362318840584e-05,
      "loss": 0.7276,
      "step": 40
    },
    {
      "epoch": 0.05434782608695652,
      "grad_norm": 0.754091203212738,
      "learning_rate": 4.909420289855073e-05,
      "loss": 0.6419,
      "step": 50
    },
    {
      "epoch": 0.05434782608695652,
      "eval_loss": 0.7905128598213196,
      "eval_runtime": 10.8418,
      "eval_samples_per_second": 44.273,
      "eval_steps_per_second": 2.767,
      "step": 50
    },
    {
      "epoch": 0.06521739130434782,
      "grad_norm": 0.7751966714859009,
      "learning_rate": 4.891304347826087e-05,
      "loss": 0.5971,
      "step": 60
    },
    {
      "epoch": 0.07608695652173914,
      "grad_norm": 0.6874057650566101,
      "learning_rate": 4.873188405797102e-05,
      "loss": 0.608,
      "step": 70
    },
    {
      "epoch": 0.08695652173913043,
      "grad_norm": 0.7145748734474182,
      "learning_rate": 4.855072463768116e-05,
      "loss": 0.6111,
      "step": 80
    },
    {
      "epoch": 0.09782608695652174,
      "grad_norm": 0.8841484189033508,
      "learning_rate": 4.836956521739131e-05,
      "loss": 0.5848,
      "step": 90
    },
    {
      "epoch": 0.10869565217391304,
      "grad_norm": 0.7937784790992737,
      "learning_rate": 4.818840579710145e-05,
      "loss": 0.5699,
      "step": 100
    },
    {
      "epoch": 0.10869565217391304,
      "eval_loss": 0.6956210732460022,
      "eval_runtime": 10.863,
      "eval_samples_per_second": 44.187,
      "eval_steps_per_second": 2.762,
      "step": 100
    },
    {
      "epoch": 0.11956521739130435,
      "grad_norm": 0.7818441987037659,
      "learning_rate": 4.80072463768116e-05,
      "loss": 0.5946,
      "step": 110
    },
    {
      "epoch": 0.13043478260869565,
      "grad_norm": 0.8666340112686157,
      "learning_rate": 4.782608695652174e-05,
      "loss": 0.5469,
      "step": 120
    },
    {
      "epoch": 0.14130434782608695,
      "grad_norm": 0.7637468576431274,
      "learning_rate": 4.764492753623189e-05,
      "loss": 0.5307,
      "step": 130
    },
    {
      "epoch": 0.15217391304347827,
      "grad_norm": 0.8282362222671509,
      "learning_rate": 4.746376811594203e-05,
      "loss": 0.5312,
      "step": 140
    },
    {
      "epoch": 0.16304347826086957,
      "grad_norm": 0.9675197601318359,
      "learning_rate": 4.7282608695652177e-05,
      "loss": 0.5486,
      "step": 150
    },
    {
      "epoch": 0.16304347826086957,
      "eval_loss": 0.6567058563232422,
      "eval_runtime": 10.8376,
      "eval_samples_per_second": 44.29,
      "eval_steps_per_second": 2.768,
      "step": 150
    },
    {
      "epoch": 0.17391304347826086,
      "grad_norm": 0.7035924792289734,
      "learning_rate": 4.710144927536232e-05,
      "loss": 0.5553,
      "step": 160
    },
    {
      "epoch": 0.18478260869565216,
      "grad_norm": 0.8957257866859436,
      "learning_rate": 4.6920289855072464e-05,
      "loss": 0.54,
      "step": 170
    },
    {
      "epoch": 0.1956521739130435,
      "grad_norm": 0.8544663190841675,
      "learning_rate": 4.673913043478261e-05,
      "loss": 0.55,
      "step": 180
    },
    {
      "epoch": 0.20652173913043478,
      "grad_norm": 0.7599456310272217,
      "learning_rate": 4.655797101449276e-05,
      "loss": 0.5102,
      "step": 190
    },
    {
      "epoch": 0.21739130434782608,
      "grad_norm": 0.9151259064674377,
      "learning_rate": 4.63768115942029e-05,
      "loss": 0.5372,
      "step": 200
    },
    {
      "epoch": 0.21739130434782608,
      "eval_loss": 0.6319016218185425,
      "eval_runtime": 10.8237,
      "eval_samples_per_second": 44.347,
      "eval_steps_per_second": 2.772,
      "step": 200
    },
    {
      "epoch": 0.22826086956521738,
      "grad_norm": 0.911018431186676,
      "learning_rate": 4.6195652173913046e-05,
      "loss": 0.5354,
      "step": 210
    },
    {
      "epoch": 0.2391304347826087,
      "grad_norm": 0.7792191505432129,
      "learning_rate": 4.601449275362319e-05,
      "loss": 0.5297,
      "step": 220
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.7631202340126038,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 0.5258,
      "step": 230
    },
    {
      "epoch": 0.2608695652173913,
      "grad_norm": 0.7436497211456299,
      "learning_rate": 4.565217391304348e-05,
      "loss": 0.4849,
      "step": 240
    },
    {
      "epoch": 0.2717391304347826,
      "grad_norm": 0.8372677564620972,
      "learning_rate": 4.547101449275363e-05,
      "loss": 0.5131,
      "step": 250
    },
    {
      "epoch": 0.2717391304347826,
      "eval_loss": 0.6155125498771667,
      "eval_runtime": 10.8327,
      "eval_samples_per_second": 44.31,
      "eval_steps_per_second": 2.769,
      "step": 250
    },
    {
      "epoch": 0.2826086956521739,
      "grad_norm": 0.8056390285491943,
      "learning_rate": 4.528985507246377e-05,
      "loss": 0.4675,
      "step": 260
    },
    {
      "epoch": 0.29347826086956524,
      "grad_norm": 0.8015912175178528,
      "learning_rate": 4.5108695652173916e-05,
      "loss": 0.4942,
      "step": 270
    },
    {
      "epoch": 0.30434782608695654,
      "grad_norm": 0.8316887021064758,
      "learning_rate": 4.492753623188406e-05,
      "loss": 0.5025,
      "step": 280
    },
    {
      "epoch": 0.31521739130434784,
      "grad_norm": 0.8440260887145996,
      "learning_rate": 4.4746376811594203e-05,
      "loss": 0.5112,
      "step": 290
    },
    {
      "epoch": 0.32608695652173914,
      "grad_norm": 0.8523954153060913,
      "learning_rate": 4.456521739130435e-05,
      "loss": 0.4973,
      "step": 300
    },
    {
      "epoch": 0.32608695652173914,
      "eval_loss": 0.6024141907691956,
      "eval_runtime": 10.839,
      "eval_samples_per_second": 44.284,
      "eval_steps_per_second": 2.768,
      "step": 300
    },
    {
      "epoch": 0.33695652173913043,
      "grad_norm": 0.8427216410636902,
      "learning_rate": 4.438405797101449e-05,
      "loss": 0.5271,
      "step": 310
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 0.8126628398895264,
      "learning_rate": 4.4202898550724645e-05,
      "loss": 0.4608,
      "step": 320
    },
    {
      "epoch": 0.358695652173913,
      "grad_norm": 0.9947460293769836,
      "learning_rate": 4.4021739130434786e-05,
      "loss": 0.4762,
      "step": 330
    },
    {
      "epoch": 0.3695652173913043,
      "grad_norm": 0.7744721174240112,
      "learning_rate": 4.384057971014493e-05,
      "loss": 0.4993,
      "step": 340
    },
    {
      "epoch": 0.3804347826086957,
      "grad_norm": 0.9483833909034729,
      "learning_rate": 4.365942028985507e-05,
      "loss": 0.497,
