{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.444444444444445,
  "eval_steps": 10,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08888888888888889,
      "grad_norm": 3.339972496032715,
      "learning_rate": 4e-05,
      "loss": 2.2556,
      "step": 10
    },
    {
      "epoch": 0.08888888888888889,
      "eval_loss": 1.899280071258545,
      "eval_runtime": 8.2861,
      "eval_samples_per_second": 12.068,
      "eval_steps_per_second": 12.068,
      "step": 10
    },
    {
      "epoch": 0.17777777777777778,
      "grad_norm": 4.27776575088501,
      "learning_rate": 8e-05,
      "loss": 1.5026,
      "step": 20
    },
    {
      "epoch": 0.17777777777777778,
      "eval_loss": 0.8436458706855774,
      "eval_runtime": 8.2809,
      "eval_samples_per_second": 12.076,
      "eval_steps_per_second": 12.076,
      "step": 20
    },
    {
      "epoch": 0.26666666666666666,
      "grad_norm": 2.9827308654785156,
      "learning_rate": 0.00012,
      "loss": 0.8291,
      "step": 30
    },
    {
      "epoch": 0.26666666666666666,
      "eval_loss": 0.6633767485618591,
      "eval_runtime": 8.2735,
      "eval_samples_per_second": 12.087,
      "eval_steps_per_second": 12.087,
      "step": 30
    },
    {
      "epoch": 0.35555555555555557,
      "grad_norm": 2.1213743686676025,
      "learning_rate": 0.00016,
      "loss": 0.6688,
      "step": 40
    },
    {
      "epoch": 0.35555555555555557,
      "eval_loss": 0.5723012685775757,
      "eval_runtime": 8.273,
      "eval_samples_per_second": 12.087,
      "eval_steps_per_second": 12.087,
      "step": 40
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 2.0819756984710693,
      "learning_rate": 0.0002,
      "loss": 0.6339,
      "step": 50
    },
    {
      "epoch": 0.4444444444444444,
      "eval_loss": 0.5510777831077576,
      "eval_runtime": 8.2675,
      "eval_samples_per_second": 12.096,
      "eval_steps_per_second": 12.096,
      "step": 50
    },
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 1.9045137166976929,
      "learning_rate": 0.00019975640502598244,
      "loss": 0.5258,
      "step": 60
    },
    {
      "epoch": 0.5333333333333333,
      "eval_loss": 0.4759778082370758,
      "eval_runtime": 8.2161,
      "eval_samples_per_second": 12.171,
      "eval_steps_per_second": 12.171,
      "step": 60
    },
    {
      "epoch": 0.6222222222222222,
      "grad_norm": 1.4711352586746216,
      "learning_rate": 0.00019902680687415705,
      "loss": 0.4825,
      "step": 70
    },
    {
      "epoch": 0.6222222222222222,
      "eval_loss": 0.4696222245693207,
      "eval_runtime": 8.0384,
      "eval_samples_per_second": 12.44,
      "eval_steps_per_second": 12.44,
      "step": 70
    },
    {
      "epoch": 0.7111111111111111,
      "grad_norm": 1.3467007875442505,
      "learning_rate": 0.00019781476007338058,
      "loss": 0.5488,
      "step": 80
    },
    {
      "epoch": 0.7111111111111111,
      "eval_loss": 0.4698783755302429,
      "eval_runtime": 8.0224,
      "eval_samples_per_second": 12.465,
      "eval_steps_per_second": 12.465,
      "step": 80
    },
    {
      "epoch": 0.8,
      "grad_norm": 1.3427356481552124,
      "learning_rate": 0.0001961261695938319,
      "loss": 0.4231,
      "step": 90
    },
    {
      "epoch": 0.8,
      "eval_loss": 0.45969489216804504,
      "eval_runtime": 8.0315,
      "eval_samples_per_second": 12.451,
      "eval_steps_per_second": 12.451,
      "step": 90
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 1.3022139072418213,
      "learning_rate": 0.00019396926207859084,
      "loss": 0.4558,
      "step": 100
    },
    {
      "epoch": 0.8888888888888888,
      "eval_loss": 0.4219018220901489,
      "eval_runtime": 8.0414,
      "eval_samples_per_second": 12.436,
      "eval_steps_per_second": 12.436,
      "step": 100
    },
    {
      "epoch": 0.9777777777777777,
