{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.5384615384615383,
  "eval_steps": 10,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003076923076923077,
      "eval_loss": 1.4537198543548584,
      "eval_runtime": 2.2676,
      "eval_samples_per_second": 50.275,
      "eval_steps_per_second": 2.646,
      "step": 2
    },
    {
      "epoch": 0.015384615384615385,
      "grad_norm": 11.972827911376953,
      "learning_rate": 4.395604395604396e-07,
      "loss": 1.3965,
      "step": 10
    },
    {
      "epoch": 0.015384615384615385,
      "eval_loss": 1.444534182548523,
      "eval_runtime": 2.0489,
      "eval_samples_per_second": 55.638,
      "eval_steps_per_second": 2.928,
      "step": 10
    },
    {
      "epoch": 0.03076923076923077,
      "grad_norm": 14.562355995178223,
      "learning_rate": 8.791208791208792e-07,
      "loss": 1.3983,
      "step": 20
    },
    {
      "epoch": 0.03076923076923077,
      "eval_loss": 1.3285874128341675,
      "eval_runtime": 2.033,
      "eval_samples_per_second": 56.075,
      "eval_steps_per_second": 2.951,
      "step": 20
    },
    {
      "epoch": 0.046153846153846156,
      "grad_norm": 13.817586898803711,
      "learning_rate": 1.3186813186813187e-06,
      "loss": 1.2094,
      "step": 30
    },
    {
      "epoch": 0.046153846153846156,
      "eval_loss": 0.9445623159408569,
      "eval_runtime": 2.0359,
      "eval_samples_per_second": 55.994,
      "eval_steps_per_second": 2.947,
      "step": 30
    },
    {
      "epoch": 0.06153846153846154,
      "grad_norm": 7.5655837059021,
      "learning_rate": 1.7582417582417585e-06,
      "loss": 0.7584,
      "step": 40
    },
    {
      "epoch": 0.06153846153846154,
      "eval_loss": 0.8109635710716248,
      "eval_runtime": 2.0521,
      "eval_samples_per_second": 55.553,
      "eval_steps_per_second": 2.924,
      "step": 40
    },
    {
      "epoch": 0.07692307692307693,
      "grad_norm": 12.146793365478516,
      "learning_rate": 2.197802197802198e-06,
      "loss": 0.7401,
      "step": 50
    },
    {
      "epoch": 0.07692307692307693,
      "eval_loss": 0.7573030591011047,
      "eval_runtime": 2.0481,
      "eval_samples_per_second": 55.661,
      "eval_steps_per_second": 2.93,
      "step": 50
    },
    {
      "epoch": 0.09230769230769231,
      "grad_norm": 9.491755485534668,
      "learning_rate": 2.6373626373626375e-06,
      "loss": 0.6296,
      "step": 60
    },
    {
      "epoch": 0.09230769230769231,
      "eval_loss": 0.7523270845413208,
      "eval_runtime": 2.0501,
      "eval_samples_per_second": 55.606,
      "eval_steps_per_second": 2.927,
      "step": 60
    },
    {
      "epoch": 0.1076923076923077,
      "grad_norm": 5.857011795043945,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.7523,
      "step": 70
    },
    {
      "epoch": 0.1076923076923077,
      "eval_loss": 0.7429385185241699,
      "eval_runtime": 2.051,
      "eval_samples_per_second": 55.583,
      "eval_steps_per_second": 2.925,
      "step": 70
    },
    {
      "epoch": 0.12307692307692308,
      "grad_norm": 4.9665045738220215,
      "learning_rate": 3.516483516483517e-06,
      "loss": 0.6073,
      "step": 80
    },
    {
      "epoch": 0.12307692307692308,
      "eval_loss": 0.7138068079948425,
      "eval_runtime": 2.0478,
      "eval_samples_per_second": 55.67,
      "eval_steps_per_second": 2.93,
      "step": 80
    },
    {
      "epoch": 0.13846153846153847,
      "grad_norm": 11.936062812805176,
      "learning_rate": 3.9560439560439565e-06,
      "loss": 0.6577,
      "step": 90
    },
    {
      "epoch": 0.13846153846153847,
      "eval_loss": 0.6989916563034058,
      "eval_runtime": 2.0579,
      "eval_samples_per_second": 55.397,
      "eval_steps_per_second": 2.916,
      "step": 90
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 13.15188980102539,
      "learning_rate": 4.395604395604396e-06,
      "loss": 0.698,
      "step": 100
    },
    {
      "epoch": 0.15384615384615385,
      "eval_loss": 0.7036928534507751,
      "eval_runtime": 2.0376,
      "eval_samples_per_second": 55.949,
      "eval_steps_per_second": 2.945,
      "step": 100
    },
    {
      "epoch": 0.16923076923076924,
      "grad_norm": 10.141645431518555,
      "learning_rate": 4.8351648351648355e-06,
      "loss": 0.6925,
      "step": 110
    },
    {
      "epoch": 0.16923076923076924,
      "eval_loss": 0.6939440369606018,
      "eval_runtime": 2.0357,
      "eval_samples_per_second": 56.002,
      "eval_steps_per_second": 2.947,
      "step": 110
    },
    {
      "epoch": 0.18461538461538463,
      "grad_norm": 3.710402488708496,
      "learning_rate": 5.274725274725275e-06,
      "loss": 0.6055,
      "step": 120
    },
    {
      "epoch": 0.18461538461538463,
      "eval_loss": 0.7082720994949341,
      "eval_runtime": 2.0484,