      "step": 350
    },
    {
      "epoch": 0.3804347826086957,
      "eval_loss": 0.596432626247406,
      "eval_runtime": 10.8191,
      "eval_samples_per_second": 44.366,
      "eval_steps_per_second": 2.773,
      "step": 350
    },
    {
      "epoch": 0.391304347826087,
      "grad_norm": 0.7573915123939514,
      "learning_rate": 4.347826086956522e-05,
      "loss": 0.469,
      "step": 360
    },
    {
      "epoch": 0.40217391304347827,
      "grad_norm": 0.8013887405395508,
      "learning_rate": 4.329710144927536e-05,
      "loss": 0.4704,
      "step": 370
    },
    {
      "epoch": 0.41304347826086957,
      "grad_norm": 0.8412826657295227,
      "learning_rate": 4.3115942028985515e-05,
      "loss": 0.5033,
      "step": 380
    },
    {
      "epoch": 0.42391304347826086,
      "grad_norm": 0.8203413486480713,
      "learning_rate": 4.2934782608695655e-05,
      "loss": 0.5063,
      "step": 390
    },
    {
      "epoch": 0.43478260869565216,
      "grad_norm": 0.7952508926391602,
      "learning_rate": 4.27536231884058e-05,
      "loss": 0.4706,
      "step": 400
    },
    {
      "epoch": 0.43478260869565216,
      "eval_loss": 0.5875406861305237,
      "eval_runtime": 10.8328,
      "eval_samples_per_second": 44.31,
      "eval_steps_per_second": 2.769,
      "step": 400
    },
    {
      "epoch": 0.44565217391304346,
      "grad_norm": 0.9020015597343445,
      "learning_rate": 4.257246376811594e-05,
      "loss": 0.4786,
      "step": 410
    },
    {
      "epoch": 0.45652173913043476,
      "grad_norm": 0.8588363528251648,
      "learning_rate": 4.239130434782609e-05,
      "loss": 0.482,
      "step": 420
    },
    {
      "epoch": 0.4673913043478261,
      "grad_norm": 0.816198468208313,
      "learning_rate": 4.221014492753623e-05,
      "loss": 0.4683,
      "step": 430
    },
    {
      "epoch": 0.4782608695652174,
      "grad_norm": 0.8713563680648804,
      "learning_rate": 4.202898550724638e-05,
      "loss": 0.4422,
      "step": 440
    },
    {
      "epoch": 0.4891304347826087,
      "grad_norm": 0.8633397817611694,
      "learning_rate": 4.1847826086956525e-05,
      "loss": 0.45,
      "step": 450
    },
    {
      "epoch": 0.4891304347826087,
      "eval_loss": 0.5812788605690002,
      "eval_runtime": 10.8048,
      "eval_samples_per_second": 44.425,
      "eval_steps_per_second": 2.777,
      "step": 450
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.8244722485542297,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.4911,
      "step": 460
    },
    {
      "epoch": 0.5108695652173914,
      "grad_norm": 0.8201929926872253,
      "learning_rate": 4.148550724637681e-05,
      "loss": 0.4611,
      "step": 470
    },
    {
      "epoch": 0.5217391304347826,
      "grad_norm": 0.9646862149238586,
      "learning_rate": 4.130434782608696e-05,
      "loss": 0.4738,
      "step": 480
    },
    {
      "epoch": 0.532608695652174,
      "grad_norm": 0.9149807095527649,
      "learning_rate": 4.11231884057971e-05,
      "loss": 0.4603,
      "step": 490
    },
    {
      "epoch": 0.5434782608695652,
      "grad_norm": 0.8054277896881104,
      "learning_rate": 4.094202898550725e-05,
      "loss": 0.466,
      "step": 500
    },
    {
      "epoch": 0.5434782608695652,
      "eval_loss": 0.573806881904602,
      "eval_runtime": 10.7906,
      "eval_samples_per_second": 44.483,
      "eval_steps_per_second": 2.78,
      "step": 500
    },
    {
      "epoch": 0.5543478260869565,
      "grad_norm": 0.9110711812973022,
      "learning_rate": 4.076086956521739e-05,
      "loss": 0.4777,
      "step": 510
    },
    {
      "epoch": 0.5652173913043478,
      "grad_norm": 0.8030044436454773,
      "learning_rate": 4.057971014492754e-05,
      "loss": 0.4965,
      "step": 520
    },
    {
      "epoch": 0.5760869565217391,
      "grad_norm": 0.8847204446792603,
      "learning_rate": 4.039855072463768e-05,
      "loss": 0.4843,
      "step": 530
    },
    {
      "epoch": 0.5869565217391305,
      "grad_norm": 0.9041047692298889,
      "learning_rate": 4.021739130434783e-05,
      "loss": 0.4548,
      "step": 540
    },
    {
      "epoch": 0.5978260869565217,
      "grad_norm": 0.8136462569236755,
      "learning_rate": 4.003623188405797e-05,
      "loss": 0.4543,
      "step": 550
    },
    {
      "epoch": 0.5978260869565217,
      "eval_loss": 0.5662083029747009,
      "eval_runtime": 10.7748,
      "eval_samples_per_second": 44.549,
      "eval_steps_per_second": 2.784,
      "step": 550
    },
    {
      "epoch": 0.6086956521739131,
      "grad_norm": 0.9470314383506775,
      "learning_rate": 3.985507246376812e-05,
      "loss": 0.4852,
      "step": 560
    },
    {
      "epoch": 0.6195652173913043,
      "grad_norm": 0.9280235171318054,
      "learning_rate": 3.9673913043478264e-05,
      "loss": 0.4655,
      "step": 570
    },
    {
      "epoch": 0.6304347826086957,
      "grad_norm": 0.9118008017539978,
      "learning_rate": 3.9492753623188405e-05,
      "loss": 0.449,
      "step": 580
    },
    {
      "epoch": 0.6413043478260869,
      "grad_norm": 0.9286589026451111,
      "learning_rate": 3.931159420289855e-05,
      "loss": 0.448,
      "step": 590
    },
    {
      "epoch": 0.6521739130434783,
      "grad_norm": 0.7761940956115723,
      "learning_rate": 3.91304347826087e-05,
      "loss": 0.4363,
      "step": 600
    },
    {
      "epoch": 0.6521739130434783,
      "eval_loss": 0.5645204186439514,
      "eval_runtime": 10.7763,
      "eval_samples_per_second": 44.542,
      "eval_steps_per_second": 2.784,
      "step": 600
    },
    {
      "epoch": 0.6630434782608695,
      "grad_norm": 0.9244002103805542,
      "learning_rate": 3.8949275362318846e-05,
      "loss": 0.4354,
      "step": 610
    },
    {
      "epoch": 0.6739130434782609,
      "grad_norm": 0.8263306021690369,
      "learning_rate": 3.876811594202899e-05,
      "loss": 0.4951,
      "step": 620
    },
    {
      "epoch": 0.6847826086956522,
      "grad_norm": 0.8908408284187317,
      "learning_rate": 3.8586956521739134e-05,
      "loss": 0.4401,
      "step": 630
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 0.8497889637947083,
      "learning_rate": 3.8405797101449274e-05,
      "loss": 0.4514,
      "step": 640
    },
    {
      "epoch": 0.7065217391304348,
      "grad_norm": 0.8306752443313599,
      "learning_rate": 3.822463768115942e-05,
      "loss": 0.4503,
      "step": 650
    },
    {
      "epoch": 0.7065217391304348,
      "eval_loss": 0.5576011538505554,
      "eval_runtime": 10.7704,
      "eval_samples_per_second": 44.567,
      "eval_steps_per_second": 2.785,
      "step": 650
    },
    {
