{
  "best_global_step": 174,
  "best_metric": 0.14457035064697266,
  "best_model_checkpoint": "saves/lora/llama-3-8b-instruct/train_cb_1753094175/checkpoint-174",
  "epoch": 10.0,
  "eval_steps": 29,
  "global_step": 570,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.08771929824561403,
      "grad_norm": 7.463611125946045,
      "learning_rate": 3.5087719298245615e-06,
      "loss": 1.1748,
      "num_input_tokens_seen": 3552,
      "step": 5
    },
    {
      "epoch": 0.17543859649122806,
      "grad_norm": 6.506534576416016,
      "learning_rate": 7.894736842105263e-06,
      "loss": 1.2825,
      "num_input_tokens_seen": 7264,
      "step": 10
    },
    {
      "epoch": 0.2631578947368421,
      "grad_norm": 8.0122652053833,
      "learning_rate": 1.2280701754385964e-05,
      "loss": 1.0854,
      "num_input_tokens_seen": 10528,
      "step": 15
    },
    {
      "epoch": 0.3508771929824561,
      "grad_norm": 6.1492414474487305,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.7203,
      "num_input_tokens_seen": 14720,
      "step": 20
    },
    {
      "epoch": 0.43859649122807015,
      "grad_norm": 3.373422145843506,
      "learning_rate": 2.105263157894737e-05,
      "loss": 0.3206,
      "num_input_tokens_seen": 18016,
      "step": 25
    },
    {
      "epoch": 0.5087719298245614,
      "eval_loss": 0.21824319660663605,
      "eval_runtime": 0.648,
      "eval_samples_per_second": 38.579,
      "eval_steps_per_second": 10.802,
      "num_input_tokens_seen": 20064,
      "step": 29
    },
    {
      "epoch": 0.5263157894736842,
      "grad_norm": 1.7657883167266846,
      "learning_rate": 2.5438596491228074e-05,
      "loss": 0.2968,
      "num_input_tokens_seen": 20640,
      "step": 30
    },
    {
      "epoch": 0.6140350877192983,
      "grad_norm": 13.480871200561523,
      "learning_rate": 2.9824561403508772e-05,
      "loss": 0.1392,
      "num_input_tokens_seen": 24800,
      "step": 35
    },
    {
      "epoch": 0.7017543859649122,
      "grad_norm": 1.2553735971450806,
      "learning_rate": 3.421052631578947e-05,
      "loss": 0.0784,
      "num_input_tokens_seen": 28064,
      "step": 40
    },
    {
      "epoch": 0.7894736842105263,
      "grad_norm": 2.4903719425201416,
      "learning_rate": 3.859649122807018e-05,
      "loss": 0.3173,
      "num_input_tokens_seen": 30944,
      "step": 45
    },
    {
      "epoch": 0.8771929824561403,
      "grad_norm": 0.7033002376556396,
      "learning_rate": 4.298245614035088e-05,
      "loss": 0.0997,
      "num_input_tokens_seen": 33664,
      "step": 50
    },
    {
      "epoch": 0.9649122807017544,
      "grad_norm": 0.124003104865551,
      "learning_rate": 4.736842105263158e-05,
      "loss": 0.1923,
      "num_input_tokens_seen": 36320,
      "step": 55
    },
    {
      "epoch": 1.0175438596491229,
      "eval_loss": 0.24051079154014587,
      "eval_runtime": 0.6526,
      "eval_samples_per_second": 38.31,
      "eval_steps_per_second": 10.727,
      "num_input_tokens_seen": 37832,
      "step": 58
    },
    {
      "epoch": 1.0526315789473684,
      "grad_norm": 0.4836447238922119,
      "learning_rate": 4.999812487773597e-05,
      "loss": 0.0054,
      "num_input_tokens_seen": 39080,
      "step": 60
    },
    {
      "epoch": 1.1403508771929824,
      "grad_norm": 8.141298294067383,
      "learning_rate": 4.997703298253406e-05,
      "loss": 0.1719,
      "num_input_tokens_seen": 42536,
      "step": 65
    },
    {
      "epoch": 1.2280701754385965,
      "grad_norm": 0.14007717370986938,
      "learning_rate": 4.993252512887069e-05,
      "loss": 0.4013,
      "num_input_tokens_seen": 45608,
      "step": 70
    },
    {
