{
  "best_metric": 0.8779178261756897,
  "best_model_checkpoint": "saves/CADICA_qwenvl_stenosis_classily_scale4_frozenVision/lora/sft/checkpoint-850",
  "epoch": 0.43780582024208087,
  "eval_steps": 50,
  "global_step": 850,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0025753283543651817,
      "grad_norm": 21.336819681898895,
      "learning_rate": 2.9411764705882355e-06,
      "loss": 3.0444,
      "num_input_tokens_seen": 58496,
      "step": 5
    },
    {
      "epoch": 0.0051506567087303634,
      "grad_norm": 20.576623155848594,
      "learning_rate": 5.882352941176471e-06,
      "loss": 2.9824,
      "num_input_tokens_seen": 116960,
      "step": 10
    },
    {
      "epoch": 0.007725985063095545,
      "grad_norm": 22.989873871108518,
      "learning_rate": 8.823529411764707e-06,
      "loss": 2.8371,
      "num_input_tokens_seen": 175448,
      "step": 15
    },
    {
      "epoch": 0.010301313417460727,
      "grad_norm": 19.533434089690918,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 2.5198,
      "num_input_tokens_seen": 233944,
      "step": 20
    },
    {
      "epoch": 0.012876641771825908,
      "grad_norm": 12.509494197145006,
      "learning_rate": 1.4705882352941177e-05,
      "loss": 1.772,
      "num_input_tokens_seen": 292416,
      "step": 25
    },
    {
      "epoch": 0.01545197012619109,
      "grad_norm": 3.6901887027066667,
      "learning_rate": 1.7647058823529414e-05,
      "loss": 1.2263,
      "num_input_tokens_seen": 350904,
      "step": 30
    },
    {
      "epoch": 0.018027298480556272,
      "grad_norm": 2.3996076770849744,
      "learning_rate": 2.058823529411765e-05,
      "loss": 1.0102,
      "num_input_tokens_seen": 409384,
      "step": 35
    },
    {
      "epoch": 0.020602626834921454,
      "grad_norm": 0.9253415848864577,
      "learning_rate": 2.3529411764705884e-05,
      "loss": 0.9378,
      "num_input_tokens_seen": 467864,
      "step": 40
    },
    {
      "epoch": 0.023177955189286635,
      "grad_norm": 1.1966244115097795,
      "learning_rate": 2.647058823529412e-05,
      "loss": 0.9265,
      "num_input_tokens_seen": 526384,
      "step": 45
    },
    {
      "epoch": 0.025753283543651816,
      "grad_norm": 1.853648349752417,
      "learning_rate": 2.9411764705882354e-05,
      "loss": 0.9157,
      "num_input_tokens_seen": 584856,
      "step": 50
    },
    {
      "epoch": 0.025753283543651816,
      "eval_loss": 0.9191630482673645,
      "eval_runtime": 36.6123,
      "eval_samples_per_second": 1.639,
      "eval_steps_per_second": 0.41,
      "num_input_tokens_seen": 584856,
      "step": 50
    },
    {
      "epoch": 0.028328611898016998,
      "grad_norm": 0.8294990584587586,
      "learning_rate": 3.235294117647059e-05,
      "loss": 0.9009,
      "num_input_tokens_seen": 643344,
      "step": 55
    },
    {
      "epoch": 0.03090394025238218,
      "grad_norm": 0.8278765532866457,
      "learning_rate": 3.529411764705883e-05,
      "loss": 0.9063,
      "num_input_tokens_seen": 701808,
      "step": 60
    },
    {
      "epoch": 0.03347926860674736,
      "grad_norm": 0.7285901101792476,
      "learning_rate": 3.8235294117647055e-05,
      "loss": 0.9031,
      "num_input_tokens_seen": 760304,
      "step": 65
    },
    {
      "epoch": 0.036054596961112545,
      "grad_norm": 0.5341783688819233,
      "learning_rate": 4.11764705882353e-05,
      "loss": 0.8991,
      "num_input_tokens_seen": 818760,
      "step": 70
    },
    {
      "epoch": 0.03862992531547772,
      "grad_norm": 0.46059313680988906,
      "learning_rate": 4.411764705882353e-05,
      "loss": 0.9055,
      "num_input_tokens_seen": 877256,
      "step": 75
    },
    {
      "epoch": 0.04120525366984291,
      "grad_norm": 0.8194379237293679,
      "learning_rate": 4.705882352941177e-05,
      "loss": 0.9092,
      "num_input_tokens_seen": 935752,
      "step": 80
    },
    {
      "epoch": 0.043780582024208085,
      "grad_norm": 0.6745093544830881,
      "learning_rate": 5e-05,
      "loss": 0.9069,
      "num_input_tokens_seen": 994216,
      "step": 85
    },
    {
      "epoch": 0.04635591037857327,
      "grad_norm": 0.2894672897884604,
      "learning_rate": 5.294117647058824e-05,
      "loss": 0.8924,
      "num_input_tokens_seen": 1052704,
      "step": 90
    },
    {
      "epoch": 0.04893123873293845,
      "grad_norm": 0.5108489024576455,
      "learning_rate": 5.588235294117647e-05,
      "loss": 0.9059,
      "num_input_tokens_seen": 1111176,
      "step": 95
    },
    {
      "epoch": 0.05150656708730363,
      "grad_norm": 0.40317180386305224,
      "learning_rate": 5.882352941176471e-05,
      "loss": 0.901,
      "num_input_tokens_seen": 1169664,
      "step": 100
    },
    {
      "epoch": 0.05150656708730363,
      "eval_loss": 0.9077914953231812,
      "eval_runtime": 16.8879,
      "eval_samples_per_second": 3.553,
      "eval_steps_per_second": 0.888,
      "num_input_tokens_seen": 1169664,
      "step": 100
    },
    {
      "epoch": 0.05408189544166881,
      "grad_norm": 0.412918917979438,
      "learning_rate": 6.176470588235295e-05,
