{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.5384615384615383,
  "eval_steps": 10,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.006153846153846154,
      "eval_loss": 1.487821340560913,
      "eval_runtime": 1.2374,
      "eval_samples_per_second": 92.131,
      "eval_steps_per_second": 4.849,
      "step": 2
    },
    {
      "epoch": 0.03076923076923077,
      "grad_norm": 20.007999420166016,
      "learning_rate": 6.153846153846155e-07,
      "loss": 1.4703,
      "step": 10
    },
    {
      "epoch": 0.03076923076923077,
      "eval_loss": 1.4221965074539185,
      "eval_runtime": 1.1701,
      "eval_samples_per_second": 97.431,
      "eval_steps_per_second": 5.128,
      "step": 10
    },
    {
      "epoch": 0.06153846153846154,
      "grad_norm": 29.657033920288086,
      "learning_rate": 1.230769230769231e-06,
      "loss": 1.4402,
      "step": 20
    },
    {
      "epoch": 0.06153846153846154,
      "eval_loss": 1.238189935684204,
      "eval_runtime": 1.1729,
      "eval_samples_per_second": 97.195,
      "eval_steps_per_second": 5.116,
      "step": 20
    },
    {
      "epoch": 0.09230769230769231,
      "grad_norm": 10.80688190460205,
      "learning_rate": 1.8461538461538465e-06,
      "loss": 1.0957,
      "step": 30
    },
    {
      "epoch": 0.09230769230769231,
      "eval_loss": 0.908875048160553,
      "eval_runtime": 1.1739,
      "eval_samples_per_second": 97.111,
      "eval_steps_per_second": 5.111,
      "step": 30
    },
    {
      "epoch": 0.12307692307692308,
      "grad_norm": 12.914667129516602,
      "learning_rate": 2.461538461538462e-06,
      "loss": 0.8136,
      "step": 40
    },
    {
      "epoch": 0.12307692307692308,
      "eval_loss": 0.7436326146125793,
      "eval_runtime": 1.1712,
      "eval_samples_per_second": 97.34,
      "eval_steps_per_second": 5.123,
      "step": 40
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 5.206438064575195,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.7864,
      "step": 50
    },
    {
      "epoch": 0.15384615384615385,
      "eval_loss": 0.6924600601196289,
      "eval_runtime": 1.1718,
      "eval_samples_per_second": 97.287,
      "eval_steps_per_second": 5.12,
      "step": 50
    },
    {
      "epoch": 0.18461538461538463,
      "grad_norm": 5.670599460601807,
      "learning_rate": 3.692307692307693e-06,
      "loss": 0.622,
      "step": 60
    },
    {
      "epoch": 0.18461538461538463,
      "eval_loss": 0.6791077852249146,
      "eval_runtime": 1.1678,
      "eval_samples_per_second": 97.62,
      "eval_steps_per_second": 5.138,
      "step": 60
    },
    {
      "epoch": 0.2153846153846154,
      "grad_norm": 3.6266868114471436,
      "learning_rate": 4.307692307692308e-06,
      "loss": 0.6672,
      "step": 70
    },
    {
      "epoch": 0.2153846153846154,
      "eval_loss": 0.6631948947906494,
      "eval_runtime": 1.1626,
      "eval_samples_per_second": 98.06,
      "eval_steps_per_second": 5.161,
      "step": 70
    },
    {
      "epoch": 0.24615384615384617,
      "grad_norm": 3.1027884483337402,
      "learning_rate": 4.923076923076924e-06,
      "loss": 0.6097,
      "step": 80
    },
    {
      "epoch": 0.24615384615384617,
      "eval_loss": 0.6625475883483887,
      "eval_runtime": 1.1717,
      "eval_samples_per_second": 97.294,
      "eval_steps_per_second": 5.121,
      "step": 80
    },
    {
      "epoch": 0.27692307692307694,
      "grad_norm": 3.233839511871338,
      "learning_rate": 5.538461538461539e-06,
      "loss": 0.6957,
      "step": 90
    },
    {
      "epoch": 0.27692307692307694,
      "eval_loss": 0.6590337157249451,
      "eval_runtime": 1.1664,
      "eval_samples_per_second": 97.738,
      "eval_steps_per_second": 5.144,
      "step": 90
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 4.932614803314209,
      "learning_rate": 6.153846153846155e-06,
      "loss": 0.6994,