      "epoch": 1.3157894736842106,
      "grad_norm": 0.6049325466156006,
      "learning_rate": 4.986464304284091e-05,
      "loss": 0.1023,
      "num_input_tokens_seen": 49352,
      "step": 75
    },
    {
      "epoch": 1.4035087719298245,
      "grad_norm": 10.203009605407715,
      "learning_rate": 4.977345036387331e-05,
      "loss": 0.1665,
      "num_input_tokens_seen": 52328,
      "step": 80
    },
    {
      "epoch": 1.4912280701754386,
      "grad_norm": 0.1354743093252182,
      "learning_rate": 4.965903258506806e-05,
      "loss": 0.0899,
      "num_input_tokens_seen": 56328,
      "step": 85
    },
    {
      "epoch": 1.526315789473684,
      "eval_loss": 0.20537658035755157,
      "eval_runtime": 0.6661,
      "eval_samples_per_second": 37.534,
      "eval_steps_per_second": 10.51,
      "num_input_tokens_seen": 57288,
      "step": 87
    },
    {
      "epoch": 1.5789473684210527,
      "grad_norm": 4.1772894859313965,
      "learning_rate": 4.952149697304716e-05,
      "loss": 0.3031,
      "num_input_tokens_seen": 59048,
      "step": 90
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.6734763979911804,
      "learning_rate": 4.9360972467392056e-05,
      "loss": 0.0187,
      "num_input_tokens_seen": 62504,
      "step": 95
    },
    {
      "epoch": 1.7543859649122808,
      "grad_norm": 0.049418896436691284,
      "learning_rate": 4.917760955976277e-05,
      "loss": 0.0838,
      "num_input_tokens_seen": 65832,
      "step": 100
    },
    {
      "epoch": 1.8421052631578947,
      "grad_norm": 3.6227242946624756,
      "learning_rate": 4.897158015281209e-05,
      "loss": 0.2078,
      "num_input_tokens_seen": 68808,
      "step": 105
    },
    {
      "epoch": 1.9298245614035088,
      "grad_norm": 0.028476709499955177,
      "learning_rate": 4.874307739902689e-05,
      "loss": 0.0884,
      "num_input_tokens_seen": 71848,
      "step": 110
    },
    {
      "epoch": 2.017543859649123,
      "grad_norm": 0.10558129101991653,
      "learning_rate": 4.849231551964771e-05,
      "loss": 0.0185,
      "num_input_tokens_seen": 74040,
      "step": 115
    },
    {
      "epoch": 2.0350877192982457,
      "eval_loss": 0.16205132007598877,
      "eval_runtime": 0.6617,
      "eval_samples_per_second": 37.781,
      "eval_steps_per_second": 10.579,
      "num_input_tokens_seen": 74520,
      "step": 116
    },
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 0.13294285535812378,
      "learning_rate": 4.821952960383649e-05,
      "loss": 0.0018,
      "num_input_tokens_seen": 77272,
      "step": 120
    },
    {
      "epoch": 2.192982456140351,
      "grad_norm": 0.0844767689704895,
      "learning_rate": 4.7924975388280524e-05,
      "loss": 0.1122,
      "num_input_tokens_seen": 80280,
      "step": 125
    },
    {
      "epoch": 2.280701754385965,
      "grad_norm": 0.09769173711538315,
      "learning_rate": 4.760892901743944e-05,
      "loss": 0.0236,
      "num_input_tokens_seen": 83480,
      "step": 130
    },
    {
      "epoch": 2.3684210526315788,
      "grad_norm": 2.08616042137146,
      "learning_rate": 4.727168678465988e-05,
      "loss": 0.1384,
      "num_input_tokens_seen": 86232,
      "step": 135
    },
    {
      "epoch": 2.456140350877193,
      "grad_norm": 0.8602607846260071,
      "learning_rate": 4.6913564854400595e-05,
      "loss": 0.0229,
      "num_input_tokens_seen": 89656,
      "step": 140
    },
    {
      "epoch": 2.543859649122807,
      "grad_norm": 0.8418223857879639,
      "learning_rate": 4.6534898965828405e-05,
      "loss": 0.2439,
      "num_input_tokens_seen": 93080,
      "step": 145
    },
    {
      "epoch": 2.543859649122807,