      "eval_samples_per_second": 55.652,
      "eval_steps_per_second": 2.929,
      "step": 120
    },
    {
      "epoch": 0.2,
      "grad_norm": 4.347709655761719,
      "learning_rate": 5.7142857142857145e-06,
      "loss": 0.5347,
      "step": 130
    },
    {
      "epoch": 0.2,
      "eval_loss": 0.6877034306526184,
      "eval_runtime": 2.0344,
      "eval_samples_per_second": 56.037,
      "eval_steps_per_second": 2.949,
      "step": 130
    },
    {
      "epoch": 0.2153846153846154,
      "grad_norm": 8.17250919342041,
      "learning_rate": 6.153846153846155e-06,
      "loss": 0.6877,
      "step": 140
    },
    {
      "epoch": 0.2153846153846154,
      "eval_loss": 0.6877639889717102,
      "eval_runtime": 2.0484,
      "eval_samples_per_second": 55.652,
      "eval_steps_per_second": 2.929,
      "step": 140
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 6.628317356109619,
      "learning_rate": 6.5934065934065935e-06,
      "loss": 0.5965,
      "step": 150
    },
    {
      "epoch": 0.23076923076923078,
      "eval_loss": 0.6875787973403931,
      "eval_runtime": 2.0524,
      "eval_samples_per_second": 55.545,
      "eval_steps_per_second": 2.923,
      "step": 150
    },
    {
      "epoch": 0.24615384615384617,
      "grad_norm": 4.867484092712402,
      "learning_rate": 7.032967032967034e-06,
      "loss": 0.703,
      "step": 160
    },
    {
      "epoch": 0.24615384615384617,
      "eval_loss": 0.6956614851951599,
      "eval_runtime": 2.0284,
      "eval_samples_per_second": 56.201,
      "eval_steps_per_second": 2.958,
      "step": 160
    },
    {
      "epoch": 0.26153846153846155,
      "grad_norm": 3.452112913131714,
      "learning_rate": 7.472527472527473e-06,
      "loss": 0.6539,
      "step": 170
    },
    {
      "epoch": 0.26153846153846155,
      "eval_loss": 0.6873570084571838,
      "eval_runtime": 2.0311,
      "eval_samples_per_second": 56.127,
      "eval_steps_per_second": 2.954,
      "step": 170
    },
    {
      "epoch": 0.27692307692307694,
      "grad_norm": 5.422184944152832,
      "learning_rate": 7.912087912087913e-06,
      "loss": 0.6788,
      "step": 180
    },
    {
      "epoch": 0.27692307692307694,
      "eval_loss": 0.6954818367958069,
      "eval_runtime": 2.051,
      "eval_samples_per_second": 55.583,
      "eval_steps_per_second": 2.925,
      "step": 180
    },
    {
      "epoch": 0.2923076923076923,
      "grad_norm": 4.284252643585205,
      "learning_rate": 8.351648351648353e-06,
      "loss": 0.659,
      "step": 190
    },
    {
      "epoch": 0.2923076923076923,
      "eval_loss": 0.6905339360237122,
      "eval_runtime": 2.0478,
      "eval_samples_per_second": 55.67,
      "eval_steps_per_second": 2.93,
      "step": 190
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 4.645473003387451,
      "learning_rate": 8.791208791208792e-06,
      "loss": 0.6795,
      "step": 200
    },
    {
      "epoch": 0.3076923076923077,
      "eval_loss": 0.6910640001296997,
      "eval_runtime": 2.0348,
      "eval_samples_per_second": 56.026,
      "eval_steps_per_second": 2.949,
      "step": 200
    },
    {
      "epoch": 0.3230769230769231,
      "grad_norm": 3.9407358169555664,
      "learning_rate": 9.230769230769232e-06,
      "loss": 0.6608,
      "step": 210
    },
    {
      "epoch": 0.3230769230769231,
      "eval_loss": 0.6903011798858643,
      "eval_runtime": 2.0478,
      "eval_samples_per_second": 55.669,
      "eval_steps_per_second": 2.93,
      "step": 210
    },
    {
      "epoch": 0.3384615384615385,
      "grad_norm": 3.000676393508911,
      "learning_rate": 9.670329670329671e-06,
      "loss": 0.647,
      "step": 220
    },
    {
      "epoch": 0.3384615384615385,
      "eval_loss": 0.6933804154396057,
      "eval_runtime": 2.0377,
      "eval_samples_per_second": 55.946,
      "eval_steps_per_second": 2.945,
      "step": 220
    },
    {
      "epoch": 0.35384615384615387,
      "grad_norm": 5.05721378326416,
      "learning_rate": 1.010989010989011e-05,
      "loss": 0.669,
      "step": 230
    },
    {
      "epoch": 0.35384615384615387,
      "eval_loss": 0.6953541040420532,
      "eval_runtime": 2.0502,
      "eval_samples_per_second": 55.604,
      "eval_steps_per_second": 2.927,
      "step": 230
    },
    {
      "epoch": 0.36923076923076925,
      "grad_norm": 2.5314972400665283,
      "learning_rate": 1.054945054945055e-05,
      "loss": 0.7088,
      "step": 240
    },
    {
      "epoch": 0.36923076923076925,
      "eval_loss": 0.6936004757881165,
      "eval_runtime": 2.0376,
      "eval_samples_per_second": 55.948,
      "eval_steps_per_second": 2.945,
      "step": 240
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 9.507554054260254,
      "learning_rate": 1.098901098901099e-05,
      "loss": 0.7007,