      "loss": 0.9159,
      "num_input_tokens_seen": 1228112,
      "step": 105
    },
    {
      "epoch": 0.056657223796033995,
      "grad_norm": 0.34797408069968117,
      "learning_rate": 6.470588235294118e-05,
      "loss": 0.91,
      "num_input_tokens_seen": 1286608,
      "step": 110
    },
    {
      "epoch": 0.05923255215039917,
      "grad_norm": 0.27558494796967653,
      "learning_rate": 6.764705882352942e-05,
      "loss": 0.9047,
      "num_input_tokens_seen": 1345072,
      "step": 115
    },
    {
      "epoch": 0.06180788050476436,
      "grad_norm": 0.5422134023513459,
      "learning_rate": 7.058823529411765e-05,
      "loss": 0.9022,
      "num_input_tokens_seen": 1403544,
      "step": 120
    },
    {
      "epoch": 0.06438320885912954,
      "grad_norm": 0.4452796218739235,
      "learning_rate": 7.352941176470589e-05,
      "loss": 0.9081,
      "num_input_tokens_seen": 1462024,
      "step": 125
    },
    {
      "epoch": 0.06695853721349472,
      "grad_norm": 0.5632558160730559,
      "learning_rate": 7.647058823529411e-05,
      "loss": 0.8939,
      "num_input_tokens_seen": 1520528,
      "step": 130
    },
    {
      "epoch": 0.0695338655678599,
      "grad_norm": 0.3383115884436812,
      "learning_rate": 7.941176470588235e-05,
      "loss": 0.9029,
      "num_input_tokens_seen": 1579024,
      "step": 135
    },
    {
      "epoch": 0.07210919392222509,
      "grad_norm": 0.3506611095466577,
      "learning_rate": 8.23529411764706e-05,
      "loss": 0.9014,
      "num_input_tokens_seen": 1637504,
      "step": 140
    },
    {
      "epoch": 0.07468452227659027,
      "grad_norm": 0.6328034405712752,
      "learning_rate": 8.529411764705883e-05,
      "loss": 0.9053,
      "num_input_tokens_seen": 1696024,
      "step": 145
    },
    {
      "epoch": 0.07725985063095545,
      "grad_norm": 0.3511657661506363,
      "learning_rate": 8.823529411764706e-05,
      "loss": 0.9032,
      "num_input_tokens_seen": 1754512,
      "step": 150
    },
    {
      "epoch": 0.07725985063095545,
      "eval_loss": 0.8962129950523376,
      "eval_runtime": 17.0673,
      "eval_samples_per_second": 3.515,
      "eval_steps_per_second": 0.879,
      "num_input_tokens_seen": 1754512,
      "step": 150
    },
    {
      "epoch": 0.07983517898532062,
      "grad_norm": 0.4047681172482029,
      "learning_rate": 9.11764705882353e-05,
      "loss": 0.8985,
      "num_input_tokens_seen": 1812976,
      "step": 155
    },
    {
      "epoch": 0.08241050733968582,
      "grad_norm": 0.37729033726569733,
      "learning_rate": 9.411764705882353e-05,
      "loss": 0.8949,
      "num_input_tokens_seen": 1871464,
      "step": 160
    },
    {
      "epoch": 0.08498583569405099,
      "grad_norm": 0.4655744785034158,
      "learning_rate": 9.705882352941177e-05,
      "loss": 0.9069,
      "num_input_tokens_seen": 1929928,
      "step": 165
    },
    {
      "epoch": 0.08756116404841617,
      "grad_norm": 0.30643056878817176,
      "learning_rate": 0.0001,
      "loss": 0.9049,
      "num_input_tokens_seen": 1988432,
      "step": 170
    },
    {
      "epoch": 0.09013649240278135,
      "grad_norm": 0.39944696269496754,
      "learning_rate": 9.999940874631277e-05,
      "loss": 0.9026,
      "num_input_tokens_seen": 2046920,
      "step": 175
    },
    {
      "epoch": 0.09271182075714654,
      "grad_norm": 0.31301259106593154,
      "learning_rate": 9.999763499923432e-05,
      "loss": 0.8984,
      "num_input_tokens_seen": 2105392,
      "step": 180
    },
    {
      "epoch": 0.09528714911151172,
      "grad_norm": 0.4309753054454554,
      "learning_rate": 9.999467880071402e-05,
      "loss": 0.9057,
      "num_input_tokens_seen": 2163872,
      "step": 185
    },
    {
      "epoch": 0.0978624774658769,
      "grad_norm": 0.262930252305763,
      "learning_rate": 9.999054022066641e-05,
      "loss": 0.9078,
      "num_input_tokens_seen": 2222352,
      "step": 190
    },
    {
      "epoch": 0.10043780582024209,
      "grad_norm": 0.22073598270887426,
      "learning_rate": 9.998521935696953e-05,
      "loss": 0.9028,
      "num_input_tokens_seen": 2280800,
      "step": 195
    },
    {
      "epoch": 0.10301313417460727,
      "grad_norm": 0.23764668792524696,
      "learning_rate": 9.997871633546257e-05,
      "loss": 0.9053,
      "num_input_tokens_seen": 2339304,
      "step": 200
    },
    {
      "epoch": 0.10301313417460727,
      "eval_loss": 0.8982028961181641,
      "eval_runtime": 16.9118,
      "eval_samples_per_second": 3.548,
      "eval_steps_per_second": 0.887,
      "num_input_tokens_seen": 2339304,
      "step": 200
    },
    {
      "epoch": 0.10558846252897244,
      "grad_norm": 0.6222576114383499,
      "learning_rate": 9.997103130994296e-05,
      "loss": 0.9003,
      "num_input_tokens_seen": 2397808,
      "step": 205
    },
    {
      "epoch": 0.10816379088333762,
      "grad_norm": 0.2983149992592585,
      "learning_rate": 9.996216446216267e-05,
      "loss": 0.8969,
      "num_input_tokens_seen": 2456288,
      "step": 210
    },
    {
      "epoch": 0.11073911923770281,
      "grad_norm": 0.3505370510576513,