      "eval_loss": 0.22532634437084198,
      "eval_runtime": 0.6656,
      "eval_samples_per_second": 37.559,
      "eval_steps_per_second": 10.516,
      "num_input_tokens_seen": 93080,
      "step": 145
    },
    {
      "epoch": 2.6315789473684212,
      "grad_norm": 5.491137981414795,
      "learning_rate": 4.613604411806285e-05,
      "loss": 0.0258,
      "num_input_tokens_seen": 96440,
      "step": 150
    },
    {
      "epoch": 2.719298245614035,
      "grad_norm": 10.605635643005371,
      "learning_rate": 4.5717374237364665e-05,
      "loss": 0.0493,
      "num_input_tokens_seen": 100280,
      "step": 155
    },
    {
      "epoch": 2.807017543859649,
      "grad_norm": 0.011853941716253757,
      "learning_rate": 4.5279281826580056e-05,
      "loss": 0.0705,
      "num_input_tokens_seen": 103512,
      "step": 160
    },
    {
      "epoch": 2.8947368421052633,
      "grad_norm": 0.43100544810295105,
      "learning_rate": 4.482217759716946e-05,
      "loss": 0.0222,
      "num_input_tokens_seen": 106168,
      "step": 165
    },
    {
      "epoch": 2.982456140350877,
      "grad_norm": 0.012999014928936958,
      "learning_rate": 4.434649008416565e-05,
      "loss": 0.0828,
      "num_input_tokens_seen": 109624,
      "step": 170
    },
    {
      "epoch": 3.0526315789473686,
      "eval_loss": 0.14457035064697266,
      "eval_runtime": 0.6637,
      "eval_samples_per_second": 37.667,
      "eval_steps_per_second": 10.547,
      "num_input_tokens_seen": 111928,
      "step": 174
    },
    {
      "epoch": 3.0701754385964914,
      "grad_norm": 0.11448172479867935,
      "learning_rate": 4.385266524442241e-05,
      "loss": 0.0035,
      "num_input_tokens_seen": 112472,
      "step": 175
    },
    {
      "epoch": 3.1578947368421053,
      "grad_norm": 3.0042924880981445,
      "learning_rate": 4.334116603853007e-05,
      "loss": 0.0055,
      "num_input_tokens_seen": 115576,
      "step": 180
    },
    {
      "epoch": 3.245614035087719,
      "grad_norm": 0.006559327244758606,
      "learning_rate": 4.2812471996790206e-05,
      "loss": 0.0026,
      "num_input_tokens_seen": 119000,
      "step": 185
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 6.159653186798096,
      "learning_rate": 4.226707876965611e-05,
      "loss": 0.0181,
      "num_input_tokens_seen": 122360,
      "step": 190
    },
    {
      "epoch": 3.4210526315789473,
      "grad_norm": 0.010262958705425262,
      "learning_rate": 4.1705497663060767e-05,
      "loss": 0.1453,
      "num_input_tokens_seen": 125496,
      "step": 195
    },
    {
      "epoch": 3.5087719298245617,
      "grad_norm": 8.110419273376465,
      "learning_rate": 4.1128255159067665e-05,
      "loss": 0.0129,
      "num_input_tokens_seen": 128760,
      "step": 200
    },
    {
      "epoch": 3.56140350877193,
      "eval_loss": 0.16937656700611115,
      "eval_runtime": 0.6627,
      "eval_samples_per_second": 37.726,
      "eval_steps_per_second": 10.563,
      "num_input_tokens_seen": 131160,
      "step": 203
    },
    {
      "epoch": 3.5964912280701755,
      "grad_norm": 0.3041260242462158,
      "learning_rate": 4.053589242229412e-05,
      "loss": 0.0698,
      "num_input_tokens_seen": 132248,
      "step": 205
    },
    {
      "epoch": 3.6842105263157894,
      "grad_norm": 0.07038825005292892,
      "learning_rate": 3.9928964792569655e-05,
      "loss": 0.0012,
      "num_input_tokens_seen": 135160,
      "step": 210
    },
    {
      "epoch": 3.7719298245614032,
      "grad_norm": 1.2545245885849,
      "learning_rate": 3.930804126430513e-05,
      "loss": 0.0704,
      "num_input_tokens_seen": 138200,
      "step": 215
    },
    {
      "epoch": 3.8596491228070176,