      "epoch": 0.717391304347826,
      "grad_norm": 0.898961067199707,
      "learning_rate": 3.804347826086957e-05,
      "loss": 0.4185,
      "step": 660
    },
    {
      "epoch": 0.7282608695652174,
      "grad_norm": 0.8870590925216675,
      "learning_rate": 3.7862318840579716e-05,
      "loss": 0.4303,
      "step": 670
    },
    {
      "epoch": 0.7391304347826086,
      "grad_norm": 0.8863015174865723,
      "learning_rate": 3.7681159420289856e-05,
      "loss": 0.4356,
      "step": 680
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.852130651473999,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.4756,
      "step": 690
    },
    {
      "epoch": 0.7608695652173914,
      "grad_norm": 0.8342716097831726,
      "learning_rate": 3.7318840579710144e-05,
      "loss": 0.4955,
      "step": 700
    },
    {
      "epoch": 0.7608695652173914,
      "eval_loss": 0.5521832704544067,
      "eval_runtime": 10.7828,
      "eval_samples_per_second": 44.515,
      "eval_steps_per_second": 2.782,
      "step": 700
    },
    {
      "epoch": 0.7717391304347826,
      "grad_norm": 0.8382502794265747,
      "learning_rate": 3.713768115942029e-05,
      "loss": 0.4435,
      "step": 710
    },
    {
      "epoch": 0.782608695652174,
      "grad_norm": 0.9789420962333679,
      "learning_rate": 3.695652173913043e-05,
      "loss": 0.3957,
      "step": 720
    },
    {
      "epoch": 0.7934782608695652,
      "grad_norm": 0.8945524096488953,
      "learning_rate": 3.6775362318840586e-05,
      "loss": 0.4565,
      "step": 730
    },
    {
      "epoch": 0.8043478260869565,
      "grad_norm": 0.9632206559181213,
      "learning_rate": 3.6594202898550726e-05,
      "loss": 0.4312,
      "step": 740
    },
    {
      "epoch": 0.8152173913043478,
      "grad_norm": 0.8000399470329285,
      "learning_rate": 3.641304347826087e-05,
      "loss": 0.429,
      "step": 750
    },
    {
      "epoch": 0.8152173913043478,
      "eval_loss": 0.550889790058136,
      "eval_runtime": 10.7987,
      "eval_samples_per_second": 44.45,
      "eval_steps_per_second": 2.778,
      "step": 750
    },
    {
      "epoch": 0.8260869565217391,
      "grad_norm": 0.8817518353462219,
      "learning_rate": 3.6231884057971014e-05,
      "loss": 0.4464,
      "step": 760
    },
    {
      "epoch": 0.8369565217391305,
      "grad_norm": 0.9800861477851868,
      "learning_rate": 3.605072463768116e-05,
      "loss": 0.4577,
      "step": 770
    },
    {
      "epoch": 0.8478260869565217,
      "grad_norm": 0.846571683883667,
      "learning_rate": 3.58695652173913e-05,
      "loss": 0.4169,
      "step": 780
    },
    {
      "epoch": 0.8586956521739131,
      "grad_norm": 0.8653192520141602,
      "learning_rate": 3.568840579710145e-05,
      "loss": 0.4195,
      "step": 790
    },
    {
      "epoch": 0.8695652173913043,
      "grad_norm": 0.8515974283218384,
      "learning_rate": 3.5507246376811596e-05,
      "loss": 0.4366,
      "step": 800
    },
    {
      "epoch": 0.8695652173913043,
      "eval_loss": 0.543337881565094,
      "eval_runtime": 10.8166,
      "eval_samples_per_second": 44.376,
      "eval_steps_per_second": 2.774,
      "step": 800
    },
    {
      "epoch": 0.8804347826086957,
      "grad_norm": 0.947827935218811,
      "learning_rate": 3.532608695652174e-05,
      "loss": 0.4498,
      "step": 810
    },
    {
      "epoch": 0.8913043478260869,
      "grad_norm": 0.866033136844635,
      "learning_rate": 3.514492753623188e-05,
      "loss": 0.4495,
      "step": 820
    },
    {
      "epoch": 0.9021739130434783,
      "grad_norm": 0.8692104816436768,
      "learning_rate": 3.496376811594203e-05,
      "loss": 0.4502,
      "step": 830
    },
    {
      "epoch": 0.9130434782608695,
      "grad_norm": 0.8841784596443176,
      "learning_rate": 3.478260869565218e-05,
      "loss": 0.4577,
      "step": 840
    },
    {
      "epoch": 0.9239130434782609,
      "grad_norm": 0.9257445335388184,
      "learning_rate": 3.460144927536232e-05,
      "loss": 0.4329,
      "step": 850
    },
    {
      "epoch": 0.9239130434782609,
      "eval_loss": 0.5429030060768127,
      "eval_runtime": 10.8284,
      "eval_samples_per_second": 44.328,
      "eval_steps_per_second": 2.77,
      "step": 850
    },
    {
      "epoch": 0.9347826086956522,
      "grad_norm": 0.8753738403320312,
      "learning_rate": 3.4420289855072465e-05,
      "loss": 0.422,
      "step": 860
    },
    {
      "epoch": 0.9456521739130435,
      "grad_norm": 0.8853139877319336,
      "learning_rate": 3.423913043478261e-05,
      "loss": 0.4357,
      "step": 870
    },
    {
      "epoch": 0.9565217391304348,
      "grad_norm": 0.8637978434562683,
      "learning_rate": 3.405797101449276e-05,
      "loss": 0.4254,
      "step": 880
    },
    {
      "epoch": 0.967391304347826,
      "grad_norm": 0.9090889096260071,
      "learning_rate": 3.38768115942029e-05,
      "loss": 0.4526,
      "step": 890
    },
    {
      "epoch": 0.9782608695652174,
      "grad_norm": 0.8591863512992859,
      "learning_rate": 3.369565217391305e-05,
      "loss": 0.4566,
      "step": 900
    },
    {
      "epoch": 0.9782608695652174,
      "eval_loss": 0.5399300456047058,
      "eval_runtime": 10.8533,
      "eval_samples_per_second": 44.226,
      "eval_steps_per_second": 2.764,
      "step": 900
    },
    {
      "epoch": 0.9891304347826086,
      "grad_norm": 0.8273764848709106,
      "learning_rate": 3.351449275362319e-05,
      "loss": 0.4182,
      "step": 910
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.5736610889434814,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.4548,
      "step": 920
    },
    {
      "epoch": 1.0108695652173914,
      "grad_norm": 0.8933672904968262,
      "learning_rate": 3.3152173913043475e-05,
      "loss": 0.3834,
      "step": 930
    },
    {
      "epoch": 1.0217391304347827,
      "grad_norm": 0.9641056060791016,
      "learning_rate": 3.297101449275363e-05,
      "loss": 0.3962,
      "step": 940
    },
    {
      "epoch": 1.0326086956521738,
      "grad_norm": 0.8840826153755188,
      "learning_rate": 3.278985507246377e-05,
      "loss": 0.3886,
      "step": 950
    },
    {
      "epoch": 1.0326086956521738,
      "eval_loss": 0.5417191982269287,
      "eval_runtime": 10.828,
      "eval_samples_per_second": 44.33,
      "eval_steps_per_second": 2.771,
      "step": 950
    },
    {
      "epoch": 1.0434782608695652,
      "grad_norm": 0.9005197286605835,
      "learning_rate": 3.260869565217392e-05,
      "loss": 0.3887,
      "step": 960
    },
    {
      "epoch": 1.0543478260869565,
      "grad_norm": 1.0132256746292114,