      "step": 250
    },
    {
      "epoch": 0.38461538461538464,
      "eval_loss": 0.7747458815574646,
      "eval_runtime": 2.042,
      "eval_samples_per_second": 55.827,
      "eval_steps_per_second": 2.938,
      "step": 250
    },
    {
      "epoch": 0.4,
      "grad_norm": 3.4734535217285156,
      "learning_rate": 1.1428571428571429e-05,
      "loss": 0.6827,
      "step": 260
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.6972140669822693,
      "eval_runtime": 2.0689,
      "eval_samples_per_second": 55.101,
      "eval_steps_per_second": 2.9,
      "step": 260
    },
    {
      "epoch": 0.4153846153846154,
      "grad_norm": 6.09801721572876,
      "learning_rate": 1.186813186813187e-05,
      "loss": 0.6556,
      "step": 270
    },
    {
      "epoch": 0.4153846153846154,
      "eval_loss": 0.7027987837791443,
      "eval_runtime": 2.0282,
      "eval_samples_per_second": 56.208,
      "eval_steps_per_second": 2.958,
      "step": 270
    },
    {
      "epoch": 0.4307692307692308,
      "grad_norm": 4.644292831420898,
      "learning_rate": 1.230769230769231e-05,
      "loss": 0.6851,
      "step": 280
    },
    {
      "epoch": 0.4307692307692308,
      "eval_loss": 0.7334415316581726,
      "eval_runtime": 2.0372,
      "eval_samples_per_second": 55.958,
      "eval_steps_per_second": 2.945,
      "step": 280
    },
    {
      "epoch": 0.4461538461538462,
      "grad_norm": 3.16902232170105,
      "learning_rate": 1.2747252747252747e-05,
      "loss": 0.7118,
      "step": 290
    },
    {
      "epoch": 0.4461538461538462,
      "eval_loss": 0.7018752694129944,
      "eval_runtime": 2.0467,
      "eval_samples_per_second": 55.699,
      "eval_steps_per_second": 2.932,
      "step": 290
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 3.3035457134246826,
      "learning_rate": 1.3186813186813187e-05,
      "loss": 0.6381,
      "step": 300
    },
    {
      "epoch": 0.46153846153846156,
      "eval_loss": 0.7047535181045532,
      "eval_runtime": 2.0352,
      "eval_samples_per_second": 56.015,
      "eval_steps_per_second": 2.948,
      "step": 300
    },
    {
      "epoch": 0.47692307692307695,
      "grad_norm": 2.248196840286255,
      "learning_rate": 1.3626373626373627e-05,
      "loss": 0.6167,
      "step": 310
    },
    {
      "epoch": 0.47692307692307695,
      "eval_loss": 0.7115533351898193,
      "eval_runtime": 2.0425,
      "eval_samples_per_second": 55.815,
      "eval_steps_per_second": 2.938,
      "step": 310
    },
    {
      "epoch": 0.49230769230769234,
      "grad_norm": 6.9102678298950195,
      "learning_rate": 1.4065934065934068e-05,
      "loss": 0.7308,
      "step": 320
    },
    {
      "epoch": 0.49230769230769234,
      "eval_loss": 0.7213383913040161,
      "eval_runtime": 2.0226,
      "eval_samples_per_second": 56.362,
      "eval_steps_per_second": 2.966,
      "step": 320
    },
    {
      "epoch": 0.5076923076923077,
      "grad_norm": 3.5638887882232666,
      "learning_rate": 1.4505494505494506e-05,
      "loss": 0.6035,
      "step": 330
    },
    {
      "epoch": 0.5076923076923077,
      "eval_loss": 0.7181587815284729,
      "eval_runtime": 2.06,
      "eval_samples_per_second": 55.34,
      "eval_steps_per_second": 2.913,
      "step": 330
    },
    {
      "epoch": 0.5230769230769231,
      "grad_norm": 4.234628677368164,
      "learning_rate": 1.4945054945054947e-05,
      "loss": 0.5433,
      "step": 340
    },
    {
      "epoch": 0.5230769230769231,
      "eval_loss": 0.7234803438186646,
      "eval_runtime": 2.0282,
      "eval_samples_per_second": 56.207,
      "eval_steps_per_second": 2.958,
      "step": 340
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 2.9397354125976562,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.6332,
      "step": 350
    },
    {
      "epoch": 0.5384615384615384,
      "eval_loss": 0.7224182486534119,
      "eval_runtime": 2.1213,
      "eval_samples_per_second": 53.74,
      "eval_steps_per_second": 2.828,
      "step": 350
    },
    {
      "epoch": 0.5538461538461539,
      "grad_norm": 3.2468433380126953,
      "learning_rate": 1.5824175824175826e-05,
      "loss": 0.7297,
      "step": 360
    },
    {
      "epoch": 0.5538461538461539,
      "eval_loss": 0.7258623838424683,
      "eval_runtime": 2.0648,
      "eval_samples_per_second": 55.211,
      "eval_steps_per_second": 2.906,
      "step": 360
    },
    {
      "epoch": 0.5692307692307692,
      "grad_norm": 4.74401330947876,
      "learning_rate": 1.6263736263736265e-05,
      "loss": 0.6346,
      "step": 370
    },
    {
      "epoch": 0.5692307692307692,
      "eval_loss": 0.7451629638671875,
      "eval_runtime": 2.143,
      "eval_samples_per_second": 53.196,
      "eval_steps_per_second": 2.8,
      "step": 370
    },
    {
      "epoch": 0.5846153846153846,