      "step": 100
    },
    {
      "epoch": 0.3076923076923077,
      "eval_loss": 0.6592673659324646,
      "eval_runtime": 1.1692,
      "eval_samples_per_second": 97.502,
      "eval_steps_per_second": 5.132,
      "step": 100
    },
    {
      "epoch": 0.3384615384615385,
      "grad_norm": 2.642303466796875,
      "learning_rate": 6.76923076923077e-06,
      "loss": 0.6501,
      "step": 110
    },
    {
      "epoch": 0.3384615384615385,
      "eval_loss": 0.6589992642402649,
      "eval_runtime": 1.1669,
      "eval_samples_per_second": 97.691,
      "eval_steps_per_second": 5.142,
      "step": 110
    },
    {
      "epoch": 0.36923076923076925,
      "grad_norm": 3.45409893989563,
      "learning_rate": 7.384615384615386e-06,
      "loss": 0.7512,
      "step": 120
    },
    {
      "epoch": 0.36923076923076925,
      "eval_loss": 0.6568591594696045,
      "eval_runtime": 1.1778,
      "eval_samples_per_second": 96.79,
      "eval_steps_per_second": 5.094,
      "step": 120
    },
    {
      "epoch": 0.4,
      "grad_norm": 8.460351943969727,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.5996,
      "step": 130
    },
    {
      "epoch": 0.4,
      "eval_loss": 0.6578297019004822,
      "eval_runtime": 1.1683,
      "eval_samples_per_second": 97.575,
      "eval_steps_per_second": 5.136,
      "step": 130
    },
    {
      "epoch": 0.4307692307692308,
      "grad_norm": 2.4495465755462646,
      "learning_rate": 8.615384615384617e-06,
      "loss": 0.6498,
      "step": 140
    },
    {
      "epoch": 0.4307692307692308,
      "eval_loss": 0.6588199138641357,
      "eval_runtime": 1.1664,
      "eval_samples_per_second": 97.739,
      "eval_steps_per_second": 5.144,
      "step": 140
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 2.6779162883758545,
      "learning_rate": 9.230769230769232e-06,
      "loss": 0.703,
      "step": 150
    },
    {
      "epoch": 0.46153846153846156,
      "eval_loss": 0.6626676321029663,
      "eval_runtime": 1.1764,
      "eval_samples_per_second": 96.904,
      "eval_steps_per_second": 5.1,
      "step": 150
    },
    {
      "epoch": 0.49230769230769234,
      "grad_norm": 8.147880554199219,
      "learning_rate": 9.846153846153848e-06,
      "loss": 0.6649,
      "step": 160
    },
    {
      "epoch": 0.49230769230769234,
      "eval_loss": 0.6644802689552307,
      "eval_runtime": 1.1701,
      "eval_samples_per_second": 97.424,
      "eval_steps_per_second": 5.128,
      "step": 160
    },
    {
      "epoch": 0.5230769230769231,
      "grad_norm": 2.7410728931427,
      "learning_rate": 1.0461538461538463e-05,
      "loss": 0.6704,
      "step": 170
    },
    {
      "epoch": 0.5230769230769231,
      "eval_loss": 0.6591455340385437,
      "eval_runtime": 1.1722,
      "eval_samples_per_second": 97.253,
      "eval_steps_per_second": 5.119,
      "step": 170
    },
    {
      "epoch": 0.5538461538461539,
      "grad_norm": 6.886598587036133,
      "learning_rate": 1.1076923076923079e-05,
      "loss": 0.6073,
      "step": 180
    },
    {
      "epoch": 0.5538461538461539,
      "eval_loss": 0.6617754697799683,
      "eval_runtime": 1.1682,
      "eval_samples_per_second": 97.588,
      "eval_steps_per_second": 5.136,
      "step": 180
    },
    {
      "epoch": 0.5846153846153846,
      "grad_norm": 4.26249361038208,
      "learning_rate": 1.1692307692307694e-05,
      "loss": 0.7556,
      "step": 190
    },
    {
      "epoch": 0.5846153846153846,
      "eval_loss": 0.6659030318260193,
      "eval_runtime": 1.1721,
      "eval_samples_per_second": 97.265,
      "eval_steps_per_second": 5.119,
      "step": 190
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 4.204222202301025,
      "learning_rate": 1.230769230769231e-05,
      "loss": 0.6583,
      "step": 200
    },
    {
      "epoch": 0.6153846153846154,
      "eval_loss": 0.6660778522491455,