      "learning_rate": 3.242753623188406e-05,
      "loss": 0.4026,
      "step": 970
    },
    {
      "epoch": 1.065217391304348,
      "grad_norm": 1.1139558553695679,
      "learning_rate": 3.2246376811594205e-05,
      "loss": 0.3703,
      "step": 980
    },
    {
      "epoch": 1.0760869565217392,
      "grad_norm": 0.9440574049949646,
      "learning_rate": 3.2065217391304345e-05,
      "loss": 0.3766,
      "step": 990
    },
    {
      "epoch": 1.0869565217391304,
      "grad_norm": 1.0028133392333984,
      "learning_rate": 3.188405797101449e-05,
      "loss": 0.4038,
      "step": 1000
    },
    {
      "epoch": 1.0869565217391304,
      "eval_loss": 0.5413796901702881,
      "eval_runtime": 10.822,
      "eval_samples_per_second": 44.354,
      "eval_steps_per_second": 2.772,
      "step": 1000
    },
    {
      "epoch": 1.0978260869565217,
      "grad_norm": 0.9884430766105652,
      "learning_rate": 3.170289855072464e-05,
      "loss": 0.4134,
      "step": 1010
    },
    {
      "epoch": 1.108695652173913,
      "grad_norm": 0.920870840549469,
      "learning_rate": 3.152173913043479e-05,
      "loss": 0.3837,
      "step": 1020
    },
    {
      "epoch": 1.1195652173913044,
      "grad_norm": 0.942890465259552,
      "learning_rate": 3.134057971014493e-05,
      "loss": 0.38,
      "step": 1030
    },
    {
      "epoch": 1.1304347826086956,
      "grad_norm": 1.0265600681304932,
      "learning_rate": 3.1159420289855074e-05,
      "loss": 0.3988,
      "step": 1040
    },
    {
      "epoch": 1.141304347826087,
      "grad_norm": 0.9994137287139893,
      "learning_rate": 3.0978260869565215e-05,
      "loss": 0.3649,
      "step": 1050
    },
    {
      "epoch": 1.141304347826087,
      "eval_loss": 0.537377655506134,
      "eval_runtime": 10.7968,
      "eval_samples_per_second": 44.458,
      "eval_steps_per_second": 2.779,
      "step": 1050
    },
    {
      "epoch": 1.1521739130434783,
      "grad_norm": 0.9697039723396301,
      "learning_rate": 3.079710144927536e-05,
      "loss": 0.393,
      "step": 1060
    },
    {
      "epoch": 1.1630434782608696,
      "grad_norm": 0.9805816411972046,
      "learning_rate": 3.061594202898551e-05,
      "loss": 0.3594,
      "step": 1070
    },
    {
      "epoch": 1.1739130434782608,
      "grad_norm": 0.9799211621284485,
      "learning_rate": 3.0434782608695656e-05,
      "loss": 0.387,
      "step": 1080
    },
    {
      "epoch": 1.184782608695652,
      "grad_norm": 0.9287502765655518,
      "learning_rate": 3.02536231884058e-05,
      "loss": 0.3889,
      "step": 1090
    },
    {
      "epoch": 1.1956521739130435,
      "grad_norm": 1.042643666267395,
      "learning_rate": 3.0072463768115944e-05,
      "loss": 0.3894,
      "step": 1100
    },
    {
      "epoch": 1.1956521739130435,
      "eval_loss": 0.5378063917160034,
      "eval_runtime": 10.77,
      "eval_samples_per_second": 44.568,
      "eval_steps_per_second": 2.786,
      "step": 1100
    },
    {
      "epoch": 1.2065217391304348,
      "grad_norm": 0.963258683681488,
      "learning_rate": 2.9891304347826088e-05,
      "loss": 0.3824,
      "step": 1110
    },
    {
      "epoch": 1.2173913043478262,
      "grad_norm": 0.939275860786438,
      "learning_rate": 2.971014492753623e-05,
      "loss": 0.3898,
      "step": 1120
    },
    {
      "epoch": 1.2282608695652173,
      "grad_norm": 1.1144332885742188,
      "learning_rate": 2.9528985507246375e-05,
      "loss": 0.3754,
      "step": 1130
    },
    {
      "epoch": 1.2391304347826086,
      "grad_norm": 0.999047577381134,
      "learning_rate": 2.9347826086956526e-05,
      "loss": 0.3876,
      "step": 1140
    },
    {
      "epoch": 1.25,
      "grad_norm": 1.024510383605957,
      "learning_rate": 2.916666666666667e-05,
      "loss": 0.3782,
      "step": 1150
    },
    {
      "epoch": 1.25,
      "eval_loss": 0.5348225831985474,
      "eval_runtime": 10.7748,
      "eval_samples_per_second": 44.548,
      "eval_steps_per_second": 2.784,
      "step": 1150
    },
    {
      "epoch": 1.2608695652173914,
      "grad_norm": 1.0736693143844604,
      "learning_rate": 2.8985507246376814e-05,
      "loss": 0.3895,
      "step": 1160
    },
    {
      "epoch": 1.2717391304347827,
      "grad_norm": 0.9759789705276489,
      "learning_rate": 2.8804347826086957e-05,
      "loss": 0.3989,
      "step": 1170
    },
    {
      "epoch": 1.2826086956521738,
      "grad_norm": 0.8532370328903198,
      "learning_rate": 2.86231884057971e-05,
      "loss": 0.3814,
      "step": 1180
    },
    {
      "epoch": 1.2934782608695652,
      "grad_norm": 0.9838298559188843,
      "learning_rate": 2.8442028985507245e-05,
      "loss": 0.3864,
      "step": 1190
    },
    {
      "epoch": 1.3043478260869565,
      "grad_norm": 0.8785775899887085,
      "learning_rate": 2.826086956521739e-05,
      "loss": 0.3676,
      "step": 1200
    },
    {
      "epoch": 1.3043478260869565,
      "eval_loss": 0.5335877537727356,
      "eval_runtime": 10.7959,
      "eval_samples_per_second": 44.461,
      "eval_steps_per_second": 2.779,
      "step": 1200
    },
    {
      "epoch": 1.315217391304348,
      "grad_norm": 0.9719653129577637,
      "learning_rate": 2.807971014492754e-05,
      "loss": 0.3913,
      "step": 1210
    },
    {
      "epoch": 1.3260869565217392,
      "grad_norm": 0.9099355340003967,
      "learning_rate": 2.7898550724637683e-05,
      "loss": 0.4019,
      "step": 1220
    },
    {
      "epoch": 1.3369565217391304,
      "grad_norm": 0.9192841649055481,
      "learning_rate": 2.7717391304347827e-05,
      "loss": 0.3921,
      "step": 1230
    },
    {
      "epoch": 1.3478260869565217,
      "grad_norm": 1.1005280017852783,
      "learning_rate": 2.753623188405797e-05,
      "loss": 0.3892,
      "step": 1240
    },
    {
      "epoch": 1.358695652173913,
      "grad_norm": 0.9969606995582581,
      "learning_rate": 2.7355072463768118e-05,
      "loss": 0.3756,
      "step": 1250
    },
    {
      "epoch": 1.358695652173913,
      "eval_loss": 0.5348997712135315,
      "eval_runtime": 10.8018,
      "eval_samples_per_second": 44.437,
      "eval_steps_per_second": 2.777,
      "step": 1250
    },
    {
      "epoch": 1.3695652173913042,
      "grad_norm": 1.1843461990356445,
      "learning_rate": 2.7173913043478262e-05,
      "loss": 0.3751,
      "step": 1260
    },
    {
      "epoch": 1.3804347826086958,
      "grad_norm": 0.9097041487693787,
      "learning_rate": 2.6992753623188406e-05,
      "loss": 0.3852,
      "step": 1270
    },
    {
      "epoch": 1.391304347826087,
      "grad_norm": 0.9780197739601135,
      "learning_rate": 2.6811594202898553e-05,