      "grad_norm": 3.286860942840576,
      "learning_rate": 1.6703296703296707e-05,
      "loss": 0.6819,
      "step": 380
    },
    {
      "epoch": 0.5846153846153846,
      "eval_loss": 0.7284204959869385,
      "eval_runtime": 2.0361,
      "eval_samples_per_second": 55.989,
      "eval_steps_per_second": 2.947,
      "step": 380
    },
    {
      "epoch": 0.6,
      "grad_norm": 4.673974514007568,
      "learning_rate": 1.7142857142857142e-05,
      "loss": 0.6864,
      "step": 390
    },
    {
      "epoch": 0.6,
      "eval_loss": 0.7300311923027039,
      "eval_runtime": 2.1078,
      "eval_samples_per_second": 54.086,
      "eval_steps_per_second": 2.847,
      "step": 390
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 6.604970932006836,
      "learning_rate": 1.7582417582417584e-05,
      "loss": 0.6694,
      "step": 400
    },
    {
      "epoch": 0.6153846153846154,
      "eval_loss": 0.7328219413757324,
      "eval_runtime": 2.1266,
      "eval_samples_per_second": 53.608,
      "eval_steps_per_second": 2.821,
      "step": 400
    },
    {
      "epoch": 0.6307692307692307,
      "grad_norm": 2.9900333881378174,
      "learning_rate": 1.8021978021978023e-05,
      "loss": 0.513,
      "step": 410
    },
    {
      "epoch": 0.6307692307692307,
      "eval_loss": 0.7355452179908752,
      "eval_runtime": 2.0335,
      "eval_samples_per_second": 56.062,
      "eval_steps_per_second": 2.951,
      "step": 410
    },
    {
      "epoch": 0.6461538461538462,
      "grad_norm": 4.437044143676758,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.599,
      "step": 420
    },
    {
      "epoch": 0.6461538461538462,
      "eval_loss": 0.7413352727890015,
      "eval_runtime": 2.0588,
      "eval_samples_per_second": 55.371,
      "eval_steps_per_second": 2.914,
      "step": 420
    },
    {
      "epoch": 0.6615384615384615,
      "grad_norm": 2.5650463104248047,
      "learning_rate": 1.8901098901098903e-05,
      "loss": 0.765,
      "step": 430
    },
    {
      "epoch": 0.6615384615384615,
      "eval_loss": 0.735922634601593,
      "eval_runtime": 2.0718,
      "eval_samples_per_second": 55.025,
      "eval_steps_per_second": 2.896,
      "step": 430
    },
    {
      "epoch": 0.676923076923077,
      "grad_norm": 6.075387477874756,
      "learning_rate": 1.9340659340659342e-05,
      "loss": 0.6436,
      "step": 440
    },
    {
      "epoch": 0.676923076923077,
      "eval_loss": 0.7397021651268005,
      "eval_runtime": 2.0454,
      "eval_samples_per_second": 55.734,
      "eval_steps_per_second": 2.933,
      "step": 440
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 4.346602916717529,
      "learning_rate": 1.9780219780219784e-05,
      "loss": 0.7325,
      "step": 450
    },
    {
      "epoch": 0.6923076923076923,
      "eval_loss": 0.7424116134643555,
      "eval_runtime": 2.1114,
      "eval_samples_per_second": 53.993,
      "eval_steps_per_second": 2.842,
      "step": 450
    },
    {
      "epoch": 0.7076923076923077,
      "grad_norm": 2.532031774520874,
      "learning_rate": 1.9999926429888597e-05,
      "loss": 0.675,
      "step": 460
    },
    {
      "epoch": 0.7076923076923077,
      "eval_loss": 0.7457496523857117,
      "eval_runtime": 2.069,
      "eval_samples_per_second": 55.099,
      "eval_steps_per_second": 2.9,
      "step": 460
    },
    {
      "epoch": 0.7230769230769231,
      "grad_norm": 3.643791675567627,
      "learning_rate": 1.9999337875492412e-05,
      "loss": 0.7264,
      "step": 470
    },
    {
      "epoch": 0.7230769230769231,
      "eval_loss": 0.7425748705863953,
      "eval_runtime": 2.0709,
      "eval_samples_per_second": 55.05,
      "eval_steps_per_second": 2.897,
      "step": 470
    },
    {
      "epoch": 0.7384615384615385,
      "grad_norm": 1.3282146453857422,
      "learning_rate": 1.999816080133992e-05,
      "loss": 0.5571,
      "step": 480
    },
    {
      "epoch": 0.7384615384615385,
      "eval_loss": 0.7501851916313171,
      "eval_runtime": 2.107,
      "eval_samples_per_second": 54.106,
      "eval_steps_per_second": 2.848,
      "step": 480
    },
    {
      "epoch": 0.7538461538461538,
      "grad_norm": 2.987581968307495,
      "learning_rate": 1.9996395276708856e-05,
      "loss": 0.7058,
      "step": 490
    },
    {
      "epoch": 0.7538461538461538,
      "eval_loss": 0.7441815733909607,
      "eval_runtime": 2.0491,
      "eval_samples_per_second": 55.634,
      "eval_steps_per_second": 2.928,
      "step": 490
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 2.506946563720703,
      "learning_rate": 1.9994041405510705e-05,
      "loss": 0.6256,
      "step": 500
    },
    {
      "epoch": 0.7692307692307693,
      "eval_loss": 0.7526156902313232,
      "eval_runtime": 2.0407,
      "eval_samples_per_second": 55.864,