      "learning_rate": 9.995211600182397e-05,
      "loss": 0.9114,
      "num_input_tokens_seen": 2514784,
      "step": 215
    },
    {
      "epoch": 0.11331444759206799,
      "grad_norm": 0.3683806652106065,
      "learning_rate": 9.994088616657444e-05,
      "loss": 0.899,
      "num_input_tokens_seen": 2573240,
      "step": 220
    },
    {
      "epoch": 0.11588977594643317,
      "grad_norm": 0.21111769827155855,
      "learning_rate": 9.992847522200133e-05,
      "loss": 0.898,
      "num_input_tokens_seen": 2631672,
      "step": 225
    },
    {
      "epoch": 0.11846510430079835,
      "grad_norm": 0.3426987181783304,
      "learning_rate": 9.99148834616253e-05,
      "loss": 0.9006,
      "num_input_tokens_seen": 2690112,
      "step": 230
    },
    {
      "epoch": 0.12104043265516354,
      "grad_norm": 0.236983209071443,
      "learning_rate": 9.990011120689351e-05,
      "loss": 0.8973,
      "num_input_tokens_seen": 2748608,
      "step": 235
    },
    {
      "epoch": 0.12361576100952872,
      "grad_norm": 0.4575208248826409,
      "learning_rate": 9.988415880717194e-05,
      "loss": 0.8885,
      "num_input_tokens_seen": 2807080,
      "step": 240
    },
    {
      "epoch": 0.1261910893638939,
      "grad_norm": 0.5470317919414993,
      "learning_rate": 9.986702663973722e-05,
      "loss": 0.9066,
      "num_input_tokens_seen": 2865520,
      "step": 245
    },
    {
      "epoch": 0.12876641771825909,
      "grad_norm": 0.4992479706331095,
      "learning_rate": 9.98487151097676e-05,
      "loss": 0.9098,
      "num_input_tokens_seen": 2924016,
      "step": 250
    },
    {
      "epoch": 0.12876641771825909,
      "eval_loss": 0.8956434726715088,
      "eval_runtime": 17.4804,
      "eval_samples_per_second": 3.432,
      "eval_steps_per_second": 0.858,
      "num_input_tokens_seen": 2924016,
      "step": 250
    },
    {
      "epoch": 0.13134174607262425,
      "grad_norm": 0.3762164361984238,
      "learning_rate": 9.98292246503335e-05,
      "loss": 0.8987,
      "num_input_tokens_seen": 2982520,
      "step": 255
    },
    {
      "epoch": 0.13391707442698944,
      "grad_norm": 0.6447043002410199,
      "learning_rate": 9.980855572238714e-05,
      "loss": 0.9036,
      "num_input_tokens_seen": 3041008,
      "step": 260
    },
    {
      "epoch": 0.13649240278135463,
      "grad_norm": 0.5308092769971742,
      "learning_rate": 9.978670881475172e-05,
      "loss": 0.8961,
      "num_input_tokens_seen": 3099464,
      "step": 265
    },
    {
      "epoch": 0.1390677311357198,
      "grad_norm": 0.508333330469703,
      "learning_rate": 9.976368444410985e-05,
      "loss": 0.9012,
      "num_input_tokens_seen": 3157944,
      "step": 270
    },
    {
      "epoch": 0.141643059490085,
      "grad_norm": 0.6801788563719119,
      "learning_rate": 9.973948315499126e-05,
      "loss": 0.8985,
      "num_input_tokens_seen": 3216448,
      "step": 275
    },
    {
      "epoch": 0.14421838784445018,
      "grad_norm": 0.6933074703933572,
      "learning_rate": 9.971410551976002e-05,
      "loss": 0.9114,
      "num_input_tokens_seen": 3274928,
      "step": 280
    },
    {
      "epoch": 0.14679371619881534,
      "grad_norm": 0.21208820897494882,
      "learning_rate": 9.968755213860094e-05,
      "loss": 0.8886,
      "num_input_tokens_seen": 3333408,
      "step": 285
    },
    {
      "epoch": 0.14936904455318054,
      "grad_norm": 0.5791422669000065,
      "learning_rate": 9.96598236395054e-05,
      "loss": 0.8929,
      "num_input_tokens_seen": 3391896,
      "step": 290
    },
    {
      "epoch": 0.1519443729075457,
      "grad_norm": 0.3460368893191152,
      "learning_rate": 9.96309206782565e-05,
      "loss": 0.9091,
      "num_input_tokens_seen": 3450392,
      "step": 295
    },
    {
      "epoch": 0.1545197012619109,
      "grad_norm": 0.22425222135997747,
      "learning_rate": 9.960084393841355e-05,
      "loss": 0.8893,
      "num_input_tokens_seen": 3508888,
      "step": 300
    },
    {
      "epoch": 0.1545197012619109,
      "eval_loss": 0.8908902406692505,
      "eval_runtime": 16.9521,
      "eval_samples_per_second": 3.539,
      "eval_steps_per_second": 0.885,
      "num_input_tokens_seen": 3508888,
      "step": 300
    },
    {
      "epoch": 0.15709502961627608,
      "grad_norm": 0.23111596622064604,
      "learning_rate": 9.956959413129585e-05,
      "loss": 0.9056,
      "num_input_tokens_seen": 3567368,
      "step": 305
    },
    {
      "epoch": 0.15967035797064125,
      "grad_norm": 0.3918406894807393,
      "learning_rate": 9.953717199596598e-05,
      "loss": 0.8982,
      "num_input_tokens_seen": 3625848,
      "step": 310
    },
    {
      "epoch": 0.16224568632500644,
      "grad_norm": 0.22081666860189372,
      "learning_rate": 9.95035782992122e-05,
      "loss": 0.8968,
      "num_input_tokens_seen": 3684336,
      "step": 315
    },
    {
      "epoch": 0.16482101467937163,
      "grad_norm": 0.18024383676398176,
      "learning_rate": 9.94688138355304e-05,
      "loss": 0.8975,
      "num_input_tokens_seen": 3742800,
      "step": 320
