{
  "dataset_name": "ToT-Math-V1",
  "group_index": 3,
  "prompt_group_file": "/hkfs/work/workspace/scratch/tum_fmp0582-dndworkspace/自己训练lora/train_lora/prompt_groups/ToT-Math-V1/group_03.json",
  "output_dir": "/hkfs/work/workspace/scratch/tum_fmp0582-dndworkspace/自己训练lora/train_lora/outputs/ToT-Math-V1/group_03",
  "checkpoint_root": "/hkfs/work/workspace/scratch/tum_fmp0582-dndworkspace/自己训练lora/train_lora/outputs/ToT-Math-V1/group_03/checkpoints",
  "generated_at": "2025-11-06T12:10:08Z",
  "train_loss": 0.03864684621699982,
  "metrics": {
    "train_runtime": 30821.8954,
    "train_samples_per_second": 1.869,
    "train_steps_per_second": 0.058,
    "total_flos": 1.5308141101056e+18,
    "train_loss": 0.03864684621699982,
    "epoch": 450.0
  },
  "trainer_state": [
    {
      "loss": 1.3545,
      "grad_norm": 3.3054826259613037,
      "learning_rate": 0.0,
      "epoch": 0.32,
      "step": 1
    },
    {
      "loss": 1.2974,
      "grad_norm": 2.2654924392700195,
      "learning_rate": 7.4074074074074075e-06,
      "epoch": 1.32,
      "step": 5
    },
    {
      "loss": 1.1861,
      "grad_norm": 1.0169233083724976,
      "learning_rate": 1.6666666666666667e-05,
      "epoch": 2.64,
      "step": 10
    },
    {
      "loss": 1.1839,
      "grad_norm": 0.5837942361831665,
      "learning_rate": 2.5925925925925925e-05,
      "epoch": 3.96,
      "step": 15
    },
    {
      "loss": 1.0151,
      "grad_norm": 0.6558277010917664,
      "learning_rate": 3.518518518518519e-05,
      "epoch": 5.0,
      "step": 20
    },
    {
      "loss": 0.9023,
      "grad_norm": 0.4020928740501404,
      "learning_rate": 4.4444444444444447e-05,
      "epoch": 6.32,
      "step": 25
    },
    {
      "loss": 0.8593,
      "grad_norm": 0.32427164912223816,
      "learning_rate": 5.370370370370371e-05,
      "epoch": 7.64,
      "step": 30
    },
    {
      "loss": 0.7702,
      "grad_norm": 0.3604109287261963,
      "learning_rate": 6.296296296296296e-05,
      "epoch": 8.96,
      "step": 35
    },
    {
      "loss": 0.7169,
      "grad_norm": 0.7918370366096497,
      "learning_rate": 7.222222222222222e-05,
      "epoch": 10.0,
      "step": 40
    },
    {
      "loss": 0.7023,
      "grad_norm": 0.3114716112613678,
      "learning_rate": 8.148148148148148e-05,
      "epoch": 11.32,
      "step": 45
    },
    {
      "loss": 0.6026,
      "grad_norm": 0.34100472927093506,
      "learning_rate": 9.074074074074075e-05,
      "epoch": 12.64,
      "step": 50
    },
    {
      "loss": 0.4889,
      "grad_norm": 0.3940233290195465,
      "learning_rate": 0.0001,
      "epoch": 13.96,
      "step": 55
    },
    {
      "loss": 0.4029,
      "grad_norm": 1.0256505012512207,
      "learning_rate": 9.971363115693013e-05,
      "epoch": 15.0,
      "step": 60
    },
    {
      "loss": 0.3254,
      "grad_norm": 0.4910169839859009,
      "learning_rate": 9.942726231386026e-05,
      "epoch": 16.32,
      "step": 65
    },
    {
      "loss": 0.2307,
      "grad_norm": 0.514342725276947,
      "learning_rate": 9.914089347079038e-05,
      "epoch": 17.64,
      "step": 70
    },
    {
      "loss": 0.1762,
      "grad_norm": 0.4822510778903961,
      "learning_rate": 9.885452462772051e-05,
      "epoch": 18.96,
      "step": 75
    },
    {