      "grad_norm": 1.149550199508667,
      "learning_rate": 0.0001913545457642601,
      "loss": 0.4588,
      "step": 110
    },
    {
      "epoch": 0.9777777777777777,
      "eval_loss": 0.41754281520843506,
      "eval_runtime": 8.0271,
      "eval_samples_per_second": 12.458,
      "eval_steps_per_second": 12.458,
      "step": 110
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 1.335131049156189,
      "learning_rate": 0.00018829475928589271,
      "loss": 0.4592,
      "step": 120
    },
    {
      "epoch": 1.0666666666666667,
      "eval_loss": 0.4288429617881775,
      "eval_runtime": 8.0429,
      "eval_samples_per_second": 12.433,
      "eval_steps_per_second": 12.433,
      "step": 120
    },
    {
      "epoch": 1.1555555555555554,
      "grad_norm": 1.0977760553359985,
      "learning_rate": 0.0001848048096156426,
      "loss": 0.2996,
      "step": 130
    },
    {
      "epoch": 1.1555555555555554,
      "eval_loss": 0.3918496072292328,
      "eval_runtime": 8.0388,
      "eval_samples_per_second": 12.44,
      "eval_steps_per_second": 12.44,
      "step": 130
    },
    {
      "epoch": 1.2444444444444445,
      "grad_norm": 1.511498212814331,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.3269,
      "step": 140
    },
    {
      "epoch": 1.2444444444444445,
      "eval_loss": 0.41831883788108826,
      "eval_runtime": 8.0461,
      "eval_samples_per_second": 12.428,
      "eval_steps_per_second": 12.428,
      "step": 140
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 1.6732897758483887,
      "learning_rate": 0.0001766044443118978,
      "loss": 0.347,
      "step": 150
    },
    {
      "epoch": 1.3333333333333333,
      "eval_loss": 0.43138808012008667,
      "eval_runtime": 8.0313,
      "eval_samples_per_second": 12.451,
      "eval_steps_per_second": 12.451,
      "step": 150
    },
    {
      "epoch": 1.4222222222222223,
      "grad_norm": 1.5496058464050293,
      "learning_rate": 0.0001719339800338651,
      "loss": 0.3251,
      "step": 160
    },
    {
      "epoch": 1.4222222222222223,
      "eval_loss": 0.3889056444168091,
      "eval_runtime": 8.0322,
      "eval_samples_per_second": 12.45,
      "eval_steps_per_second": 12.45,
      "step": 160
    },
    {
      "epoch": 1.511111111111111,
      "grad_norm": 1.3534138202667236,
      "learning_rate": 0.00016691306063588583,
      "loss": 0.3035,
      "step": 170
    },
    {
      "epoch": 1.511111111111111,
      "eval_loss": 0.37887275218963623,
      "eval_runtime": 8.0354,
      "eval_samples_per_second": 12.445,
      "eval_steps_per_second": 12.445,
      "step": 170
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.4648184776306152,
      "learning_rate": 0.0001615661475325658,
      "loss": 0.3141,
      "step": 180
    },
    {
      "epoch": 1.6,
      "eval_loss": 0.38691258430480957,
      "eval_runtime": 8.0374,
      "eval_samples_per_second": 12.442,
      "eval_steps_per_second": 12.442,
      "step": 180
    },
    {
      "epoch": 1.6888888888888889,
      "grad_norm": 1.0812690258026123,
      "learning_rate": 0.0001559192903470747,
      "loss": 0.2878,
      "step": 190
    },
    {
      "epoch": 1.6888888888888889,
      "eval_loss": 0.3909819722175598,
      "eval_runtime": 8.0411,
      "eval_samples_per_second": 12.436,
      "eval_steps_per_second": 12.436,
      "step": 190
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 1.936132788658142,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.3063,
      "step": 200
    },
    {
      "epoch": 1.7777777777777777,
      "eval_loss": 0.39576366543769836,
      "eval_runtime": 8.0433,
      "eval_samples_per_second": 12.433,
      "eval_steps_per_second": 12.433,
      "step": 200
    },
    {
      "epoch": 1.8666666666666667,
      "grad_norm": 1.551062822341919,
      "learning_rate": 0.00014383711467890774,
      "loss": 0.2748,
      "step": 210