    },
    {
      "epoch": 0.1673963430337368,
      "grad_norm": 0.3866897344302321,
      "learning_rate": 9.943287942710527e-05,
      "loss": 0.9061,
      "num_input_tokens_seen": 3801280,
      "step": 325
    },
    {
      "epoch": 0.16997167138810199,
      "grad_norm": 0.4804151381712559,
      "learning_rate": 9.939577592379088e-05,
      "loss": 0.8948,
      "num_input_tokens_seen": 3859792,
      "step": 330
    },
    {
      "epoch": 0.17254699974246718,
      "grad_norm": 0.35878231707669056,
      "learning_rate": 9.935750420309055e-05,
      "loss": 0.9063,
      "num_input_tokens_seen": 3918272,
      "step": 335
    },
    {
      "epoch": 0.17512232809683234,
      "grad_norm": 0.8713957774909928,
      "learning_rate": 9.931806517013612e-05,
      "loss": 0.8952,
      "num_input_tokens_seen": 3976760,
      "step": 340
    },
    {
      "epoch": 0.17769765645119753,
      "grad_norm": 0.6671526212854116,
      "learning_rate": 9.927745975766654e-05,
      "loss": 0.9136,
      "num_input_tokens_seen": 4035240,
      "step": 345
    },
    {
      "epoch": 0.1802729848055627,
      "grad_norm": 0.28702679234521244,
      "learning_rate": 9.923568892600578e-05,
      "loss": 0.9075,
      "num_input_tokens_seen": 4093688,
      "step": 350
    },
    {
      "epoch": 0.1802729848055627,
      "eval_loss": 0.89204341173172,
      "eval_runtime": 16.5819,
      "eval_samples_per_second": 3.618,
      "eval_steps_per_second": 0.905,
      "num_input_tokens_seen": 4093688,
      "step": 350
    },
    {
      "epoch": 0.1828483131599279,
      "grad_norm": 0.32233149132200706,
      "learning_rate": 9.91927536630402e-05,
      "loss": 0.8812,
      "num_input_tokens_seen": 4152160,
      "step": 355
    },
    {
      "epoch": 0.18542364151429308,
      "grad_norm": 0.5071871697326992,
      "learning_rate": 9.91486549841951e-05,
      "loss": 0.9109,
      "num_input_tokens_seen": 4210648,
      "step": 360
    },
    {
      "epoch": 0.18799896986865824,
      "grad_norm": 0.4532792519849944,
      "learning_rate": 9.91033939324107e-05,
      "loss": 0.9176,
      "num_input_tokens_seen": 4269136,
      "step": 365
    },
    {
      "epoch": 0.19057429822302344,
      "grad_norm": 0.5409761562534501,
      "learning_rate": 9.905697157811761e-05,
      "loss": 0.9077,
      "num_input_tokens_seen": 4327664,
      "step": 370
    },
    {
      "epoch": 0.19314962657738863,
      "grad_norm": 0.3432361562809093,
      "learning_rate": 9.900938901921131e-05,
      "loss": 0.893,
      "num_input_tokens_seen": 4386120,
      "step": 375
    },
    {
      "epoch": 0.1957249549317538,
      "grad_norm": 0.4756530294720616,
      "learning_rate": 9.896064738102635e-05,
      "loss": 0.9094,
      "num_input_tokens_seen": 4444560,
      "step": 380
    },
    {
      "epoch": 0.19830028328611898,
      "grad_norm": 0.424836974193983,
      "learning_rate": 9.891074781630966e-05,
      "loss": 0.9091,
      "num_input_tokens_seen": 4503016,
      "step": 385
    },
    {
      "epoch": 0.20087561164048418,
      "grad_norm": 0.31316926977469683,
      "learning_rate": 9.885969150519331e-05,
      "loss": 0.9033,
      "num_input_tokens_seen": 4561496,
      "step": 390
    },
    {
      "epoch": 0.20345093999484934,
      "grad_norm": 0.6108378682480797,
      "learning_rate": 9.88074796551666e-05,
      "loss": 0.8851,
      "num_input_tokens_seen": 4619944,
      "step": 395
    },
    {
      "epoch": 0.20602626834921453,
      "grad_norm": 0.38294566619219206,
      "learning_rate": 9.875411350104744e-05,
      "loss": 0.9004,
      "num_input_tokens_seen": 4678384,
      "step": 400
    },
    {
      "epoch": 0.20602626834921453,
      "eval_loss": 0.9086406826972961,
      "eval_runtime": 16.7827,
      "eval_samples_per_second": 3.575,
      "eval_steps_per_second": 0.894,
      "num_input_tokens_seen": 4678384,
      "step": 400
    },
    {
      "epoch": 0.2086015967035797,
      "grad_norm": 0.4283475401297436,
      "learning_rate": 9.86995943049533e-05,
      "loss": 0.8976,
      "num_input_tokens_seen": 4736904,
      "step": 405
    },
    {
      "epoch": 0.2111769250579449,
      "grad_norm": 0.40329738287583206,
      "learning_rate": 9.864392335627117e-05,
      "loss": 0.9134,
      "num_input_tokens_seen": 4795376,
      "step": 410
    },
    {
      "epoch": 0.21375225341231008,
      "grad_norm": 0.37890634863656475,
      "learning_rate": 9.858710197162721e-05,
      "loss": 0.8955,
      "num_input_tokens_seen": 4853880,
      "step": 415
    },
    {
      "epoch": 0.21632758176667524,
      "grad_norm": 0.32402245835420784,
      "learning_rate": 9.852913149485556e-05,
      "loss": 0.9014,
      "num_input_tokens_seen": 4912360,
      "step": 420
    },
    {
      "epoch": 0.21890291012104043,
      "grad_norm": 0.49572499508345125,
      "learning_rate": 9.847001329696653e-05,
      "loss": 0.9065,
      "num_input_tokens_seen": 4970872,
      "step": 425
    },
    {
      "epoch": 0.22147823847540563,
      "grad_norm": 0.11883567118448765,
      "learning_rate": 9.840974877611422e-05,