      "grad_norm": 0.24765923619270325,
      "learning_rate": 3.867370395306068e-05,
      "loss": 0.0016,
      "num_input_tokens_seen": 141752,
      "step": 220
    },
    {
      "epoch": 3.9473684210526314,
      "grad_norm": 9.69217300415039,
      "learning_rate": 3.8026547549812665e-05,
      "loss": 0.0516,
      "num_input_tokens_seen": 146008,
      "step": 225
    },
    {
      "epoch": 4.035087719298246,
      "grad_norm": 0.01317143626511097,
      "learning_rate": 3.736717876343106e-05,
      "loss": 0.0193,
      "num_input_tokens_seen": 148648,
      "step": 230
    },
    {
      "epoch": 4.0701754385964914,
      "eval_loss": 0.17531278729438782,
      "eval_runtime": 0.6645,
      "eval_samples_per_second": 37.624,
      "eval_steps_per_second": 10.535,
      "num_input_tokens_seen": 150056,
      "step": 232
    },
    {
      "epoch": 4.12280701754386,
      "grad_norm": 0.20597302913665771,
      "learning_rate": 3.66962157518902e-05,
      "loss": 0.0014,
      "num_input_tokens_seen": 151656,
      "step": 235
    },
    {
      "epoch": 4.2105263157894735,
      "grad_norm": 0.26287728548049927,
      "learning_rate": 3.601428754274584e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 154280,
      "step": 240
    },
    {
      "epoch": 4.298245614035087,
      "grad_norm": 0.265356183052063,
      "learning_rate": 3.532203344342212e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 157160,
      "step": 245
    },
    {
      "epoch": 4.385964912280702,
      "grad_norm": 0.007273579481989145,
      "learning_rate": 3.4620102441861143e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 160072,
      "step": 250
    },
    {
      "epoch": 4.473684210526316,
      "grad_norm": 0.012681229040026665,
      "learning_rate": 3.390915259809696e-05,
      "loss": 0.0003,
      "num_input_tokens_seen": 163752,
      "step": 255
    },
    {
      "epoch": 4.56140350877193,
      "grad_norm": 0.07570221275091171,
      "learning_rate": 3.318985042732461e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 166600,
      "step": 260
    },
    {
      "epoch": 4.578947368421053,
      "eval_loss": 0.19881293177604675,
      "eval_runtime": 0.6652,
      "eval_samples_per_second": 37.584,
      "eval_steps_per_second": 10.523,
      "num_input_tokens_seen": 167208,
      "step": 261
    },
    {
      "epoch": 4.649122807017544,
      "grad_norm": 0.010587863624095917,
      "learning_rate": 3.246287027504237e-05,
      "loss": 0.0006,
      "num_input_tokens_seen": 169704,
      "step": 265
    },
    {
      "epoch": 4.7368421052631575,
      "grad_norm": 0.10257233679294586,
      "learning_rate": 3.172889368485311e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 173000,
      "step": 270
    },
    {
      "epoch": 4.824561403508772,
      "grad_norm": 0.005364793818444014,
      "learning_rate": 3.0988608759517475e-05,
      "loss": 0.0542,
      "num_input_tokens_seen": 177128,
      "step": 275
    },
    {
      "epoch": 4.912280701754386,
      "grad_norm": 0.01404496654868126,
      "learning_rate": 3.0242709515857758e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 180680,
      "step": 280
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.07565370202064514,
      "learning_rate": 2.949189523411747e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 183280,
      "step": 285
    },
    {
      "epoch": 5.087719298245614,
      "grad_norm": 0.0016017368761822581,
      "learning_rate": 2.8736869802386364e-05,
      "loss": 0.001,
      "num_input_tokens_seen": 186160,
      "step": 290
    },
    {
      "epoch": 5.087719298245614,
      "eval_loss": 0.24563251435756683,