    },
    {
      "epoch": 1.8666666666666667,
      "eval_loss": 0.3819361627101898,
      "eval_runtime": 8.0342,
      "eval_samples_per_second": 12.447,
      "eval_steps_per_second": 12.447,
      "step": 210
    },
    {
      "epoch": 1.9555555555555557,
      "grad_norm": 1.2996041774749756,
      "learning_rate": 0.00013746065934159123,
      "loss": 0.2725,
      "step": 220
    },
    {
      "epoch": 1.9555555555555557,
      "eval_loss": 0.4039897620677948,
      "eval_runtime": 8.024,
      "eval_samples_per_second": 12.463,
      "eval_steps_per_second": 12.463,
      "step": 220
    },
    {
      "epoch": 2.0444444444444443,
      "grad_norm": 1.0756527185440063,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.2897,
      "step": 230
    },
    {
      "epoch": 2.0444444444444443,
      "eval_loss": 0.3928260803222656,
      "eval_runtime": 8.0625,
      "eval_samples_per_second": 12.403,
      "eval_steps_per_second": 12.403,
      "step": 230
    },
    {
      "epoch": 2.1333333333333333,
      "grad_norm": 0.8828668594360352,
      "learning_rate": 0.00012419218955996676,
      "loss": 0.1813,
      "step": 240
    },
    {
      "epoch": 2.1333333333333333,
      "eval_loss": 0.4048071801662445,
      "eval_runtime": 8.0422,
      "eval_samples_per_second": 12.434,
      "eval_steps_per_second": 12.434,
      "step": 240
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 1.6169154644012451,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.1965,
      "step": 250
    },
    {
      "epoch": 2.2222222222222223,
      "eval_loss": 0.4035675525665283,
      "eval_runtime": 8.0344,
      "eval_samples_per_second": 12.447,
      "eval_steps_per_second": 12.447,
      "step": 250
    },
    {
      "epoch": 2.311111111111111,
      "grad_norm": 1.6981542110443115,
      "learning_rate": 0.00011045284632676536,
      "loss": 0.1751,
      "step": 260
    },
    {
      "epoch": 2.311111111111111,
      "eval_loss": 0.4221409857273102,
      "eval_runtime": 8.0296,
      "eval_samples_per_second": 12.454,
      "eval_steps_per_second": 12.454,
      "step": 260
    },
    {
      "epoch": 2.4,
      "grad_norm": 1.5531835556030273,
      "learning_rate": 0.00010348994967025012,
      "loss": 0.1739,
      "step": 270
    },
    {
      "epoch": 2.4,
      "eval_loss": 0.40371406078338623,
      "eval_runtime": 8.017,
      "eval_samples_per_second": 12.474,
      "eval_steps_per_second": 12.474,
      "step": 270
    },
    {
      "epoch": 2.488888888888889,
      "grad_norm": 1.6275184154510498,
      "learning_rate": 9.651005032974994e-05,
      "loss": 0.1629,
      "step": 280
    },
    {
      "epoch": 2.488888888888889,
      "eval_loss": 0.41767382621765137,
      "eval_runtime": 8.026,
      "eval_samples_per_second": 12.46,
      "eval_steps_per_second": 12.46,
      "step": 280
    },
    {
      "epoch": 2.5777777777777775,
      "grad_norm": 1.6192083358764648,
      "learning_rate": 8.954715367323468e-05,
      "loss": 0.1919,
      "step": 290
    },
    {
      "epoch": 2.5777777777777775,
      "eval_loss": 0.4002458155155182,
      "eval_runtime": 8.0399,
      "eval_samples_per_second": 12.438,
      "eval_steps_per_second": 12.438,
      "step": 290
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 1.4782391786575317,
      "learning_rate": 8.263518223330697e-05,
      "loss": 0.1804,
      "step": 300
    },
    {
      "epoch": 2.6666666666666665,
      "eval_loss": 0.4098145365715027,
      "eval_runtime": 8.0329,
      "eval_samples_per_second": 12.449,
      "eval_steps_per_second": 12.449,
      "step": 300
    },
    {
      "epoch": 2.7555555555555555,
      "grad_norm": 1.2380214929580688,
      "learning_rate": 7.580781044003324e-05,
      "loss": 0.1569,
      "step": 310
    },
    {
      "epoch": 2.7555555555555555,
      "eval_loss": 0.41249945759773254,