      "loss": 0.8952,
      "num_input_tokens_seen": 5029304,
      "step": 430
    },
    {
      "epoch": 0.2240535668297708,
      "grad_norm": 0.7105724703149633,
      "learning_rate": 9.834833935756344e-05,
      "loss": 0.9106,
      "num_input_tokens_seen": 5087800,
      "step": 435
    },
    {
      "epoch": 0.22662889518413598,
      "grad_norm": 0.708953365388227,
      "learning_rate": 9.828578649365601e-05,
      "loss": 0.8996,
      "num_input_tokens_seen": 5146312,
      "step": 440
    },
    {
      "epoch": 0.22920422353850115,
      "grad_norm": 0.4503080730364326,
      "learning_rate": 9.822209166377635e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 5204800,
      "step": 445
    },
    {
      "epoch": 0.23177955189286634,
      "grad_norm": 0.20754132336834788,
      "learning_rate": 9.815725637431662e-05,
      "loss": 0.9076,
      "num_input_tokens_seen": 5263304,
      "step": 450
    },
    {
      "epoch": 0.23177955189286634,
      "eval_loss": 0.8962157368659973,
      "eval_runtime": 17.2029,
      "eval_samples_per_second": 3.488,
      "eval_steps_per_second": 0.872,
      "num_input_tokens_seen": 5263304,
      "step": 450
    },
    {
      "epoch": 0.23435488024723153,
      "grad_norm": 0.5906403377099594,
      "learning_rate": 9.809128215864097e-05,
      "loss": 0.8942,
      "num_input_tokens_seen": 5321760,
      "step": 455
    },
    {
      "epoch": 0.2369302086015967,
      "grad_norm": 0.5706805631290568,
      "learning_rate": 9.802417057704931e-05,
      "loss": 0.9099,
      "num_input_tokens_seen": 5380224,
      "step": 460
    },
    {
      "epoch": 0.23950553695596188,
      "grad_norm": 0.164631948732384,
      "learning_rate": 9.795592321674045e-05,
      "loss": 0.8981,
      "num_input_tokens_seen": 5438704,
      "step": 465
    },
    {
      "epoch": 0.24208086531032708,
      "grad_norm": 0.32986780285522194,
      "learning_rate": 9.788654169177453e-05,
      "loss": 0.8952,
      "num_input_tokens_seen": 5497208,
      "step": 470
    },
    {
      "epoch": 0.24465619366469224,
      "grad_norm": 0.40551569446674784,
      "learning_rate": 9.781602764303487e-05,
      "loss": 0.8959,
      "num_input_tokens_seen": 5555704,
      "step": 475
    },
    {
      "epoch": 0.24723152201905743,
      "grad_norm": 0.20928586231326682,
      "learning_rate": 9.774438273818911e-05,
      "loss": 0.901,
      "num_input_tokens_seen": 5614160,
      "step": 480
    },
    {
      "epoch": 0.24980685037342262,
      "grad_norm": 0.34365307116824517,
      "learning_rate": 9.767160867164979e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 5672640,
      "step": 485
    },
    {
      "epoch": 0.2523821787277878,
      "grad_norm": 0.4212274243028996,
      "learning_rate": 9.759770716453436e-05,
      "loss": 0.9016,
      "num_input_tokens_seen": 5731072,
      "step": 490
    },
    {
      "epoch": 0.254957507082153,
      "grad_norm": 0.39823625576558597,
      "learning_rate": 9.752267996462434e-05,
      "loss": 0.9132,
      "num_input_tokens_seen": 5789544,
      "step": 495
    },
    {
      "epoch": 0.25753283543651817,
      "grad_norm": 0.24856324117583653,
      "learning_rate": 9.744652884632406e-05,
      "loss": 0.8962,
      "num_input_tokens_seen": 5848048,
      "step": 500
    },
    {
      "epoch": 0.25753283543651817,
      "eval_loss": 0.8987945914268494,
      "eval_runtime": 17.1622,
      "eval_samples_per_second": 3.496,
      "eval_steps_per_second": 0.874,
      "num_input_tokens_seen": 5848048,
      "step": 500
    },
    {
      "epoch": 0.26010816379088336,
      "grad_norm": 0.25461397268106634,
      "learning_rate": 9.736925561061871e-05,
      "loss": 0.8954,
      "num_input_tokens_seen": 5906512,
      "step": 505
    },
    {
      "epoch": 0.2626834921452485,
      "grad_norm": 0.38602603275675745,
      "learning_rate": 9.729086208503174e-05,
      "loss": 0.8927,
      "num_input_tokens_seen": 5965024,
      "step": 510
    },
    {
      "epoch": 0.2652588204996137,
      "grad_norm": 0.150082825225123,
      "learning_rate": 9.721135012358156e-05,
      "loss": 0.898,
      "num_input_tokens_seen": 6023496,
      "step": 515
    },
    {
      "epoch": 0.2678341488539789,
      "grad_norm": 0.26881662025899655,
      "learning_rate": 9.713072160673777e-05,
      "loss": 0.9016,
      "num_input_tokens_seen": 6082000,
      "step": 520
    },
    {
      "epoch": 0.2704094772083441,
      "grad_norm": 0.5039123575147229,
      "learning_rate": 9.704897844137673e-05,
      "loss": 0.8842,
      "num_input_tokens_seen": 6140480,
      "step": 525
    },
    {
      "epoch": 0.27298480556270927,
      "grad_norm": 0.27836945453098666,
      "learning_rate": 9.696612256073633e-05,
      "loss": 0.8921,
      "num_input_tokens_seen": 6198968,
      "step": 530
    },
    {
      "epoch": 0.2755601339170744,
      "grad_norm": 0.22936338891946384,
      "learning_rate": 9.688215592437039e-05,
      "loss": 0.8979,
      "num_input_tokens_seen": 6257464,
      "step": 535
    },
    {
      "epoch": 0.2781354622714396,