      "eval_runtime": 0.6686,
      "eval_samples_per_second": 37.392,
      "eval_steps_per_second": 10.47,
      "num_input_tokens_seen": 186160,
      "step": 290
    },
    {
      "epoch": 5.175438596491228,
      "grad_norm": 0.02681572549045086,
      "learning_rate": 2.797834105670559e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 189584,
      "step": 295
    },
    {
      "epoch": 5.2631578947368425,
      "grad_norm": 0.05616031214594841,
      "learning_rate": 2.7217020117471793e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 193296,
      "step": 300
    },
    {
      "epoch": 5.350877192982456,
      "grad_norm": 0.0030404762364923954,
      "learning_rate": 2.6453620722761896e-05,
      "loss": 0.0065,
      "num_input_tokens_seen": 197328,
      "step": 305
    },
    {
      "epoch": 5.43859649122807,
      "grad_norm": 0.0018681815126910806,
      "learning_rate": 2.5688858559204053e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 200560,
      "step": 310
    },
    {
      "epoch": 5.526315789473684,
      "grad_norm": 0.003526732325553894,
      "learning_rate": 2.492345059102164e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 203632,
      "step": 315
    },
    {
      "epoch": 5.5964912280701755,
      "eval_loss": 0.26281487941741943,
      "eval_runtime": 0.6639,
      "eval_samples_per_second": 37.656,
      "eval_steps_per_second": 10.544,
      "num_input_tokens_seen": 206000,
      "step": 319
    },
    {
      "epoch": 5.614035087719298,
      "grad_norm": 0.3430648148059845,
      "learning_rate": 2.4158114387879616e-05,
      "loss": 0.0004,
      "num_input_tokens_seen": 206480,
      "step": 320
    },
    {
      "epoch": 5.701754385964913,
      "grad_norm": 0.003832002403214574,
      "learning_rate": 2.3393567452163252e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 209232,
      "step": 325
    },
    {
      "epoch": 5.7894736842105265,
      "grad_norm": 0.11517422646284103,
      "learning_rate": 2.2630526546319914e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 213168,
      "step": 330
    },
    {
      "epoch": 5.87719298245614,
      "grad_norm": 0.008495246060192585,
      "learning_rate": 2.186970702089457e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 216752,
      "step": 335
    },
    {
      "epoch": 5.964912280701754,
      "grad_norm": 0.001990408403798938,
      "learning_rate": 2.111182214388893e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 219536,
      "step": 340
    },
    {
      "epoch": 6.052631578947368,
      "grad_norm": 0.019063444808125496,
      "learning_rate": 2.0357582432072957e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 222272,
      "step": 345
    },
    {
      "epoch": 6.105263157894737,
      "eval_loss": 0.28356415033340454,
      "eval_runtime": 0.6499,
      "eval_samples_per_second": 38.465,
      "eval_steps_per_second": 10.77,
      "num_input_tokens_seen": 224064,
      "step": 348
    },
    {
      "epoch": 6.140350877192983,
      "grad_norm": 0.019526321440935135,
      "learning_rate": 1.9607694984875754e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 225664,
      "step": 350
    },
    {
      "epoch": 6.228070175438597,
      "grad_norm": 0.00943941343575716,
      "learning_rate": 1.8862862821480025e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 229408,
      "step": 355
    },
    {
      "epoch": 6.315789473684211,
      "grad_norm": 0.004007827956229448,
      "learning_rate": 1.8123784221741964e-05,
      "loss": 0.0015,
      "num_input_tokens_seen": 232832,
      "step": 360
    },
    {
      "epoch": 6.4035087719298245,