      "loss": 0.3776,
      "step": 1280
    },
    {
      "epoch": 1.4021739130434783,
      "grad_norm": 1.0949311256408691,
      "learning_rate": 2.66304347826087e-05,
      "loss": 0.3517,
      "step": 1290
    },
    {
      "epoch": 1.4130434782608696,
      "grad_norm": 1.0105839967727661,
      "learning_rate": 2.6449275362318844e-05,
      "loss": 0.384,
      "step": 1300
    },
    {
      "epoch": 1.4130434782608696,
      "eval_loss": 0.5327685475349426,
      "eval_runtime": 10.8023,
      "eval_samples_per_second": 44.435,
      "eval_steps_per_second": 2.777,
      "step": 1300
    },
    {
      "epoch": 1.4239130434782608,
      "grad_norm": 1.0113478899002075,
      "learning_rate": 2.6268115942028988e-05,
      "loss": 0.3774,
      "step": 1310
    },
    {
      "epoch": 1.434782608695652,
      "grad_norm": 1.0456714630126953,
      "learning_rate": 2.608695652173913e-05,
      "loss": 0.3618,
      "step": 1320
    },
    {
      "epoch": 1.4456521739130435,
      "grad_norm": 1.048971176147461,
      "learning_rate": 2.5905797101449275e-05,
      "loss": 0.3707,
      "step": 1330
    },
    {
      "epoch": 1.4565217391304348,
      "grad_norm": 0.9918339252471924,
      "learning_rate": 2.572463768115942e-05,
      "loss": 0.3979,
      "step": 1340
    },
    {
      "epoch": 1.4673913043478262,
      "grad_norm": 0.8943142890930176,
      "learning_rate": 2.554347826086957e-05,
      "loss": 0.3647,
      "step": 1350
    },
    {
      "epoch": 1.4673913043478262,
      "eval_loss": 0.530137836933136,
      "eval_runtime": 10.8212,
      "eval_samples_per_second": 44.357,
      "eval_steps_per_second": 2.772,
      "step": 1350
    },
    {
      "epoch": 1.4782608695652173,
      "grad_norm": 1.1170648336410522,
      "learning_rate": 2.5362318840579714e-05,
      "loss": 0.3752,
      "step": 1360
    },
    {
      "epoch": 1.4891304347826086,
      "grad_norm": 1.0574833154678345,
      "learning_rate": 2.5181159420289857e-05,
      "loss": 0.4005,
      "step": 1370
    },
    {
      "epoch": 1.5,
      "grad_norm": 1.0728522539138794,
      "learning_rate": 2.5e-05,
      "loss": 0.404,
      "step": 1380
    },
    {
      "epoch": 1.5108695652173914,
      "grad_norm": 1.0159659385681152,
      "learning_rate": 2.4818840579710145e-05,
      "loss": 0.4051,
      "step": 1390
    },
    {
      "epoch": 1.5217391304347827,
      "grad_norm": 0.9714246392250061,
      "learning_rate": 2.4637681159420292e-05,
      "loss": 0.3775,
      "step": 1400
    },
    {
      "epoch": 1.5217391304347827,
      "eval_loss": 0.528946578502655,
      "eval_runtime": 10.8248,
      "eval_samples_per_second": 44.343,
      "eval_steps_per_second": 2.771,
      "step": 1400
    },
    {
      "epoch": 1.5326086956521738,
      "grad_norm": 1.0848588943481445,
      "learning_rate": 2.4456521739130436e-05,
      "loss": 0.3703,
      "step": 1410
    },
    {
      "epoch": 1.5434782608695652,
      "grad_norm": 1.0333784818649292,
      "learning_rate": 2.427536231884058e-05,
      "loss": 0.3692,
      "step": 1420
    },
    {
      "epoch": 1.5543478260869565,
      "grad_norm": 0.9682796597480774,
      "learning_rate": 2.4094202898550724e-05,
      "loss": 0.3722,
      "step": 1430
    },
    {
      "epoch": 1.5652173913043477,
      "grad_norm": 1.0484014749526978,
      "learning_rate": 2.391304347826087e-05,
      "loss": 0.3763,
      "step": 1440
    },
    {
      "epoch": 1.5760869565217392,
      "grad_norm": 0.9075008630752563,
      "learning_rate": 2.3731884057971015e-05,
      "loss": 0.3594,
      "step": 1450
    },
    {
      "epoch": 1.5760869565217392,
      "eval_loss": 0.5278663039207458,
      "eval_runtime": 10.831,
      "eval_samples_per_second": 44.317,
      "eval_steps_per_second": 2.77,
      "step": 1450
    },
    {
      "epoch": 1.5869565217391304,
      "grad_norm": 0.9710797667503357,
      "learning_rate": 2.355072463768116e-05,
      "loss": 0.3821,
      "step": 1460
    },
    {
      "epoch": 1.5978260869565217,
      "grad_norm": 1.0220218896865845,
      "learning_rate": 2.3369565217391306e-05,
      "loss": 0.3623,
      "step": 1470
    },
    {
      "epoch": 1.608695652173913,
      "grad_norm": 1.0994117259979248,
      "learning_rate": 2.318840579710145e-05,
      "loss": 0.3735,
      "step": 1480
    },
    {
      "epoch": 1.6195652173913042,
      "grad_norm": 0.9513503313064575,
      "learning_rate": 2.3007246376811593e-05,
      "loss": 0.3743,
      "step": 1490
    },
    {
      "epoch": 1.6304347826086958,
      "grad_norm": 0.9903764128684998,
      "learning_rate": 2.282608695652174e-05,
      "loss": 0.3661,
      "step": 1500
    },
    {
      "epoch": 1.6304347826086958,
      "eval_loss": 0.5267059803009033,
      "eval_runtime": 10.8461,
      "eval_samples_per_second": 44.256,
      "eval_steps_per_second": 2.766,
      "step": 1500
    },
    {
      "epoch": 1.641304347826087,
      "grad_norm": 0.9712284207344055,
      "learning_rate": 2.2644927536231884e-05,
      "loss": 0.3662,
      "step": 1510
    },
    {
      "epoch": 1.6521739130434783,
      "grad_norm": 0.9667128920555115,
      "learning_rate": 2.246376811594203e-05,
      "loss": 0.3767,
      "step": 1520
    },
    {
      "epoch": 1.6630434782608696,
      "grad_norm": 0.9828950762748718,
      "learning_rate": 2.2282608695652175e-05,
      "loss": 0.3766,
      "step": 1530
    },
    {
      "epoch": 1.6739130434782608,
      "grad_norm": 1.0777678489685059,
      "learning_rate": 2.2101449275362323e-05,
      "loss": 0.3791,
      "step": 1540
    },
    {
      "epoch": 1.6847826086956523,
      "grad_norm": 1.006035566329956,
      "learning_rate": 2.1920289855072466e-05,
      "loss": 0.3961,
      "step": 1550
    },
    {
      "epoch": 1.6847826086956523,
      "eval_loss": 0.5241792798042297,
      "eval_runtime": 10.8288,
      "eval_samples_per_second": 44.326,
      "eval_steps_per_second": 2.77,
      "step": 1550
    },
    {
      "epoch": 1.6956521739130435,
      "grad_norm": 0.9397144913673401,
      "learning_rate": 2.173913043478261e-05,
      "loss": 0.3659,
      "step": 1560
    },
    {
      "epoch": 1.7065217391304348,
      "grad_norm": 1.1106013059616089,
      "learning_rate": 2.1557971014492757e-05,
      "loss": 0.381,
      "step": 1570
    },
    {
      "epoch": 1.7173913043478262,
      "grad_norm": 0.9626741409301758,
      "learning_rate": 2.13768115942029e-05,
      "loss": 0.3751,
      "step": 1580
    },
    {
      "epoch": 1.7282608695652173,
      "grad_norm": 1.139076828956604,
      "learning_rate": 2.1195652173913045e-05,