      "grad_norm": 0.396486857609105,
      "learning_rate": 9.679708051810221e-05,
      "loss": 0.8951,
      "num_input_tokens_seen": 6315944,
      "step": 540
    },
    {
      "epoch": 0.2807107906258048,
      "grad_norm": 0.4751226662261396,
      "learning_rate": 9.67108983539777e-05,
      "loss": 0.9149,
      "num_input_tokens_seen": 6374408,
      "step": 545
    },
    {
      "epoch": 0.28328611898017,
      "grad_norm": 0.26829103885131056,
      "learning_rate": 9.662361147021779e-05,
      "loss": 0.9013,
      "num_input_tokens_seen": 6432936,
      "step": 550
    },
    {
      "epoch": 0.28328611898017,
      "eval_loss": 0.9001271724700928,
      "eval_runtime": 16.9878,
      "eval_samples_per_second": 3.532,
      "eval_steps_per_second": 0.883,
      "num_input_tokens_seen": 6432936,
      "step": 550
    },
    {
      "epoch": 0.28586144733453517,
      "grad_norm": 0.5334970266367584,
      "learning_rate": 9.653522193117013e-05,
      "loss": 0.8981,
      "num_input_tokens_seen": 6491400,
      "step": 555
    },
    {
      "epoch": 0.28843677568890036,
      "grad_norm": 0.33261202813259866,
      "learning_rate": 9.644573182726035e-05,
      "loss": 0.9041,
      "num_input_tokens_seen": 6549872,
      "step": 560
    },
    {
      "epoch": 0.2910121040432655,
      "grad_norm": 0.19122862132727417,
      "learning_rate": 9.63551432749426e-05,
      "loss": 0.9024,
      "num_input_tokens_seen": 6608296,
      "step": 565
    },
    {
      "epoch": 0.2935874323976307,
      "grad_norm": 0.27778009425329764,
      "learning_rate": 9.626345841664953e-05,
      "loss": 0.9002,
      "num_input_tokens_seen": 6666768,
      "step": 570
    },
    {
      "epoch": 0.2961627607519959,
      "grad_norm": 0.3065314332046026,
      "learning_rate": 9.617067942074153e-05,
      "loss": 0.9035,
      "num_input_tokens_seen": 6725248,
      "step": 575
    },
    {
      "epoch": 0.29873808910636107,
      "grad_norm": 0.24431496415058412,
      "learning_rate": 9.607680848145558e-05,
      "loss": 0.9019,
      "num_input_tokens_seen": 6783680,
      "step": 580
    },
    {
      "epoch": 0.30131341746072626,
      "grad_norm": 0.27088193021301504,
      "learning_rate": 9.598184781885318e-05,
      "loss": 0.9001,
      "num_input_tokens_seen": 6842144,
      "step": 585
    },
    {
      "epoch": 0.3038887458150914,
      "grad_norm": 0.33893098113605125,
      "learning_rate": 9.588579967876806e-05,
      "loss": 0.8961,
      "num_input_tokens_seen": 6900656,
      "step": 590
    },
    {
      "epoch": 0.3064640741694566,
      "grad_norm": 0.3038921833221806,
      "learning_rate": 9.578866633275288e-05,
      "loss": 0.9,
      "num_input_tokens_seen": 6959128,
      "step": 595
    },
    {
      "epoch": 0.3090394025238218,
      "grad_norm": 0.48929637235055645,
      "learning_rate": 9.569045007802559e-05,
      "loss": 0.9046,
      "num_input_tokens_seen": 7017576,
      "step": 600
    },
    {
      "epoch": 0.3090394025238218,
      "eval_loss": 0.9053278565406799,
      "eval_runtime": 17.1218,
      "eval_samples_per_second": 3.504,
      "eval_steps_per_second": 0.876,
      "num_input_tokens_seen": 7017576,
      "step": 600
    },
    {
      "epoch": 0.311614730878187,
      "grad_norm": 0.3545950949033049,
      "learning_rate": 9.55911532374151e-05,
      "loss": 0.9019,
      "num_input_tokens_seen": 7076032,
      "step": 605
    },
    {
      "epoch": 0.31419005923255217,
      "grad_norm": 0.2355627006333952,
      "learning_rate": 9.549077815930636e-05,
      "loss": 0.8956,
      "num_input_tokens_seen": 7134536,
      "step": 610
    },
    {
      "epoch": 0.31676538758691736,
      "grad_norm": 0.17552483625655946,
      "learning_rate": 9.538932721758474e-05,
      "loss": 0.898,
      "num_input_tokens_seen": 7193032,
      "step": 615
    },
    {
      "epoch": 0.3193407159412825,
      "grad_norm": 0.1749010635522076,
      "learning_rate": 9.528680281157999e-05,
      "loss": 0.8991,
      "num_input_tokens_seen": 7251568,
      "step": 620
    },
    {
      "epoch": 0.3219160442956477,
      "grad_norm": 0.19885182954224315,
      "learning_rate": 9.518320736600943e-05,
      "loss": 0.8961,
      "num_input_tokens_seen": 7310072,
      "step": 625
    },
    {
      "epoch": 0.3244913726500129,
      "grad_norm": 0.4778756508206831,
      "learning_rate": 9.507854333092063e-05,
      "loss": 0.8994,
      "num_input_tokens_seen": 7368560,
      "step": 630
    },
    {
      "epoch": 0.32706670100437807,
      "grad_norm": 0.4123272743887767,
      "learning_rate": 9.497281318163346e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 7427040,
      "step": 635
    },
    {
      "epoch": 0.32964202935874326,
      "grad_norm": 0.34409942667705734,
      "learning_rate": 9.486601941868154e-05,
      "loss": 0.9087,
      "num_input_tokens_seen": 7485552,
      "step": 640
    },
    {
      "epoch": 0.3322173577131084,
      "grad_norm": 0.43327107411223276,
      "learning_rate": 9.475816456775313e-05,
      "loss": 0.8924,
      "num_input_tokens_seen": 7544040,
      "step": 645