      "grad_norm": 0.002020841231569648,
      "learning_rate": 1.73911520715541e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 236288,
      "step": 365
    },
    {
      "epoch": 6.491228070175438,
      "grad_norm": 0.0033255487214773893,
      "learning_rate": 1.666565321326512e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 239296,
      "step": 370
    },
    {
      "epoch": 6.578947368421053,
      "grad_norm": 0.006292109843343496,
      "learning_rate": 1.5947967801765345e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 242848,
      "step": 375
    },
    {
      "epoch": 6.614035087719298,
      "eval_loss": 0.28133174777030945,
      "eval_runtime": 0.6588,
      "eval_samples_per_second": 37.945,
      "eval_steps_per_second": 10.625,
      "num_input_tokens_seen": 243840,
      "step": 377
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.001143688103184104,
      "learning_rate": 1.5238768666841907e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 245344,
      "step": 380
    },
    {
      "epoch": 6.754385964912281,
      "grad_norm": 0.006525792647153139,
      "learning_rate": 1.4538720682400969e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 248256,
      "step": 385
    },
    {
      "epoch": 6.842105263157895,
      "grad_norm": 0.1522936075925827,
      "learning_rate": 1.3848480143148839e-05,
      "loss": 0.0002,
      "num_input_tokens_seen": 250976,
      "step": 390
    },
    {
      "epoch": 6.9298245614035086,
      "grad_norm": 0.004834050312638283,
      "learning_rate": 1.3168694149315796e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 254944,
      "step": 395
    },
    {
      "epoch": 7.017543859649122,
      "grad_norm": 0.0014362959191203117,
      "learning_rate": 1.2500000000000006e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 257888,
      "step": 400
    },
    {
      "epoch": 7.105263157894737,
      "grad_norm": 0.0018230185378342867,
      "learning_rate": 1.1843024595699805e-05,
      "loss": 0.0,
      "num_input_tokens_seen": 261088,
      "step": 405
    },
    {
      "epoch": 7.12280701754386,
      "eval_loss": 0.2790299952030182,
      "eval_runtime": 0.6586,
      "eval_samples_per_second": 37.957,
      "eval_steps_per_second": 10.628,
      "num_input_tokens_seen": 261504,
      "step": 406
    },
    {
      "epoch": 7.192982456140351,
      "grad_norm": 0.001772104762494564,
      "learning_rate": 1.1198383850594758e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 263840,
      "step": 410
    },
    {
      "epoch": 7.280701754385965,
      "grad_norm": 0.017873741686344147,
      "learning_rate": 1.0566682115126344e-05,
      "loss": 0.0001,
      "num_input_tokens_seen": 267424,
      "step": 415
    },
    {
      "epoch": 7.368421052631579,
      "grad_norm": 0.007278971374034882,
      "learning_rate": 9.948511609419675e-06,
      "loss": 0.0003,
      "num_input_tokens_seen": 270720,
      "step": 420
    },
    {
      "epoch": 7.456140350877193,
      "grad_norm": 0.0015650499844923615,
      "learning_rate": 9.344451868077353e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 273504,
      "step": 425
    },
    {
      "epoch": 7.543859649122807,
      "grad_norm": 0.0008619703003205359,
      "learning_rate": 8.755069196866014e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 276608,
      "step": 430
    },
    {
      "epoch": 7.631578947368421,
      "grad_norm": 0.011649594642221928,
      "learning_rate": 8.180916141804906e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 280352,
      "step": 435
    },
    {
      "epoch": 7.631578947368421,
      "eval_loss": 0.28303590416908264,
      "eval_runtime": 0.6612,