      "eval_steps_per_second": 2.94,
      "step": 500
    },
    {
      "epoch": 0.7846153846153846,
      "grad_norm": 2.2201008796691895,
      "learning_rate": 1.9991099326284616e-05,
      "loss": 0.6102,
      "step": 510
    },
    {
      "epoch": 0.7846153846153846,
      "eval_loss": 0.7576336860656738,
      "eval_runtime": 2.1027,
      "eval_samples_per_second": 54.217,
      "eval_steps_per_second": 2.854,
      "step": 510
    },
    {
      "epoch": 0.8,
      "grad_norm": 2.3949899673461914,
      "learning_rate": 1.9987569212189224e-05,
      "loss": 0.7756,
      "step": 520
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.7575399279594421,
      "eval_runtime": 2.0449,
      "eval_samples_per_second": 55.749,
      "eval_steps_per_second": 2.934,
      "step": 520
    },
    {
      "epoch": 0.8153846153846154,
      "grad_norm": 2.082437753677368,
      "learning_rate": 1.998345127099248e-05,
      "loss": 0.7127,
      "step": 530
    },
    {
      "epoch": 0.8153846153846154,
      "eval_loss": 0.7561782598495483,
      "eval_runtime": 2.048,
      "eval_samples_per_second": 55.664,
      "eval_steps_per_second": 2.93,
      "step": 530
    },
    {
      "epoch": 0.8307692307692308,
      "grad_norm": 2.132066488265991,
      "learning_rate": 1.99787457450594e-05,
      "loss": 0.6398,
      "step": 540
    },
    {
      "epoch": 0.8307692307692308,
      "eval_loss": 0.7557066679000854,
      "eval_runtime": 2.0659,
      "eval_samples_per_second": 55.182,
      "eval_steps_per_second": 2.904,
      "step": 540
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 2.816959857940674,
      "learning_rate": 1.997345291133783e-05,
      "loss": 0.6956,
      "step": 550
    },
    {
      "epoch": 0.8461538461538461,
      "eval_loss": 0.7591909170150757,
      "eval_runtime": 2.0335,
      "eval_samples_per_second": 56.061,
      "eval_steps_per_second": 2.951,
      "step": 550
    },
    {
      "epoch": 0.8615384615384616,
      "grad_norm": 1.8596259355545044,
      "learning_rate": 1.9967573081342103e-05,
      "loss": 0.656,
      "step": 560
    },
    {
      "epoch": 0.8615384615384616,
      "eval_loss": 0.76849365234375,
      "eval_runtime": 2.0556,
      "eval_samples_per_second": 55.457,
      "eval_steps_per_second": 2.919,
      "step": 560
    },
    {
      "epoch": 0.8769230769230769,
      "grad_norm": 2.643859386444092,
      "learning_rate": 1.996110660113475e-05,
      "loss": 0.6426,
      "step": 570
    },
    {
      "epoch": 0.8769230769230769,
      "eval_loss": 0.7682607769966125,
      "eval_runtime": 2.0831,
      "eval_samples_per_second": 54.725,
      "eval_steps_per_second": 2.88,
      "step": 570
    },
    {
      "epoch": 0.8923076923076924,
      "grad_norm": 3.8031773567199707,
      "learning_rate": 1.995405385130611e-05,
      "loss": 0.7259,
      "step": 580
    },
    {
      "epoch": 0.8923076923076924,
      "eval_loss": 0.7607874274253845,
      "eval_runtime": 2.0943,
      "eval_samples_per_second": 54.433,
      "eval_steps_per_second": 2.865,
      "step": 580
    },
    {
      "epoch": 0.9076923076923077,
      "grad_norm": 2.6620101928710938,
      "learning_rate": 1.9946415246951928e-05,
      "loss": 0.739,
      "step": 590
    },
    {
      "epoch": 0.9076923076923077,
      "eval_loss": 0.7598668336868286,
      "eval_runtime": 2.0418,
      "eval_samples_per_second": 55.833,
      "eval_steps_per_second": 2.939,
      "step": 590
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 4.222265720367432,
      "learning_rate": 1.9938191237648924e-05,
      "loss": 0.7451,
      "step": 600
    },
    {
      "epoch": 0.9230769230769231,
      "eval_loss": 0.7620775699615479,
      "eval_runtime": 2.0423,
      "eval_samples_per_second": 55.82,
      "eval_steps_per_second": 2.938,
      "step": 600
    },
    {
      "epoch": 0.9384615384615385,
      "grad_norm": 1.1073704957962036,
      "learning_rate": 1.992938230742835e-05,
      "loss": 0.705,
      "step": 610
    },
    {
      "epoch": 0.9384615384615385,
      "eval_loss": 0.7568953633308411,
      "eval_runtime": 2.1052,
      "eval_samples_per_second": 54.153,
      "eval_steps_per_second": 2.85,
      "step": 610
    },
    {
      "epoch": 0.9538461538461539,
      "grad_norm": 1.9715179204940796,
      "learning_rate": 1.9919988974747473e-05,
      "loss": 0.7293,
      "step": 620
    },
    {
      "epoch": 0.9538461538461539,
      "eval_loss": 0.7628914713859558,
      "eval_runtime": 2.0473,
      "eval_samples_per_second": 55.684,
      "eval_steps_per_second": 2.931,
      "step": 620
    },
    {
      "epoch": 0.9692307692307692,
      "grad_norm": 3.2774569988250732,
      "learning_rate": 1.9910011792459086e-05,
      "loss": 0.7437,
      "step": 630
    },
    {