      "loss": 0.1007,
      "grad_norm": 1.3144968748092651,
      "learning_rate": 9.856815578465064e-05,
      "epoch": 20.0,
      "step": 80
    },
    {
      "loss": 0.0721,
      "grad_norm": 0.481692910194397,
      "learning_rate": 9.828178694158075e-05,
      "epoch": 21.32,
      "step": 85
    },
    {
      "loss": 0.0512,
      "grad_norm": 0.4445790648460388,
      "learning_rate": 9.799541809851088e-05,
      "epoch": 22.64,
      "step": 90
    },
    {
      "loss": 0.0369,
      "grad_norm": 0.42874398827552795,
      "learning_rate": 9.7709049255441e-05,
      "epoch": 23.96,
      "step": 95
    },
    {
      "loss": 0.0424,
      "grad_norm": 1.2392895221710205,
      "learning_rate": 9.742268041237114e-05,
      "epoch": 25.0,
      "step": 100
    },
    {
      "loss": 0.0259,
      "grad_norm": 0.27867934107780457,
      "learning_rate": 9.713631156930127e-05,
      "epoch": 26.32,
      "step": 105
    },
    {
      "loss": 0.0235,
      "grad_norm": 0.25555816292762756,
      "learning_rate": 9.68499427262314e-05,
      "epoch": 27.64,
      "step": 110
    },
    {
      "loss": 0.024,
      "grad_norm": 0.265229731798172,
      "learning_rate": 9.656357388316152e-05,
      "epoch": 28.96,
      "step": 115
    },
    {
      "loss": 0.0269,
      "grad_norm": 0.5608186721801758,
      "learning_rate": 9.627720504009165e-05,
      "epoch": 30.0,
      "step": 120
    },
    {
      "loss": 0.017,
      "grad_norm": 0.315533846616745,
      "learning_rate": 9.599083619702178e-05,
      "epoch": 31.32,
      "step": 125
    },
    {
      "loss": 0.0157,
      "grad_norm": 0.18174390494823456,
      "learning_rate": 9.57044673539519e-05,
      "epoch": 32.64,
      "step": 130
    },
    {
      "loss": 0.0151,
      "grad_norm": 0.23531611263751984,
      "learning_rate": 9.541809851088203e-05,
      "epoch": 33.96,
      "step": 135
    },
    {
      "loss": 0.0161,
      "grad_norm": 1.0181336402893066,
      "learning_rate": 9.513172966781214e-05,
      "epoch": 35.0,
      "step": 140
    },
    {
      "loss": 0.0145,
      "grad_norm": 0.23052875697612762,
      "learning_rate": 9.484536082474227e-05,
      "epoch": 36.32,
      "step": 145
    },
    {
      "loss": 0.0153,
      "grad_norm": 0.24433749914169312,
      "learning_rate": 9.45589919816724e-05,
      "epoch": 37.64,
      "step": 150
    },
    {
      "loss": 0.0124,
      "grad_norm": 0.20692484080791473,
      "learning_rate": 9.427262313860252e-05,
      "epoch": 38.96,
      "step": 155
    },
    {
      "loss": 0.0134,
      "grad_norm": 0.5084072947502136,
      "learning_rate": 9.398625429553265e-05,
      "epoch": 40.0,
      "step": 160
    },
    {
      "loss": 0.0136,
      "grad_norm": 0.0886864885687828,
      "learning_rate": 9.369988545246277e-05,
      "epoch": 41.32,
      "step": 165
    },
    {
      "loss": 0.0121,
      "grad_norm": 0.08282119780778885,
      "learning_rate": 9.34135166093929e-05,
      "epoch": 42.64,
      "step": 170
    },
    {
      "loss": 0.0111,
      "grad_norm": 0.09241609275341034,
      "learning_rate": 9.312714776632303e-05,
      "epoch": 43.96,
      "step": 175
    },
    {
      "loss": 0.01,
      "grad_norm": 0.08246699720621109,
      "learning_rate": 9.284077892325315e-05,