      "eval_samples_per_second": 37.809,
      "eval_steps_per_second": 10.586,
      "num_input_tokens_seen": 280352,
      "step": 435
    },
    {
      "epoch": 7.719298245614035,
      "grad_norm": 0.002258860971778631,
      "learning_rate": 7.622530971154199e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 283808,
      "step": 440
    },
    {
      "epoch": 7.807017543859649,
      "grad_norm": 0.006624751258641481,
      "learning_rate": 7.080437170788723e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 286688,
      "step": 445
    },
    {
      "epoch": 7.894736842105263,
      "grad_norm": 0.01442085113376379,
      "learning_rate": 6.555142953430158e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 290240,
      "step": 450
    },
    {
      "epoch": 7.982456140350877,
      "grad_norm": 0.014971098862588406,
      "learning_rate": 6.0471407821978135e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 293568,
      "step": 455
    },
    {
      "epoch": 8.070175438596491,
      "grad_norm": 0.0020688732620328665,
      "learning_rate": 5.556906908924655e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 296720,
      "step": 460
    },
    {
      "epoch": 8.140350877192983,
      "eval_loss": 0.27810075879096985,
      "eval_runtime": 0.6632,
      "eval_samples_per_second": 37.694,
      "eval_steps_per_second": 10.554,
      "num_input_tokens_seen": 299344,
      "step": 464
    },
    {
      "epoch": 8.157894736842104,
      "grad_norm": 0.0014573490479961038,
      "learning_rate": 5.084900927671393e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 300112,
      "step": 465
    },
    {
      "epoch": 8.24561403508772,
      "grad_norm": 0.002585242036730051,
      "learning_rate": 4.631565343857239e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 303664,
      "step": 470
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 0.0032995608635246754,
      "learning_rate": 4.19732515941125e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 306480,
      "step": 475
    },
    {
      "epoch": 8.421052631578947,
      "grad_norm": 0.011491348966956139,
      "learning_rate": 3.7825874743331907e-06,
      "loss": 0.0002,
      "num_input_tokens_seen": 309744,
      "step": 480
    },
    {
      "epoch": 8.508771929824562,
      "grad_norm": 0.0024332385510206223,
      "learning_rate": 3.3877411050374424e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 313360,
      "step": 485
    },
    {
      "epoch": 8.596491228070175,
      "grad_norm": 0.001064626849256456,
      "learning_rate": 3.013156219837776e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 316784,
      "step": 490
    },
    {
      "epoch": 8.649122807017545,
      "eval_loss": 0.2798061668872833,
      "eval_runtime": 0.663,
      "eval_samples_per_second": 37.709,
      "eval_steps_per_second": 10.559,
      "num_input_tokens_seen": 318672,
      "step": 493
    },
    {
      "epoch": 8.68421052631579,
      "grad_norm": 0.008899637497961521,
      "learning_rate": 2.659183991914696e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 319760,
      "step": 495
    },
    {
      "epoch": 8.771929824561404,
      "grad_norm": 0.0012963354820385575,
      "learning_rate": 2.326156270090735e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 322512,
      "step": 500
    },
    {
      "epoch": 8.859649122807017,
      "grad_norm": 0.001958987908437848,
      "learning_rate": 2.0143852677223075e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 325936,
      "step": 505
    },
    {
      "epoch": 8.947368421052632,
      "grad_norm": 0.0009716018103063107,
      "learning_rate": 1.7241632699998123e-06,