      "eval_runtime": 1.1728,
      "eval_samples_per_second": 97.201,
      "eval_steps_per_second": 5.116,
      "step": 200
    },
    {
      "epoch": 0.6461538461538462,
      "grad_norm": 3.154658079147339,
      "learning_rate": 1.2923076923076925e-05,
      "loss": 0.7288,
      "step": 210
    },
    {
      "epoch": 0.6461538461538462,
      "eval_loss": 0.6642422676086426,
      "eval_runtime": 1.166,
      "eval_samples_per_second": 97.772,
      "eval_steps_per_second": 5.146,
      "step": 210
    },
    {
      "epoch": 0.676923076923077,
      "grad_norm": 3.248359203338623,
      "learning_rate": 1.353846153846154e-05,
      "loss": 0.6702,
      "step": 220
    },
    {
      "epoch": 0.676923076923077,
      "eval_loss": 0.6680281758308411,
      "eval_runtime": 1.1691,
      "eval_samples_per_second": 97.507,
      "eval_steps_per_second": 5.132,
      "step": 220
    },
    {
      "epoch": 0.7076923076923077,
      "grad_norm": 2.5507216453552246,
      "learning_rate": 1.4153846153846156e-05,
      "loss": 0.66,
      "step": 230
    },
    {
      "epoch": 0.7076923076923077,
      "eval_loss": 0.6713310480117798,
      "eval_runtime": 1.167,
      "eval_samples_per_second": 97.683,
      "eval_steps_per_second": 5.141,
      "step": 230
    },
    {
      "epoch": 0.7384615384615385,
      "grad_norm": 2.1180007457733154,
      "learning_rate": 1.4769230769230772e-05,
      "loss": 0.6172,
      "step": 240
    },
    {
      "epoch": 0.7384615384615385,
      "eval_loss": 0.6759226322174072,
      "eval_runtime": 1.1758,
      "eval_samples_per_second": 96.954,
      "eval_steps_per_second": 5.103,
      "step": 240
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 2.8318371772766113,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.5954,
      "step": 250
    },
    {
      "epoch": 0.7692307692307693,
      "eval_loss": 0.6854314208030701,
      "eval_runtime": 1.1698,
      "eval_samples_per_second": 97.457,
      "eval_steps_per_second": 5.129,
      "step": 250
    },
    {
      "epoch": 0.8,
      "grad_norm": 3.519829034805298,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.6639,
      "step": 260
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.6849851012229919,
      "eval_runtime": 1.1732,
      "eval_samples_per_second": 97.173,
      "eval_steps_per_second": 5.114,
      "step": 260
    },
    {
      "epoch": 0.8307692307692308,
      "grad_norm": 3.7909388542175293,
      "learning_rate": 1.6615384615384618e-05,
      "loss": 0.6892,
      "step": 270
    },
    {
      "epoch": 0.8307692307692308,
      "eval_loss": 0.6863855123519897,
      "eval_runtime": 1.1689,
      "eval_samples_per_second": 97.53,
      "eval_steps_per_second": 5.133,
      "step": 270
    },
    {
      "epoch": 0.8615384615384616,
      "grad_norm": 8.278326988220215,
      "learning_rate": 1.7230769230769234e-05,
      "loss": 0.6014,
      "step": 280
    },
    {
      "epoch": 0.8615384615384616,
      "eval_loss": 0.6890373826026917,
      "eval_runtime": 1.1722,
      "eval_samples_per_second": 97.251,
      "eval_steps_per_second": 5.118,
      "step": 280
    },
    {
      "epoch": 0.8923076923076924,
      "grad_norm": 2.616001605987549,
      "learning_rate": 1.784615384615385e-05,
      "loss": 0.6775,
      "step": 290
    },
    {
      "epoch": 0.8923076923076924,
      "eval_loss": 0.6797110438346863,
      "eval_runtime": 1.1888,
      "eval_samples_per_second": 95.897,
      "eval_steps_per_second": 5.047,
      "step": 290
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.4450528621673584,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 0.7434,
      "step": 300
    },
    {
      "epoch": 0.9230769230769231,
      "eval_loss": 0.6779448986053467,
      "eval_runtime": 1.1722,
      "eval_samples_per_second": 97.254,
      "eval_steps_per_second": 5.119,
      "step": 300
    },