      "epoch": 0.9692307692307692,
      "eval_loss": 0.7696098685264587,
      "eval_runtime": 2.0418,
      "eval_samples_per_second": 55.833,
      "eval_steps_per_second": 2.939,
      "step": 630
    },
    {
      "epoch": 0.9846153846153847,
      "grad_norm": 2.902707099914551,
      "learning_rate": 1.9899451347778962e-05,
      "loss": 0.7946,
      "step": 640
    },
    {
      "epoch": 0.9846153846153847,
      "eval_loss": 0.760672390460968,
      "eval_runtime": 2.054,
      "eval_samples_per_second": 55.501,
      "eval_steps_per_second": 2.921,
      "step": 640
    },
    {
      "epoch": 1.0,
      "grad_norm": 4.315918922424316,
      "learning_rate": 1.9888308262251286e-05,
      "loss": 0.7553,
      "step": 650
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.7670853734016418,
      "eval_runtime": 2.0439,
      "eval_samples_per_second": 55.775,
      "eval_steps_per_second": 2.936,
      "step": 650
    },
    {
      "epoch": 1.0153846153846153,
      "grad_norm": 0.9407129287719727,
      "learning_rate": 1.9876583191712083e-05,
      "loss": 0.4016,
      "step": 660
    },
    {
      "epoch": 1.0153846153846153,
      "eval_loss": 0.8053916096687317,
      "eval_runtime": 2.0973,
      "eval_samples_per_second": 54.355,
      "eval_steps_per_second": 2.861,
      "step": 660
    },
    {
      "epoch": 1.0307692307692307,
      "grad_norm": 2.96834135055542,
      "learning_rate": 1.9864276826250608e-05,
      "loss": 0.3824,
      "step": 670
    },
    {
      "epoch": 1.0307692307692307,
      "eval_loss": 0.7907722592353821,
      "eval_runtime": 2.0495,
      "eval_samples_per_second": 55.624,
      "eval_steps_per_second": 2.928,
      "step": 670
    },
    {
      "epoch": 1.0461538461538462,
      "grad_norm": 3.3183233737945557,
      "learning_rate": 1.9851389890168738e-05,
      "loss": 0.3728,
      "step": 680
    },
    {
      "epoch": 1.0461538461538462,
      "eval_loss": 0.785139799118042,
      "eval_runtime": 2.0474,
      "eval_samples_per_second": 55.682,
      "eval_steps_per_second": 2.931,
      "step": 680
    },
    {
      "epoch": 1.0615384615384615,
      "grad_norm": 2.4374091625213623,
      "learning_rate": 1.983792314193835e-05,
      "loss": 0.3597,
      "step": 690
    },
    {
      "epoch": 1.0615384615384615,
      "eval_loss": 0.7914989590644836,
      "eval_runtime": 2.0646,
      "eval_samples_per_second": 55.217,
      "eval_steps_per_second": 2.906,
      "step": 690
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 1.1855288743972778,
      "learning_rate": 1.9823877374156647e-05,
      "loss": 0.2332,
      "step": 700
    },
    {
      "epoch": 1.0769230769230769,
      "eval_loss": 0.8562365770339966,
      "eval_runtime": 2.1227,
      "eval_samples_per_second": 53.706,
      "eval_steps_per_second": 2.827,
      "step": 700
    },
    {
      "epoch": 1.0923076923076924,
      "grad_norm": 1.8067868947982788,
      "learning_rate": 1.9809253413499565e-05,
      "loss": 0.3784,
      "step": 710
    },
    {
      "epoch": 1.0923076923076924,
      "eval_loss": 0.7976874709129333,
      "eval_runtime": 2.0503,
      "eval_samples_per_second": 55.601,
      "eval_steps_per_second": 2.926,
      "step": 710
    },
    {
      "epoch": 1.1076923076923078,
      "grad_norm": 1.8051857948303223,
      "learning_rate": 1.979405212067306e-05,
      "loss": 0.4172,
      "step": 720
    },
    {
      "epoch": 1.1076923076923078,
      "eval_loss": 0.8003087639808655,
      "eval_runtime": 2.0576,
      "eval_samples_per_second": 55.403,
      "eval_steps_per_second": 2.916,
      "step": 720
    },
    {
      "epoch": 1.123076923076923,
      "grad_norm": 3.1192572116851807,
      "learning_rate": 1.9778274390362488e-05,
      "loss": 0.3841,
      "step": 730
    },
    {
      "epoch": 1.123076923076923,
      "eval_loss": 0.7872086763381958,
      "eval_runtime": 2.0523,
      "eval_samples_per_second": 55.546,
      "eval_steps_per_second": 2.923,
      "step": 730
    },
    {
      "epoch": 1.1384615384615384,
      "grad_norm": 1.4112781286239624,
      "learning_rate": 1.9761921151179937e-05,
      "loss": 0.2886,
      "step": 740
    },
    {
      "epoch": 1.1384615384615384,
      "eval_loss": 0.8142873048782349,
      "eval_runtime": 2.0242,
      "eval_samples_per_second": 56.319,
      "eval_steps_per_second": 2.964,
      "step": 740
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 2.2796831130981445,
      "learning_rate": 1.9744993365609563e-05,
      "loss": 0.4226,
      "step": 750
    },
    {
      "epoch": 1.1538461538461537,
      "eval_loss": 0.7786396741867065,
      "eval_runtime": 2.0271,
      "eval_samples_per_second": 56.237,
      "eval_steps_per_second": 2.96,
      "step": 750
    },
    {
      "epoch": 1.1692307692307693,