      "loss": 0.0001,
      "num_input_tokens_seen": 329360,
      "step": 510
    },
    {
      "epoch": 9.035087719298245,
      "grad_norm": 0.00093101552920416,
      "learning_rate": 1.4557623599303903e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 332168,
      "step": 515
    },
    {
      "epoch": 9.12280701754386,
      "grad_norm": 0.0011983189033344388,
      "learning_rate": 1.2094341632602064e-06,
      "loss": 0.0,
      "num_input_tokens_seen": 336296,
      "step": 520
    },
    {
      "epoch": 9.157894736842104,
      "eval_loss": 0.27950653433799744,
      "eval_runtime": 0.6654,
      "eval_samples_per_second": 37.57,
      "eval_steps_per_second": 10.52,
      "num_input_tokens_seen": 337480,
      "step": 522
    },
    {
      "epoch": 9.210526315789474,
      "grad_norm": 0.014866072684526443,
      "learning_rate": 9.85409612575411e-07,
      "loss": 0.0001,
      "num_input_tokens_seen": 339080,
      "step": 525
    },
    {
      "epoch": 9.298245614035087,
      "grad_norm": 0.0015312348259612918,
      "learning_rate": 7.838987308029427e-07,
      "loss": 0.0,
      "num_input_tokens_seen": 342568,
      "step": 530
    },
    {
      "epoch": 9.385964912280702,
      "grad_norm": 0.0008261326584033668,
      "learning_rate": 6.050904343141095e-07,
      "loss": 0.0001,
      "num_input_tokens_seen": 345576,
      "step": 535
    },
    {
      "epoch": 9.473684210526315,
      "grad_norm": 0.0007581053068861365,
      "learning_rate": 4.491523558155714e-07,
      "loss": 0.0001,
      "num_input_tokens_seen": 348392,
      "step": 540
    },
    {
      "epoch": 9.56140350877193,
      "grad_norm": 0.0050800396129488945,
      "learning_rate": 3.162306871937387e-07,
      "loss": 0.0001,
      "num_input_tokens_seen": 351912,
      "step": 545
    },
    {
      "epoch": 9.649122807017545,
      "grad_norm": 0.0007958117639645934,
      "learning_rate": 2.064500424599436e-07,
      "loss": 0.0001,
      "num_input_tokens_seen": 355432,
      "step": 550
    },
    {
      "epoch": 9.666666666666666,
      "eval_loss": 0.2840792238712311,
      "eval_runtime": 0.6643,
      "eval_samples_per_second": 37.636,
      "eval_steps_per_second": 10.538,
      "num_input_tokens_seen": 356456,
      "step": 551
    },
    {
      "epoch": 9.736842105263158,
      "grad_norm": 0.002019475679844618,
      "learning_rate": 1.1991334092484318e-07,
      "loss": 0.0001,
      "num_input_tokens_seen": 358760,
      "step": 555
    },
    {
      "epoch": 9.824561403508772,
      "grad_norm": 0.0015096471179276705,
      "learning_rate": 5.6701710711626334e-08,
      "loss": 0.0001,
      "num_input_tokens_seen": 362088,
      "step": 560
    },
    {
      "epoch": 9.912280701754385,
      "grad_norm": 0.004095335025340319,
      "learning_rate": 1.6874412698408836e-08,
      "loss": 0.0001,
      "num_input_tokens_seen": 365544,
      "step": 565
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.001615795074030757,
      "learning_rate": 4.687849611939576e-10,
      "loss": 0.0002,
      "num_input_tokens_seen": 367864,
      "step": 570
    },
    {
      "epoch": 10.0,
      "num_input_tokens_seen": 367864,
      "step": 570,
      "total_flos": 1.6597151259623424e+16,
      "train_loss": 0.07575432408824175,
      "train_runtime": 187.3116,
      "train_samples_per_second": 12.012,
      "train_steps_per_second": 3.043
    }
  ],
  "logging_steps": 5,
  "max_steps": 570,
  "num_input_tokens_seen": 367864,
  "num_train_epochs": 10,
  "save_steps": 29,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.6597151259623424e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}