      "eval_runtime": 8.0306,
      "eval_samples_per_second": 12.452,
      "eval_steps_per_second": 12.452,
      "step": 310
    },
    {
      "epoch": 2.8444444444444446,
      "grad_norm": 1.0430532693862915,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.1914,
      "step": 320
    },
    {
      "epoch": 2.8444444444444446,
      "eval_loss": 0.4051912724971771,
      "eval_runtime": 8.0353,
      "eval_samples_per_second": 12.445,
      "eval_steps_per_second": 12.445,
      "step": 320
    },
    {
      "epoch": 2.9333333333333336,
      "grad_norm": 0.8478929400444031,
      "learning_rate": 6.25393406584088e-05,
      "loss": 0.144,
      "step": 330
    },
    {
      "epoch": 2.9333333333333336,
      "eval_loss": 0.4041104018688202,
      "eval_runtime": 8.0292,
      "eval_samples_per_second": 12.455,
      "eval_steps_per_second": 12.455,
      "step": 330
    },
    {
      "epoch": 3.022222222222222,
      "grad_norm": 0.7308769822120667,
      "learning_rate": 5.616288532109225e-05,
      "loss": 0.1738,
      "step": 340
    },
    {
      "epoch": 3.022222222222222,
      "eval_loss": 0.42209112644195557,
      "eval_runtime": 8.0367,
      "eval_samples_per_second": 12.443,
      "eval_steps_per_second": 12.443,
      "step": 340
    },
    {
      "epoch": 3.111111111111111,
      "grad_norm": 0.47878319025039673,
      "learning_rate": 5.000000000000002e-05,
      "loss": 0.1087,
      "step": 350
    },
    {
      "epoch": 3.111111111111111,
      "eval_loss": 0.42136698961257935,
      "eval_runtime": 8.0346,
      "eval_samples_per_second": 12.446,
      "eval_steps_per_second": 12.446,
      "step": 350
    },
    {
      "epoch": 3.2,
      "grad_norm": 0.8724251389503479,
      "learning_rate": 4.4080709652925336e-05,
      "loss": 0.0876,
      "step": 360
    },
    {
      "epoch": 3.2,
      "eval_loss": 0.4379313588142395,
      "eval_runtime": 8.025,
      "eval_samples_per_second": 12.461,
      "eval_steps_per_second": 12.461,
      "step": 360
    },
    {
      "epoch": 3.2888888888888888,
      "grad_norm": 1.1260199546813965,
      "learning_rate": 3.843385246743417e-05,
      "loss": 0.0857,
      "step": 370
    },
    {
      "epoch": 3.2888888888888888,
      "eval_loss": 0.46546536684036255,
      "eval_runtime": 8.0261,
      "eval_samples_per_second": 12.459,
      "eval_steps_per_second": 12.459,
      "step": 370
    },
    {
      "epoch": 3.3777777777777778,
      "grad_norm": 0.8650846481323242,
      "learning_rate": 3.308693936411421e-05,
      "loss": 0.0978,
      "step": 380
    },
    {
      "epoch": 3.3777777777777778,
      "eval_loss": 0.4744359850883484,
      "eval_runtime": 8.0343,
      "eval_samples_per_second": 12.447,
      "eval_steps_per_second": 12.447,
      "step": 380
    },
    {
      "epoch": 3.466666666666667,
      "grad_norm": 1.4763766527175903,
      "learning_rate": 2.8066019966134904e-05,
      "loss": 0.0746,
      "step": 390
    },
    {
      "epoch": 3.466666666666667,
      "eval_loss": 0.4815245568752289,
      "eval_runtime": 8.0329,
      "eval_samples_per_second": 12.449,
      "eval_steps_per_second": 12.449,
      "step": 390
    },
    {
      "epoch": 3.5555555555555554,
      "grad_norm": 0.7600903511047363,
      "learning_rate": 2.339555568810221e-05,
      "loss": 0.0897,
      "step": 400
    },
    {
      "epoch": 3.5555555555555554,
      "eval_loss": 0.4889250695705414,
      "eval_runtime": 8.0394,
      "eval_samples_per_second": 12.439,
      "eval_steps_per_second": 12.439,
      "step": 400
    },
    {
      "epoch": 3.6444444444444444,
      "grad_norm": 0.7227004766464233,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0645,
      "step": 410
    },
    {
      "epoch": 3.6444444444444444,
      "eval_loss": 0.4995201826095581,
      "eval_runtime": 8.0321,
      "eval_samples_per_second": 12.45,