      "grad_norm": 2.9644672870635986,
      "learning_rate": 1.9727492029950965e-05,
      "loss": 0.3813,
      "step": 760
    },
    {
      "epoch": 1.1692307692307693,
      "eval_loss": 0.7970356941223145,
      "eval_runtime": 2.0329,
      "eval_samples_per_second": 56.077,
      "eval_steps_per_second": 2.951,
      "step": 760
    },
    {
      "epoch": 1.1846153846153846,
      "grad_norm": 3.2686288356781006,
      "learning_rate": 1.9709418174260523e-05,
      "loss": 0.3995,
      "step": 770
    },
    {
      "epoch": 1.1846153846153846,
      "eval_loss": 0.813422441482544,
      "eval_runtime": 2.1424,
      "eval_samples_per_second": 53.211,
      "eval_steps_per_second": 2.801,
      "step": 770
    },
    {
      "epoch": 1.2,
      "grad_norm": 7.485370635986328,
      "learning_rate": 1.969077286229078e-05,
      "loss": 0.399,
      "step": 780
    },
    {
      "epoch": 1.2,
      "eval_loss": 0.8145797252655029,
      "eval_runtime": 2.0591,
      "eval_samples_per_second": 55.365,
      "eval_steps_per_second": 2.914,
      "step": 780
    },
    {
      "epoch": 1.2153846153846155,
      "grad_norm": 6.251548767089844,
      "learning_rate": 1.967155719142785e-05,
      "loss": 0.4973,
      "step": 790
    },
    {
      "epoch": 1.2153846153846155,
      "eval_loss": 0.7992886900901794,
      "eval_runtime": 2.0414,
      "eval_samples_per_second": 55.844,
      "eval_steps_per_second": 2.939,
      "step": 790
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 3.131122589111328,
      "learning_rate": 1.9651772292626804e-05,
      "loss": 0.3737,
      "step": 800
    },
    {
      "epoch": 1.2307692307692308,
      "eval_loss": 0.7876585721969604,
      "eval_runtime": 2.0351,
      "eval_samples_per_second": 56.018,
      "eval_steps_per_second": 2.948,
      "step": 800
    },
    {
      "epoch": 1.2461538461538462,
      "grad_norm": 2.9482033252716064,
      "learning_rate": 1.9631419330345128e-05,
      "loss": 0.3664,
      "step": 810
    },
    {
      "epoch": 1.2461538461538462,
      "eval_loss": 0.8115803003311157,
      "eval_runtime": 2.0779,
      "eval_samples_per_second": 54.862,
      "eval_steps_per_second": 2.887,
      "step": 810
    },
    {
      "epoch": 1.2615384615384615,
      "grad_norm": 2.246203660964966,
      "learning_rate": 1.961049950247418e-05,
      "loss": 0.3071,
      "step": 820
    },
    {
      "epoch": 1.2615384615384615,
      "eval_loss": 0.8049949407577515,
      "eval_runtime": 2.0439,
      "eval_samples_per_second": 55.775,
      "eval_steps_per_second": 2.936,
      "step": 820
    },
    {
      "epoch": 1.2769230769230768,
      "grad_norm": 4.451783180236816,
      "learning_rate": 1.9589014040268678e-05,
      "loss": 0.3496,
      "step": 830
    },
    {
      "epoch": 1.2769230769230768,
      "eval_loss": 0.8066652417182922,
      "eval_runtime": 2.0386,
      "eval_samples_per_second": 55.92,
      "eval_steps_per_second": 2.943,
      "step": 830
    },
    {
      "epoch": 1.2923076923076924,
      "grad_norm": 2.6668195724487305,
      "learning_rate": 1.9566964208274254e-05,
      "loss": 0.3639,
      "step": 840
    },
    {
      "epoch": 1.2923076923076924,
      "eval_loss": 0.8018885850906372,
      "eval_runtime": 2.0355,
      "eval_samples_per_second": 56.007,
      "eval_steps_per_second": 2.948,
      "step": 840
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 2.0115177631378174,
      "learning_rate": 1.954435130425301e-05,
      "loss": 0.3899,
      "step": 850
    },
    {
      "epoch": 1.3076923076923077,
      "eval_loss": 0.7964260578155518,
      "eval_runtime": 2.0482,
      "eval_samples_per_second": 55.658,
      "eval_steps_per_second": 2.929,
      "step": 850
    },
    {
      "epoch": 1.323076923076923,
      "grad_norm": 2.5273962020874023,
      "learning_rate": 1.952117665910714e-05,
      "loss": 0.3742,
      "step": 860
    },
    {
      "epoch": 1.323076923076923,
      "eval_loss": 0.8197150826454163,
      "eval_runtime": 2.0642,
      "eval_samples_per_second": 55.227,
      "eval_steps_per_second": 2.907,
      "step": 860
    },
    {
      "epoch": 1.3384615384615386,
      "grad_norm": 1.823403239250183,
      "learning_rate": 1.949744163680062e-05,
      "loss": 0.3971,
      "step": 870
    },
    {
      "epoch": 1.3384615384615386,
      "eval_loss": 0.8292858004570007,
      "eval_runtime": 2.0374,
      "eval_samples_per_second": 55.955,
      "eval_steps_per_second": 2.945,
      "step": 870
    },
    {
      "epoch": 1.353846153846154,
      "grad_norm": 3.496244430541992,
      "learning_rate": 1.9473147634278884e-05,
      "loss": 0.4178,
      "step": 880
    },
    {
      "epoch": 1.353846153846154,
      "eval_loss": 0.8014177083969116,
      "eval_runtime": 2.0358,