      "loss": 0.3888,
      "step": 1590
    },
    {
      "epoch": 1.7391304347826086,
      "grad_norm": 1.0335912704467773,
      "learning_rate": 2.101449275362319e-05,
      "loss": 0.3758,
      "step": 1600
    },
    {
      "epoch": 1.7391304347826086,
      "eval_loss": 0.5224525332450867,
      "eval_runtime": 10.8033,
      "eval_samples_per_second": 44.431,
      "eval_steps_per_second": 2.777,
      "step": 1600
    },
    {
      "epoch": 1.75,
      "grad_norm": 1.0831104516983032,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 0.3429,
      "step": 1610
    },
    {
      "epoch": 1.7608695652173914,
      "grad_norm": 1.0405676364898682,
      "learning_rate": 2.065217391304348e-05,
      "loss": 0.3506,
      "step": 1620
    },
    {
      "epoch": 1.7717391304347827,
      "grad_norm": 0.9948307275772095,
      "learning_rate": 2.0471014492753624e-05,
      "loss": 0.3528,
      "step": 1630
    },
    {
      "epoch": 1.7826086956521738,
      "grad_norm": 1.103110671043396,
      "learning_rate": 2.028985507246377e-05,
      "loss": 0.3779,
      "step": 1640
    },
    {
      "epoch": 1.7934782608695652,
      "grad_norm": 1.0501149892807007,
      "learning_rate": 2.0108695652173915e-05,
      "loss": 0.3957,
      "step": 1650
    },
    {
      "epoch": 1.7934782608695652,
      "eval_loss": 0.5198243260383606,
      "eval_runtime": 10.7809,
      "eval_samples_per_second": 44.523,
      "eval_steps_per_second": 2.783,
      "step": 1650
    },
    {
      "epoch": 1.8043478260869565,
      "grad_norm": 1.0968273878097534,
      "learning_rate": 1.992753623188406e-05,
      "loss": 0.3746,
      "step": 1660
    },
    {
      "epoch": 1.8152173913043477,
      "grad_norm": 1.0902262926101685,
      "learning_rate": 1.9746376811594202e-05,
      "loss": 0.3688,
      "step": 1670
    },
    {
      "epoch": 1.8260869565217392,
      "grad_norm": 1.1077326536178589,
      "learning_rate": 1.956521739130435e-05,
      "loss": 0.3274,
      "step": 1680
    },
    {
      "epoch": 1.8369565217391304,
      "grad_norm": 1.005223035812378,
      "learning_rate": 1.9384057971014493e-05,
      "loss": 0.3858,
      "step": 1690
    },
    {
      "epoch": 1.8478260869565217,
      "grad_norm": 1.131982445716858,
      "learning_rate": 1.9202898550724637e-05,
      "loss": 0.3676,
      "step": 1700
    },
    {
      "epoch": 1.8478260869565217,
      "eval_loss": 0.5192354321479797,
      "eval_runtime": 10.7892,
      "eval_samples_per_second": 44.489,
      "eval_steps_per_second": 2.781,
      "step": 1700
    },
    {
      "epoch": 1.858695652173913,
      "grad_norm": 1.0396854877471924,
      "learning_rate": 1.9021739130434784e-05,
      "loss": 0.3738,
      "step": 1710
    },
    {
      "epoch": 1.8695652173913042,
      "grad_norm": 1.1090553998947144,
      "learning_rate": 1.8840579710144928e-05,
      "loss": 0.3695,
      "step": 1720
    },
    {
      "epoch": 1.8804347826086958,
      "grad_norm": 1.081115484237671,
      "learning_rate": 1.8659420289855072e-05,
      "loss": 0.3549,
      "step": 1730
    },
    {
      "epoch": 1.891304347826087,
      "grad_norm": 1.1276469230651855,
      "learning_rate": 1.8478260869565216e-05,
      "loss": 0.364,
      "step": 1740
    },
    {
      "epoch": 1.9021739130434783,
      "grad_norm": 1.0103682279586792,
      "learning_rate": 1.8297101449275363e-05,
      "loss": 0.3781,
      "step": 1750
    },
    {
      "epoch": 1.9021739130434783,
      "eval_loss": 0.5211134552955627,
      "eval_runtime": 10.7922,
      "eval_samples_per_second": 44.477,
      "eval_steps_per_second": 2.78,
      "step": 1750
    },
    {
      "epoch": 1.9130434782608696,
      "grad_norm": 0.9383260011672974,
      "learning_rate": 1.8115942028985507e-05,
      "loss": 0.3469,
      "step": 1760
    },
    {
      "epoch": 1.9239130434782608,
      "grad_norm": 1.095609188079834,
      "learning_rate": 1.793478260869565e-05,
      "loss": 0.3747,
      "step": 1770
    },
    {
      "epoch": 1.9347826086956523,
      "grad_norm": 1.0580706596374512,
      "learning_rate": 1.7753623188405798e-05,
      "loss": 0.3751,
      "step": 1780
    },
    {
      "epoch": 1.9456521739130435,
      "grad_norm": 0.9539552927017212,
      "learning_rate": 1.757246376811594e-05,
      "loss": 0.3638,
      "step": 1790
    },
    {
      "epoch": 1.9565217391304348,
      "grad_norm": 1.0025752782821655,
      "learning_rate": 1.739130434782609e-05,
      "loss": 0.3851,
      "step": 1800
    },
    {
      "epoch": 1.9565217391304348,
      "eval_loss": 0.516950249671936,
      "eval_runtime": 10.7904,
      "eval_samples_per_second": 44.484,
      "eval_steps_per_second": 2.78,
      "step": 1800
    },
    {
      "epoch": 1.9673913043478262,
      "grad_norm": 1.0852687358856201,
      "learning_rate": 1.7210144927536233e-05,
      "loss": 0.3472,
      "step": 1810
    },
    {
      "epoch": 1.9782608695652173,
      "grad_norm": 1.1232205629348755,
      "learning_rate": 1.702898550724638e-05,
      "loss": 0.3591,
      "step": 1820
    },
    {
      "epoch": 1.9891304347826086,
      "grad_norm": 1.0539883375167847,
      "learning_rate": 1.6847826086956524e-05,
      "loss": 0.3657,
      "step": 1830
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.387427806854248,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.393,
      "step": 1840
    },
    {
      "epoch": 2.010869565217391,
      "grad_norm": 1.0311148166656494,
      "learning_rate": 1.6485507246376815e-05,
      "loss": 0.3364,
      "step": 1850
    },
    {
      "epoch": 2.010869565217391,
      "eval_loss": 0.5237263441085815,
      "eval_runtime": 10.8116,
      "eval_samples_per_second": 44.397,
      "eval_steps_per_second": 2.775,
      "step": 1850
    },
    {
      "epoch": 2.0217391304347827,
      "grad_norm": 1.2049050331115723,
      "learning_rate": 1.630434782608696e-05,
      "loss": 0.336,
      "step": 1860
    },
    {
      "epoch": 2.032608695652174,
      "grad_norm": 1.0490036010742188,
      "learning_rate": 1.6123188405797102e-05,
      "loss": 0.3243,
      "step": 1870
    },
    {
      "epoch": 2.0434782608695654,
      "grad_norm": 1.131861686706543,
      "learning_rate": 1.5942028985507246e-05,
      "loss": 0.3311,
      "step": 1880
    },
    {
      "epoch": 2.0543478260869565,
      "grad_norm": 1.0843744277954102,
      "learning_rate": 1.5760869565217393e-05,
      "loss": 0.3107,
      "step": 1890
    },
    {
      "epoch": 2.0652173913043477,
      "grad_norm": 1.1070170402526855,
      "learning_rate": 1.5579710144927537e-05,
      "loss": 0.3357,