    {
      "epoch": 0.9538461538461539,
      "grad_norm": 4.121164798736572,
      "learning_rate": 1.907692307692308e-05,
      "loss": 0.6413,
      "step": 310
    },
    {
      "epoch": 0.9538461538461539,
      "eval_loss": 0.6819297671318054,
      "eval_runtime": 1.1683,
      "eval_samples_per_second": 97.574,
      "eval_steps_per_second": 5.135,
      "step": 310
    },
    {
      "epoch": 0.9846153846153847,
      "grad_norm": 2.057229995727539,
      "learning_rate": 1.9692307692307696e-05,
      "loss": 0.6511,
      "step": 320
    },
    {
      "epoch": 0.9846153846153847,
      "eval_loss": 0.6861481666564941,
      "eval_runtime": 1.1699,
      "eval_samples_per_second": 97.444,
      "eval_steps_per_second": 5.129,
      "step": 320
    },
    {
      "epoch": 1.0153846153846153,
      "grad_norm": 1.7295210361480713,
      "learning_rate": 1.9999855802751384e-05,
      "loss": 0.5872,
      "step": 330
    },
    {
      "epoch": 1.0153846153846153,
      "eval_loss": 0.7481781244277954,
      "eval_runtime": 1.172,
      "eval_samples_per_second": 97.273,
      "eval_steps_per_second": 5.12,
      "step": 330
    },
    {
      "epoch": 1.0461538461538462,
      "grad_norm": 3.6910336017608643,
      "learning_rate": 1.9998702249713747e-05,
      "loss": 0.438,
      "step": 340
    },
    {
      "epoch": 1.0461538461538462,
      "eval_loss": 0.7471169829368591,
      "eval_runtime": 1.175,
      "eval_samples_per_second": 97.019,
      "eval_steps_per_second": 5.106,
      "step": 340
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 3.9220831394195557,
      "learning_rate": 1.9996395276708856e-05,
      "loss": 0.4141,
      "step": 350
    },
    {
      "epoch": 1.0769230769230769,
      "eval_loss": 0.7425897121429443,
      "eval_runtime": 1.1888,
      "eval_samples_per_second": 95.893,
      "eval_steps_per_second": 5.047,
      "step": 350
    },
    {
      "epoch": 1.1076923076923078,
      "grad_norm": 2.1136879920959473,
      "learning_rate": 1.9992935149862116e-05,
      "loss": 0.4173,
      "step": 360
    },
    {
      "epoch": 1.1076923076923078,
      "eval_loss": 0.7477542161941528,
      "eval_runtime": 1.1705,
      "eval_samples_per_second": 97.398,
      "eval_steps_per_second": 5.126,
      "step": 360
    },
    {
      "epoch": 1.1384615384615384,
      "grad_norm": 2.187299966812134,
      "learning_rate": 1.998832226832327e-05,
      "loss": 0.4242,
      "step": 370
    },
    {
      "epoch": 1.1384615384615384,
      "eval_loss": 0.7475994229316711,
      "eval_runtime": 1.1683,
      "eval_samples_per_second": 97.582,
      "eval_steps_per_second": 5.136,
      "step": 370
    },
    {
      "epoch": 1.1692307692307693,
      "grad_norm": 2.683750629425049,
      "learning_rate": 1.9982557164220335e-05,
      "loss": 0.4413,
      "step": 380
    },
    {
      "epoch": 1.1692307692307693,
      "eval_loss": 0.7552060484886169,
      "eval_runtime": 1.1689,
      "eval_samples_per_second": 97.529,
      "eval_steps_per_second": 5.133,
      "step": 380
    },
    {
      "epoch": 1.2,
      "grad_norm": 3.312657117843628,
      "learning_rate": 1.9975640502598243e-05,
      "loss": 0.3361,
      "step": 390
    },
    {
      "epoch": 1.2,
      "eval_loss": 0.7630029916763306,
      "eval_runtime": 1.1689,
      "eval_samples_per_second": 97.529,
      "eval_steps_per_second": 5.133,
      "step": 390
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 2.3290536403656006,
      "learning_rate": 1.9967573081342103e-05,
      "loss": 0.4174,
      "step": 400
    },
    {
      "epoch": 1.2307692307692308,
      "eval_loss": 0.7669293880462646,
      "eval_runtime": 1.1711,
      "eval_samples_per_second": 97.345,
      "eval_steps_per_second": 5.123,
      "step": 400
    },
    {
      "epoch": 1.2615384615384615,
      "grad_norm": 2.4483697414398193,