      "eval_samples_per_second": 55.997,
      "eval_steps_per_second": 2.947,
      "step": 880
    },
    {
      "epoch": 1.3692307692307693,
      "grad_norm": 1.5359822511672974,
      "learning_rate": 1.9448296081386656e-05,
      "loss": 0.3711,
      "step": 890
    },
    {
      "epoch": 1.3692307692307693,
      "eval_loss": 0.8050973415374756,
      "eval_runtime": 2.0345,
      "eval_samples_per_second": 56.032,
      "eval_steps_per_second": 2.949,
      "step": 890
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 3.2398881912231445,
      "learning_rate": 1.9422888440783773e-05,
      "loss": 0.3689,
      "step": 900
    },
    {
      "epoch": 1.3846153846153846,
      "eval_loss": 0.7915543913841248,
      "eval_runtime": 2.042,
      "eval_samples_per_second": 55.827,
      "eval_steps_per_second": 2.938,
      "step": 900
    },
    {
      "epoch": 1.4,
      "grad_norm": 2.8013181686401367,
      "learning_rate": 1.9396926207859085e-05,
      "loss": 0.497,
      "step": 910
    },
    {
      "epoch": 1.4,
      "eval_loss": 0.7924531698226929,
      "eval_runtime": 2.0313,
      "eval_samples_per_second": 56.12,
      "eval_steps_per_second": 2.954,
      "step": 910
    },
    {
      "epoch": 1.4153846153846155,
      "grad_norm": 2.0076115131378174,
      "learning_rate": 1.9370410910642473e-05,
      "loss": 0.3979,
      "step": 920
    },
    {
      "epoch": 1.4153846153846155,
      "eval_loss": 0.8060823678970337,
      "eval_runtime": 2.0451,
      "eval_samples_per_second": 55.743,
      "eval_steps_per_second": 2.934,
      "step": 920
    },
    {
      "epoch": 1.4307692307692308,
      "grad_norm": 3.230391502380371,
      "learning_rate": 1.934334410971489e-05,
      "loss": 0.3975,
      "step": 930
    },
    {
      "epoch": 1.4307692307692308,
      "eval_loss": 0.7960232496261597,
      "eval_runtime": 2.0487,
      "eval_samples_per_second": 55.644,
      "eval_steps_per_second": 2.929,
      "step": 930
    },
    {
      "epoch": 1.4461538461538461,
      "grad_norm": 2.0214502811431885,
      "learning_rate": 1.9315727398116516e-05,
      "loss": 0.3024,
      "step": 940
    },
    {
      "epoch": 1.4461538461538461,
      "eval_loss": 0.8080164790153503,
      "eval_runtime": 2.0367,
      "eval_samples_per_second": 55.972,
      "eval_steps_per_second": 2.946,
      "step": 940
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 2.793571949005127,
      "learning_rate": 1.9287562401253023e-05,
      "loss": 0.42,
      "step": 950
    },
    {
      "epoch": 1.4615384615384617,
      "eval_loss": 0.7998307943344116,
      "eval_runtime": 2.0337,
      "eval_samples_per_second": 56.056,
      "eval_steps_per_second": 2.95,
      "step": 950
    },
    {
      "epoch": 1.476923076923077,
      "grad_norm": 3.0209238529205322,
      "learning_rate": 1.9258850776799875e-05,
      "loss": 0.4941,
      "step": 960
    },
    {
      "epoch": 1.476923076923077,
      "eval_loss": 0.7900478839874268,
      "eval_runtime": 2.08,
      "eval_samples_per_second": 54.807,
      "eval_steps_per_second": 2.885,
      "step": 960
    },
    {
      "epoch": 1.4923076923076923,
      "grad_norm": 1.660001516342163,
      "learning_rate": 1.9229594214604782e-05,
      "loss": 0.3708,
      "step": 970
    },
    {
      "epoch": 1.4923076923076923,
      "eval_loss": 0.7939229607582092,
      "eval_runtime": 2.0329,
      "eval_samples_per_second": 56.078,
      "eval_steps_per_second": 2.951,
      "step": 970
    },
    {
      "epoch": 1.5076923076923077,
      "grad_norm": 2.3737564086914062,
      "learning_rate": 1.9199794436588244e-05,
      "loss": 0.3624,
      "step": 980
    },
    {
      "epoch": 1.5076923076923077,
      "eval_loss": 0.8061041831970215,
      "eval_runtime": 2.037,
      "eval_samples_per_second": 55.965,
      "eval_steps_per_second": 2.946,
      "step": 980
    },
    {
      "epoch": 1.523076923076923,
      "grad_norm": 3.2474894523620605,
      "learning_rate": 1.9169453196642197e-05,
      "loss": 0.4532,
      "step": 990
    },
    {
      "epoch": 1.523076923076923,
      "eval_loss": 0.8081569075584412,
      "eval_runtime": 2.0354,
      "eval_samples_per_second": 56.009,
      "eval_steps_per_second": 2.948,
      "step": 990
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 1.9144352674484253,
      "learning_rate": 1.9138572280526795e-05,
      "loss": 0.3437,
      "step": 1000
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 0.799223005771637,
      "eval_runtime": 2.0389,
      "eval_samples_per_second": 55.913,
      "eval_steps_per_second": 2.943,
      "step": 1000
    }
  ],
  "logging_steps": 10,
  "max_steps": 4550,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 1000,
  "total_flos": 4.82382464459735e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}