      "step": 1900
    },
    {
      "epoch": 2.0652173913043477,
      "eval_loss": 0.5324747562408447,
      "eval_runtime": 10.8165,
      "eval_samples_per_second": 44.377,
      "eval_steps_per_second": 2.774,
      "step": 1900
    },
    {
      "epoch": 2.0760869565217392,
      "grad_norm": 1.1747304201126099,
      "learning_rate": 1.539855072463768e-05,
      "loss": 0.3301,
      "step": 1910
    },
    {
      "epoch": 2.0869565217391304,
      "grad_norm": 1.151397943496704,
      "learning_rate": 1.5217391304347828e-05,
      "loss": 0.3313,
      "step": 1920
    },
    {
      "epoch": 2.097826086956522,
      "grad_norm": 1.1628749370574951,
      "learning_rate": 1.5036231884057972e-05,
      "loss": 0.322,
      "step": 1930
    },
    {
      "epoch": 2.108695652173913,
      "grad_norm": 1.1999317407608032,
      "learning_rate": 1.4855072463768116e-05,
      "loss": 0.3174,
      "step": 1940
    },
    {
      "epoch": 2.119565217391304,
      "grad_norm": 1.1956512928009033,
      "learning_rate": 1.4673913043478263e-05,
      "loss": 0.3281,
      "step": 1950
    },
    {
      "epoch": 2.119565217391304,
      "eval_loss": 0.5320965647697449,
      "eval_runtime": 10.8261,
      "eval_samples_per_second": 44.337,
      "eval_steps_per_second": 2.771,
      "step": 1950
    },
    {
      "epoch": 2.130434782608696,
      "grad_norm": 1.0482760667800903,
      "learning_rate": 1.4492753623188407e-05,
      "loss": 0.3193,
      "step": 1960
    },
    {
      "epoch": 2.141304347826087,
      "grad_norm": 1.72954261302948,
      "learning_rate": 1.431159420289855e-05,
      "loss": 0.3418,
      "step": 1970
    },
    {
      "epoch": 2.1521739130434785,
      "grad_norm": 1.2846759557724,
      "learning_rate": 1.4130434782608694e-05,
      "loss": 0.3274,
      "step": 1980
    },
    {
      "epoch": 2.1630434782608696,
      "grad_norm": 1.2016099691390991,
      "learning_rate": 1.3949275362318842e-05,
      "loss": 0.3305,
      "step": 1990
    },
    {
      "epoch": 2.1739130434782608,
      "grad_norm": 1.1600474119186401,
      "learning_rate": 1.3768115942028985e-05,
      "loss": 0.3424,
      "step": 2000
    },
    {
      "epoch": 2.1739130434782608,
      "eval_loss": 0.5298786163330078,
      "eval_runtime": 10.8329,
      "eval_samples_per_second": 44.31,
      "eval_steps_per_second": 2.769,
      "step": 2000
    },
    {
      "epoch": 2.1847826086956523,
      "grad_norm": 1.2170861959457397,
      "learning_rate": 1.3586956521739131e-05,
      "loss": 0.3024,
      "step": 2010
    },
    {
      "epoch": 2.1956521739130435,
      "grad_norm": 1.1393592357635498,
      "learning_rate": 1.3405797101449276e-05,
      "loss": 0.3211,
      "step": 2020
    },
    {
      "epoch": 2.2065217391304346,
      "grad_norm": 1.1940656900405884,
      "learning_rate": 1.3224637681159422e-05,
      "loss": 0.3157,
      "step": 2030
    },
    {
      "epoch": 2.217391304347826,
      "grad_norm": 1.2777471542358398,
      "learning_rate": 1.3043478260869566e-05,
      "loss": 0.3382,
      "step": 2040
    },
    {
      "epoch": 2.2282608695652173,
      "grad_norm": 1.1014587879180908,
      "learning_rate": 1.286231884057971e-05,
      "loss": 0.3198,
      "step": 2050
    },
    {
      "epoch": 2.2282608695652173,
      "eval_loss": 0.5298696160316467,
      "eval_runtime": 10.844,
      "eval_samples_per_second": 44.264,
      "eval_steps_per_second": 2.767,
      "step": 2050
    },
    {
      "epoch": 2.239130434782609,
      "grad_norm": 1.056643009185791,
      "learning_rate": 1.2681159420289857e-05,
      "loss": 0.317,
      "step": 2060
    },
    {
      "epoch": 2.25,
      "grad_norm": 1.1301910877227783,
      "learning_rate": 1.25e-05,
      "loss": 0.3266,
      "step": 2070
    },
    {
      "epoch": 2.260869565217391,
      "grad_norm": 1.2433290481567383,
      "learning_rate": 1.2318840579710146e-05,
      "loss": 0.3313,
      "step": 2080
    },
    {
      "epoch": 2.2717391304347827,
      "grad_norm": 1.2331655025482178,
      "learning_rate": 1.213768115942029e-05,
      "loss": 0.3353,
      "step": 2090
    },
    {
      "epoch": 2.282608695652174,
      "grad_norm": 1.2137727737426758,
      "learning_rate": 1.1956521739130435e-05,
      "loss": 0.3183,
      "step": 2100
    },
    {
      "epoch": 2.282608695652174,
      "eval_loss": 0.5301353335380554,
      "eval_runtime": 10.829,
      "eval_samples_per_second": 44.326,
      "eval_steps_per_second": 2.77,
      "step": 2100
    },
    {
      "epoch": 2.2934782608695654,
      "grad_norm": 1.0960383415222168,
      "learning_rate": 1.177536231884058e-05,
      "loss": 0.3083,
      "step": 2110
    },
    {
      "epoch": 2.3043478260869565,
      "grad_norm": 1.2978132963180542,
      "learning_rate": 1.1594202898550725e-05,
      "loss": 0.3218,
      "step": 2120
    },
    {
      "epoch": 2.3152173913043477,
      "grad_norm": 1.1214746236801147,
      "learning_rate": 1.141304347826087e-05,
      "loss": 0.3334,
      "step": 2130
    },
    {
      "epoch": 2.3260869565217392,
      "grad_norm": 1.2013393640518188,
      "learning_rate": 1.1231884057971016e-05,
      "loss": 0.3017,
      "step": 2140
    },
    {
      "epoch": 2.3369565217391304,
      "grad_norm": 1.2782713174819946,
      "learning_rate": 1.1050724637681161e-05,
      "loss": 0.3272,
      "step": 2150
    },
    {
      "epoch": 2.3369565217391304,
      "eval_loss": 0.5313724279403687,
      "eval_runtime": 10.8265,
      "eval_samples_per_second": 44.336,
      "eval_steps_per_second": 2.771,
      "step": 2150
    },
    {
      "epoch": 2.3478260869565215,
      "grad_norm": 1.1971640586853027,
      "learning_rate": 1.0869565217391305e-05,
      "loss": 0.3295,
      "step": 2160
    },
    {
      "epoch": 2.358695652173913,
      "grad_norm": 1.1936053037643433,
      "learning_rate": 1.068840579710145e-05,
      "loss": 0.3132,
      "step": 2170
    },
    {
      "epoch": 2.369565217391304,
      "grad_norm": 1.0900537967681885,
      "learning_rate": 1.0507246376811594e-05,
      "loss": 0.3212,
      "step": 2180
    },
    {
      "epoch": 2.380434782608696,
      "grad_norm": 1.4006975889205933,
      "learning_rate": 1.032608695652174e-05,
      "loss": 0.3258,
      "step": 2190
    },
    {
      "epoch": 2.391304347826087,
      "grad_norm": 1.3683350086212158,
      "learning_rate": 1.0144927536231885e-05,
      "loss": 0.3376,
      "step": 2200
    },
    {
      "epoch": 2.391304347826087,
      "eval_loss": 0.5328701734542847,
      "eval_runtime": 10.8097,
      "eval_samples_per_second": 44.404,
      "eval_steps_per_second": 2.775,