    },
    {
      "epoch": 0.3347926860674736,
      "grad_norm": 0.6643023904352003,
      "learning_rate": 9.464925117963133e-05,
      "loss": 0.904,
      "num_input_tokens_seen": 7602512,
      "step": 650
    },
    {
      "epoch": 0.3347926860674736,
      "eval_loss": 0.90328449010849,
      "eval_runtime": 16.1444,
      "eval_samples_per_second": 3.716,
      "eval_steps_per_second": 0.929,
      "num_input_tokens_seen": 7602512,
      "step": 650
    },
    {
      "epoch": 0.3373680144218388,
      "grad_norm": 0.620349194493935,
      "learning_rate": 9.453928183013385e-05,
      "loss": 0.8929,
      "num_input_tokens_seen": 7660968,
      "step": 655
    },
    {
      "epoch": 0.33994334277620397,
      "grad_norm": 0.18611846349930314,
      "learning_rate": 9.442825912005202e-05,
      "loss": 0.9078,
      "num_input_tokens_seen": 7719448,
      "step": 660
    },
    {
      "epoch": 0.34251867113056916,
      "grad_norm": 0.4448289413172567,
      "learning_rate": 9.431618567508933e-05,
      "loss": 0.8963,
      "num_input_tokens_seen": 7777928,
      "step": 665
    },
    {
      "epoch": 0.34509399948493436,
      "grad_norm": 0.6187189362250411,
      "learning_rate": 9.420306414579925e-05,
      "loss": 0.9134,
      "num_input_tokens_seen": 7836424,
      "step": 670
    },
    {
      "epoch": 0.3476693278392995,
      "grad_norm": 0.35247743418537675,
      "learning_rate": 9.408889720752266e-05,
      "loss": 0.8984,
      "num_input_tokens_seen": 7894904,
      "step": 675
    },
    {
      "epoch": 0.3502446561936647,
      "grad_norm": 0.20652916455346712,
      "learning_rate": 9.397368756032445e-05,
      "loss": 0.8997,
      "num_input_tokens_seen": 7953432,
      "step": 680
    },
    {
      "epoch": 0.3528199845480299,
      "grad_norm": 0.4289996063998063,
      "learning_rate": 9.385743792892982e-05,
      "loss": 0.8926,
      "num_input_tokens_seen": 8011888,
      "step": 685
    },
    {
      "epoch": 0.35539531290239507,
      "grad_norm": 0.13764054506536547,
      "learning_rate": 9.374015106265968e-05,
      "loss": 0.9008,
      "num_input_tokens_seen": 8070344,
      "step": 690
    },
    {
      "epoch": 0.35797064125676026,
      "grad_norm": 0.22142459689499855,
      "learning_rate": 9.362182973536569e-05,
      "loss": 0.8986,
      "num_input_tokens_seen": 8128816,
      "step": 695
    },
    {
      "epoch": 0.3605459696111254,
      "grad_norm": 0.3234539650829873,
      "learning_rate": 9.35024767453647e-05,
      "loss": 0.8972,
      "num_input_tokens_seen": 8187320,
      "step": 700
    },
    {
      "epoch": 0.3605459696111254,
      "eval_loss": 0.9028835892677307,
      "eval_runtime": 16.1635,
      "eval_samples_per_second": 3.712,
      "eval_steps_per_second": 0.928,
      "num_input_tokens_seen": 8187320,
      "step": 700
    },
    {
      "epoch": 0.3631212979654906,
      "grad_norm": 0.3215674690491891,
      "learning_rate": 9.338209491537257e-05,
      "loss": 0.8998,
      "num_input_tokens_seen": 8245776,
      "step": 705
    },
    {
      "epoch": 0.3656966263198558,
      "grad_norm": 0.36428692362396536,
      "learning_rate": 9.326068709243727e-05,
      "loss": 0.8999,
      "num_input_tokens_seen": 8304280,
      "step": 710
    },
    {
      "epoch": 0.36827195467422097,
      "grad_norm": 0.280459809393624,
      "learning_rate": 9.313825614787177e-05,
      "loss": 0.8983,
      "num_input_tokens_seen": 8362728,
      "step": 715
    },
    {
      "epoch": 0.37084728302858616,
      "grad_norm": 0.1819339731162554,
      "learning_rate": 9.301480497718593e-05,
      "loss": 0.892,
      "num_input_tokens_seen": 8421224,
      "step": 720
    },
    {
      "epoch": 0.37342261138295135,
      "grad_norm": 0.23784840563699303,
      "learning_rate": 9.289033650001817e-05,
      "loss": 0.9034,
      "num_input_tokens_seen": 8479720,
      "step": 725
    },
    {
      "epoch": 0.3759979397373165,
      "grad_norm": 0.24070744588741375,
      "learning_rate": 9.276485366006634e-05,
      "loss": 0.895,
      "num_input_tokens_seen": 8538192,
      "step": 730
    },
    {
      "epoch": 0.3785732680916817,
      "grad_norm": 0.24846723619231478,
      "learning_rate": 9.263835942501807e-05,
      "loss": 0.8973,
      "num_input_tokens_seen": 8596664,
      "step": 735
    },
    {
      "epoch": 0.3811485964460469,
      "grad_norm": 0.2601614440419362,
      "learning_rate": 9.251085678648072e-05,
      "loss": 0.8972,
      "num_input_tokens_seen": 8655128,
      "step": 740
    },
    {
      "epoch": 0.38372392480041206,
      "grad_norm": 0.30194733839751087,
      "learning_rate": 9.238234875991046e-05,
      "loss": 0.8987,
      "num_input_tokens_seen": 8713624,
      "step": 745
    },
    {
      "epoch": 0.38629925315477726,
      "grad_norm": 0.3015609177439829,
      "learning_rate": 9.225283838454111e-05,
      "loss": 0.9005,
      "num_input_tokens_seen": 8772104,
      "step": 750
    },
    {
      "epoch": 0.38629925315477726,
      "eval_loss": 0.8981761336326599,
      "eval_runtime": 16.0177,
      "eval_samples_per_second": 3.746,