      "learning_rate": 1.9958355831085155e-05,
      "loss": 0.4045,
      "step": 410
    },
    {
      "epoch": 1.2615384615384615,
      "eval_loss": 0.7652024626731873,
      "eval_runtime": 1.1697,
      "eval_samples_per_second": 97.461,
      "eval_steps_per_second": 5.13,
      "step": 410
    },
    {
      "epoch": 1.2923076923076924,
      "grad_norm": 1.703616976737976,
      "learning_rate": 1.9947989815101444e-05,
      "loss": 0.4292,
      "step": 420
    },
    {
      "epoch": 1.2923076923076924,
      "eval_loss": 0.7553891539573669,
      "eval_runtime": 1.1719,
      "eval_samples_per_second": 97.277,
      "eval_steps_per_second": 5.12,
      "step": 420
    },
    {
      "epoch": 1.323076923076923,
      "grad_norm": 1.6245399713516235,
      "learning_rate": 1.9936476229183133e-05,
      "loss": 0.4164,
      "step": 430
    },
    {
      "epoch": 1.323076923076923,
      "eval_loss": 0.7711736559867859,
      "eval_runtime": 1.1776,
      "eval_samples_per_second": 96.806,
      "eval_steps_per_second": 5.095,
      "step": 430
    },
    {
      "epoch": 1.353846153846154,
      "grad_norm": 1.9511611461639404,
      "learning_rate": 1.992381640150257e-05,
      "loss": 0.4337,
      "step": 440
    },
    {
      "epoch": 1.353846153846154,
      "eval_loss": 0.7438372373580933,
      "eval_runtime": 1.1692,
      "eval_samples_per_second": 97.5,
      "eval_steps_per_second": 5.132,
      "step": 440
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 2.466737747192383,
      "learning_rate": 1.9910011792459086e-05,
      "loss": 0.3712,
      "step": 450
    },
    {
      "epoch": 1.3846153846153846,
      "eval_loss": 0.8045614361763,
      "eval_runtime": 1.1689,
      "eval_samples_per_second": 97.53,
      "eval_steps_per_second": 5.133,
      "step": 450
    },
    {
      "epoch": 1.4153846153846155,
      "grad_norm": 2.473851203918457,
      "learning_rate": 1.9895063994510512e-05,
      "loss": 0.516,
      "step": 460
    },
    {
      "epoch": 1.4153846153846155,
      "eval_loss": 0.7463352084159851,
      "eval_runtime": 1.1697,
      "eval_samples_per_second": 97.464,
      "eval_steps_per_second": 5.13,
      "step": 460
    },
    {
      "epoch": 1.4461538461538461,
      "grad_norm": 2.757955312728882,
      "learning_rate": 1.9878974731989487e-05,
      "loss": 0.4965,
      "step": 470
    },
    {
      "epoch": 1.4461538461538461,
      "eval_loss": 0.7571713924407959,
      "eval_runtime": 1.1688,
      "eval_samples_per_second": 97.534,
      "eval_steps_per_second": 5.133,
      "step": 470
    },
    {
      "epoch": 1.476923076923077,
      "grad_norm": 3.9884915351867676,
      "learning_rate": 1.9861745860904538e-05,
      "loss": 0.3973,
      "step": 480
    },
    {
      "epoch": 1.476923076923077,
      "eval_loss": 0.7763081789016724,
      "eval_runtime": 1.168,
      "eval_samples_per_second": 97.605,
      "eval_steps_per_second": 5.137,
      "step": 480
    },
    {
      "epoch": 1.5076923076923077,
      "grad_norm": 1.9474631547927856,
      "learning_rate": 1.9843379368725978e-05,
      "loss": 0.4625,
      "step": 490
    },
    {
      "epoch": 1.5076923076923077,
      "eval_loss": 0.7595367431640625,
      "eval_runtime": 1.1708,
      "eval_samples_per_second": 97.37,
      "eval_steps_per_second": 5.125,
      "step": 490
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 2.2907869815826416,
      "learning_rate": 1.9823877374156647e-05,
      "loss": 0.4371,
      "step": 500
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 0.762865424156189,
      "eval_runtime": 1.1766,
      "eval_samples_per_second": 96.891,
      "eval_steps_per_second": 5.1,
      "step": 500
    }
  ],
  "logging_steps": 10,
  "max_steps": 3250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 50,
  "total_flos": 3.118537414331597e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}