      "step": 2200
    },
    {
      "epoch": 2.4021739130434785,
      "grad_norm": 1.120539903640747,
      "learning_rate": 9.96376811594203e-06,
      "loss": 0.3132,
      "step": 2210
    },
    {
      "epoch": 2.4130434782608696,
      "grad_norm": 1.1731290817260742,
      "learning_rate": 9.782608695652175e-06,
      "loss": 0.329,
      "step": 2220
    },
    {
      "epoch": 2.4239130434782608,
      "grad_norm": 1.2622851133346558,
      "learning_rate": 9.601449275362319e-06,
      "loss": 0.3181,
      "step": 2230
    },
    {
      "epoch": 2.4347826086956523,
      "grad_norm": 1.2982534170150757,
      "learning_rate": 9.420289855072464e-06,
      "loss": 0.3177,
      "step": 2240
    },
    {
      "epoch": 2.4456521739130435,
      "grad_norm": 1.2450315952301025,
      "learning_rate": 9.239130434782608e-06,
      "loss": 0.3361,
      "step": 2250
    },
    {
      "epoch": 2.4456521739130435,
      "eval_loss": 0.5312691330909729,
      "eval_runtime": 10.7859,
      "eval_samples_per_second": 44.502,
      "eval_steps_per_second": 2.781,
      "step": 2250
    },
    {
      "epoch": 2.4565217391304346,
      "grad_norm": 1.2270658016204834,
      "learning_rate": 9.057971014492753e-06,
      "loss": 0.3392,
      "step": 2260
    },
    {
      "epoch": 2.467391304347826,
      "grad_norm": 1.313056468963623,
      "learning_rate": 8.876811594202899e-06,
      "loss": 0.3065,
      "step": 2270
    },
    {
      "epoch": 2.4782608695652173,
      "grad_norm": 1.255582332611084,
      "learning_rate": 8.695652173913044e-06,
      "loss": 0.3224,
      "step": 2280
    },
    {
      "epoch": 2.489130434782609,
      "grad_norm": 1.354356050491333,
      "learning_rate": 8.51449275362319e-06,
      "loss": 0.3307,
      "step": 2290
    },
    {
      "epoch": 2.5,
      "grad_norm": 1.2218225002288818,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.3146,
      "step": 2300
    },
    {
      "epoch": 2.5,
      "eval_loss": 0.5301942825317383,
      "eval_runtime": 10.778,
      "eval_samples_per_second": 44.535,
      "eval_steps_per_second": 2.783,
      "step": 2300
    },
    {
      "epoch": 2.5108695652173916,
      "grad_norm": 1.2889074087142944,
      "learning_rate": 8.15217391304348e-06,
      "loss": 0.3135,
      "step": 2310
    },
    {
      "epoch": 2.5217391304347827,
      "grad_norm": 1.1224515438079834,
      "learning_rate": 7.971014492753623e-06,
      "loss": 0.3227,
      "step": 2320
    },
    {
      "epoch": 2.532608695652174,
      "grad_norm": 1.1914352178573608,
      "learning_rate": 7.789855072463769e-06,
      "loss": 0.3138,
      "step": 2330
    },
    {
      "epoch": 2.5434782608695654,
      "grad_norm": 1.3238486051559448,
      "learning_rate": 7.608695652173914e-06,
      "loss": 0.3168,
      "step": 2340
    },
    {
      "epoch": 2.5543478260869565,
      "grad_norm": 1.3031419515609741,
      "learning_rate": 7.427536231884058e-06,
      "loss": 0.3253,
      "step": 2350
    },
    {
      "epoch": 2.5543478260869565,
      "eval_loss": 0.5280157327651978,
      "eval_runtime": 10.8003,
      "eval_samples_per_second": 44.443,
      "eval_steps_per_second": 2.778,
      "step": 2350
    },
    {
      "epoch": 2.5652173913043477,
      "grad_norm": 1.2570607662200928,
      "learning_rate": 7.246376811594203e-06,
      "loss": 0.3246,
      "step": 2360
    },
    {
      "epoch": 2.5760869565217392,
      "grad_norm": 1.1971008777618408,
      "learning_rate": 7.065217391304347e-06,
      "loss": 0.3142,
      "step": 2370
    },
    {
      "epoch": 2.5869565217391304,
      "grad_norm": 1.2740592956542969,
      "learning_rate": 6.884057971014493e-06,
      "loss": 0.3263,
      "step": 2380
    },
    {
      "epoch": 2.5978260869565215,
      "grad_norm": 1.3141279220581055,
      "learning_rate": 6.702898550724638e-06,
      "loss": 0.3141,
      "step": 2390
    },
    {
      "epoch": 2.608695652173913,
      "grad_norm": 1.3131691217422485,
      "learning_rate": 6.521739130434783e-06,
      "loss": 0.3204,
      "step": 2400
    },
    {
      "epoch": 2.608695652173913,
      "eval_loss": 0.529800295829773,
      "eval_runtime": 10.7985,
      "eval_samples_per_second": 44.451,
      "eval_steps_per_second": 2.778,
      "step": 2400
    },
    {
      "epoch": 2.619565217391304,
      "grad_norm": 1.2273128032684326,
      "learning_rate": 6.340579710144928e-06,
      "loss": 0.3089,
      "step": 2410
    },
    {
      "epoch": 2.630434782608696,
      "grad_norm": 1.304915428161621,
      "learning_rate": 6.159420289855073e-06,
      "loss": 0.3211,
      "step": 2420
    },
    {
      "epoch": 2.641304347826087,
      "grad_norm": 1.261481761932373,
      "learning_rate": 5.978260869565218e-06,
      "loss": 0.3178,
      "step": 2430
    },
    {
      "epoch": 2.6521739130434785,
      "grad_norm": 1.2726097106933594,
      "learning_rate": 5.797101449275362e-06,
      "loss": 0.3342,
      "step": 2440
    },
    {
      "epoch": 2.6630434782608696,
      "grad_norm": 1.3216400146484375,
      "learning_rate": 5.615942028985508e-06,
      "loss": 0.3245,
      "step": 2450
    },
    {
      "epoch": 2.6630434782608696,
      "eval_loss": 0.5288810133934021,
      "eval_runtime": 10.8043,
      "eval_samples_per_second": 44.427,
      "eval_steps_per_second": 2.777,
      "step": 2450
    },
    {
      "epoch": 2.6739130434782608,
      "grad_norm": 1.2873594760894775,
      "learning_rate": 5.4347826086956525e-06,
      "loss": 0.3137,
      "step": 2460
    },
    {
      "epoch": 2.6847826086956523,
      "grad_norm": 1.2318347692489624,
      "learning_rate": 5.253623188405797e-06,
      "loss": 0.3258,
      "step": 2470
    },
    {
      "epoch": 2.6956521739130435,
      "grad_norm": 1.2132396697998047,
      "learning_rate": 5.072463768115943e-06,
      "loss": 0.3168,
      "step": 2480
    },
    {
      "epoch": 2.7065217391304346,
      "grad_norm": 1.3572018146514893,
      "learning_rate": 4.891304347826087e-06,
      "loss": 0.3187,
      "step": 2490
    },
    {
      "epoch": 2.717391304347826,
      "grad_norm": 1.2623666524887085,
      "learning_rate": 4.710144927536232e-06,
      "loss": 0.3257,
      "step": 2500
    },
    {
      "epoch": 2.717391304347826,
      "eval_loss": 0.5298188924789429,
      "eval_runtime": 10.8163,
      "eval_samples_per_second": 44.377,
      "eval_steps_per_second": 2.774,
      "step": 2500
    }
  ],
  "logging_steps": 10,
  "max_steps": 2760,
  "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": 2.2027066610132582e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}