      "eval_steps_per_second": 12.45,
      "step": 410
    },
    {
      "epoch": 3.7333333333333334,
      "grad_norm": 1.2814204692840576,
      "learning_rate": 1.5195190384357404e-05,
      "loss": 0.0649,
      "step": 420
    },
    {
      "epoch": 3.7333333333333334,
      "eval_loss": 0.5078675746917725,
      "eval_runtime": 8.0419,
      "eval_samples_per_second": 12.435,
      "eval_steps_per_second": 12.435,
      "step": 420
    },
    {
      "epoch": 3.822222222222222,
      "grad_norm": 1.3571559190750122,
      "learning_rate": 1.1705240714107302e-05,
      "loss": 0.0896,
      "step": 430
    },
    {
      "epoch": 3.822222222222222,
      "eval_loss": 0.5097964406013489,
      "eval_runtime": 8.034,
      "eval_samples_per_second": 12.447,
      "eval_steps_per_second": 12.447,
      "step": 430
    },
    {
      "epoch": 3.911111111111111,
      "grad_norm": 1.125535488128662,
      "learning_rate": 8.645454235739903e-06,
      "loss": 0.0788,
      "step": 440
    },
    {
      "epoch": 3.911111111111111,
      "eval_loss": 0.5094956755638123,
      "eval_runtime": 8.0631,
      "eval_samples_per_second": 12.402,
      "eval_steps_per_second": 12.402,
      "step": 440
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.562880754470825,
      "learning_rate": 6.030737921409169e-06,
      "loss": 0.0886,
      "step": 450
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.51046222448349,
      "eval_runtime": 8.1706,
      "eval_samples_per_second": 12.239,
      "eval_steps_per_second": 12.239,
      "step": 450
    },
    {
      "epoch": 4.088888888888889,
      "grad_norm": 0.6980682611465454,
      "learning_rate": 3.873830406168111e-06,
      "loss": 0.0471,
      "step": 460
    },
    {
      "epoch": 4.088888888888889,
      "eval_loss": 0.511073887348175,
      "eval_runtime": 8.2725,
      "eval_samples_per_second": 12.088,
      "eval_steps_per_second": 12.088,
      "step": 460
    },
    {
      "epoch": 4.177777777777778,
      "grad_norm": 0.7668414115905762,
      "learning_rate": 2.1852399266194314e-06,
      "loss": 0.0461,
      "step": 470
    },
    {
      "epoch": 4.177777777777778,
      "eval_loss": 0.515235185623169,
      "eval_runtime": 8.2096,
      "eval_samples_per_second": 12.181,
      "eval_steps_per_second": 12.181,
      "step": 470
    },
    {
      "epoch": 4.266666666666667,
      "grad_norm": 1.7168527841567993,
      "learning_rate": 9.731931258429638e-07,
      "loss": 0.0607,
      "step": 480
    },
    {
      "epoch": 4.266666666666667,
      "eval_loss": 0.5151567459106445,
      "eval_runtime": 8.2315,
      "eval_samples_per_second": 12.149,
      "eval_steps_per_second": 12.149,
      "step": 480
    },
    {
      "epoch": 4.355555555555555,
      "grad_norm": 0.8711584806442261,
      "learning_rate": 2.4359497401758024e-07,
      "loss": 0.0473,
      "step": 490
    },
    {
      "epoch": 4.355555555555555,
      "eval_loss": 0.519190788269043,
      "eval_runtime": 8.2622,
      "eval_samples_per_second": 12.103,
      "eval_steps_per_second": 12.103,
      "step": 490
    },
    {
      "epoch": 4.444444444444445,
      "grad_norm": 1.0105704069137573,
      "learning_rate": 0.0,
      "loss": 0.052,
      "step": 500
    },
    {
      "epoch": 4.444444444444445,
      "eval_loss": 0.5163235664367676,
      "eval_runtime": 8.4465,
      "eval_samples_per_second": 11.839,
      "eval_steps_per_second": 11.839,
      "step": 500
    },
    {
      "epoch": 4.444444444444445,
      "step": 500,
      "total_flos": 1.496402435211264e+16,
      "train_loss": 0.3140584453344345,
      "train_runtime": 1600.6227,
      "train_samples_per_second": 2.499,
      "train_steps_per_second": 0.312
    }
  ],
  "logging_steps": 10,
  "max_steps": 500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.496402435211264e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}