      "eval_steps_per_second": 0.936,
      "num_input_tokens_seen": 8772104,
      "step": 750
    },
    {
      "epoch": 0.3888745815091424,
      "grad_norm": 0.44991480631292463,
      "learning_rate": 9.21223287233121e-05,
      "loss": 0.8973,
      "num_input_tokens_seen": 8830568,
      "step": 755
    },
    {
      "epoch": 0.3914499098635076,
      "grad_norm": 0.22570310903133853,
      "learning_rate": 9.199082286279622e-05,
      "loss": 0.8974,
      "num_input_tokens_seen": 8889072,
      "step": 760
    },
    {
      "epoch": 0.3940252382178728,
      "grad_norm": 0.22090133233732026,
      "learning_rate": 9.185832391312644e-05,
      "loss": 0.8985,
      "num_input_tokens_seen": 8947568,
      "step": 765
    },
    {
      "epoch": 0.39660056657223797,
      "grad_norm": 0.23738058530347297,
      "learning_rate": 9.172483500792244e-05,
      "loss": 0.8935,
      "num_input_tokens_seen": 9006056,
      "step": 770
    },
    {
      "epoch": 0.39917589492660316,
      "grad_norm": 0.41232659301572594,
      "learning_rate": 9.159035930421658e-05,
      "loss": 0.8985,
      "num_input_tokens_seen": 9064592,
      "step": 775
    },
    {
      "epoch": 0.40175122328096835,
      "grad_norm": 0.2004855543001356,
      "learning_rate": 9.145489998237902e-05,
      "loss": 0.9105,
      "num_input_tokens_seen": 9123096,
      "step": 780
    },
    {
      "epoch": 0.4043265516353335,
      "grad_norm": 0.16209487510237375,
      "learning_rate": 9.131846024604274e-05,
      "loss": 0.8925,
      "num_input_tokens_seen": 9181576,
      "step": 785
    },
    {
      "epoch": 0.4069018799896987,
      "grad_norm": 0.24319930530142153,
      "learning_rate": 9.11810433220276e-05,
      "loss": 0.8955,
      "num_input_tokens_seen": 9240048,
      "step": 790
    },
    {
      "epoch": 0.40947720834406387,
      "grad_norm": 0.24311562892750557,
      "learning_rate": 9.104265246026415e-05,
      "loss": 0.8986,
      "num_input_tokens_seen": 9298528,
      "step": 795
    },
    {
      "epoch": 0.41205253669842906,
      "grad_norm": 0.2891177185942039,
      "learning_rate": 9.090329093371666e-05,
      "loss": 0.8881,
      "num_input_tokens_seen": 9357016,
      "step": 800
    },
    {
      "epoch": 0.41205253669842906,
      "eval_loss": 0.8973079919815063,
      "eval_runtime": 16.1396,
      "eval_samples_per_second": 3.718,
      "eval_steps_per_second": 0.929,
      "num_input_tokens_seen": 9357016,
      "step": 800
    },
    {
      "epoch": 0.41462786505279425,
      "grad_norm": 0.4728970278357675,
      "learning_rate": 9.076296203830579e-05,
      "loss": 0.8798,
      "num_input_tokens_seen": 9415480,
      "step": 805
    },
    {
      "epoch": 0.4172031934071594,
      "grad_norm": 0.2420351489416807,
      "learning_rate": 9.062166909283062e-05,
      "loss": 0.9104,
      "num_input_tokens_seen": 9473928,
      "step": 810
    },
    {
      "epoch": 0.4197785217615246,
      "grad_norm": 0.2262623911682871,
      "learning_rate": 9.047941543889014e-05,
      "loss": 0.9007,
      "num_input_tokens_seen": 9532408,
      "step": 815
    },
    {
      "epoch": 0.4223538501158898,
      "grad_norm": 0.18258980329217392,
      "learning_rate": 9.033620444080428e-05,
      "loss": 0.8974,
      "num_input_tokens_seen": 9590920,
      "step": 820
    },
    {
      "epoch": 0.42492917847025496,
      "grad_norm": 0.2898762949979446,
      "learning_rate": 9.019203948553422e-05,
      "loss": 0.8992,
      "num_input_tokens_seen": 9649400,
      "step": 825
    },
    {
      "epoch": 0.42750450682462016,
      "grad_norm": 0.3884592601874919,
      "learning_rate": 9.004692398260244e-05,
      "loss": 0.8991,
      "num_input_tokens_seen": 9707888,
      "step": 830
    },
    {
      "epoch": 0.43007983517898535,
      "grad_norm": 0.24055719869667014,
      "learning_rate": 8.9900861364012e-05,
      "loss": 0.8964,
      "num_input_tokens_seen": 9766384,
      "step": 835
    },
    {
      "epoch": 0.4326551635333505,
      "grad_norm": 0.4482774361285702,
      "learning_rate": 8.975385508416532e-05,
      "loss": 0.8723,
      "num_input_tokens_seen": 9824896,
      "step": 840
    },
    {
      "epoch": 0.4352304918877157,
      "grad_norm": 0.4612030185875055,
      "learning_rate": 8.960590861978265e-05,
      "loss": 0.874,
      "num_input_tokens_seen": 9883408,
      "step": 845
    },
    {
      "epoch": 0.43780582024208087,
      "grad_norm": 0.44197834194509644,
      "learning_rate": 8.945702546981969e-05,
      "loss": 0.9035,
      "num_input_tokens_seen": 9941896,
      "step": 850
    },
    {
      "epoch": 0.43780582024208087,
      "eval_loss": 0.8779178261756897,
      "eval_runtime": 16.159,
      "eval_samples_per_second": 3.713,
      "eval_steps_per_second": 0.928,
      "num_input_tokens_seen": 9941896,
      "step": 850
    }
  ],
  "logging_steps": 5,
  "max_steps": 3400,
  "num_input_tokens_seen": 9941896,
  "num_train_epochs": 2,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 558240938328064.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}