      "epoch": 45.0,
      "step": 180
    },
    {
      "loss": 0.0096,
      "grad_norm": 0.03985537588596344,
      "learning_rate": 9.255441008018328e-05,
      "epoch": 46.32,
      "step": 185
    },
    {
      "loss": 0.0094,
      "grad_norm": 0.059935443103313446,
      "learning_rate": 9.22680412371134e-05,
      "epoch": 47.64,
      "step": 190
    },
    {
      "loss": 0.0091,
      "grad_norm": 0.0341961644589901,
      "learning_rate": 9.198167239404353e-05,
      "epoch": 48.96,
      "step": 195
    },
    {
      "loss": 0.0089,
      "grad_norm": 0.07446911931037903,
      "learning_rate": 9.169530355097366e-05,
      "epoch": 50.0,
      "step": 200
    },
    {
      "loss": 0.0092,
      "grad_norm": 0.03150145336985588,
      "learning_rate": 9.140893470790379e-05,
      "epoch": 51.32,
      "step": 205
    },
    {
      "loss": 0.0087,
      "grad_norm": 0.030524440109729767,
      "learning_rate": 9.112256586483391e-05,
      "epoch": 52.64,
      "step": 210
    },
    {
      "loss": 0.011,
      "grad_norm": 0.09918652474880219,
      "learning_rate": 9.083619702176404e-05,
      "epoch": 53.96,
      "step": 215
    },
    {
      "loss": 0.0098,
      "grad_norm": 0.09719116240739822,
      "learning_rate": 9.054982817869416e-05,
      "epoch": 55.0,
      "step": 220
    },
    {
      "loss": 0.0091,
      "grad_norm": 0.05922112613916397,
      "learning_rate": 9.026345933562429e-05,
      "epoch": 56.32,
      "step": 225
    },
    {
      "loss": 0.0095,
      "grad_norm": 0.058665432035923004,
      "learning_rate": 8.997709049255442e-05,
      "epoch": 57.64,
      "step": 230
    },
    {
      "loss": 0.0087,
      "grad_norm": 0.029366934671998024,
      "learning_rate": 8.969072164948454e-05,
      "epoch": 58.96,
      "step": 235
    },
    {
      "loss": 0.0094,
      "grad_norm": 0.08083733916282654,
      "learning_rate": 8.940435280641467e-05,
      "epoch": 60.0,
      "step": 240
    },
    {
      "loss": 0.0086,
      "grad_norm": 0.029681477695703506,
      "learning_rate": 8.91179839633448e-05,
      "epoch": 61.32,
      "step": 245
    },
    {
      "loss": 0.009,
      "grad_norm": 0.0322534516453743,
      "learning_rate": 8.883161512027491e-05,
      "epoch": 62.64,
      "step": 250
    },
    {
      "loss": 0.0089,
      "grad_norm": 0.030840858817100525,
      "learning_rate": 8.854524627720504e-05,
      "epoch": 63.96,
      "step": 255
    },
    {
      "loss": 0.0091,
      "grad_norm": 0.07301075756549835,
      "learning_rate": 8.825887743413516e-05,
      "epoch": 65.0,
      "step": 260
    },
    {
      "loss": 0.0084,
      "grad_norm": 0.028555437922477722,
      "learning_rate": 8.797250859106529e-05,
      "epoch": 66.32,
      "step": 265
    },
    {
      "loss": 0.0089,
      "grad_norm": 0.031545497477054596,
      "learning_rate": 8.768613974799542e-05,
      "epoch": 67.64,
      "step": 270
    },
    {
      "loss": 0.0084,
      "grad_norm": 0.025588998571038246,
      "learning_rate": 8.739977090492554e-05,
      "epoch": 68.96,
      "step": 275
    },
    {
      "loss": 0.0085,
      "grad_norm": 0.1724974364042282,
      "learning_rate": 8.711340206185567e-05,
      "epoch": 70.0,
      "step": 280
    },
    {
      "loss": 0.0085,
      "grad_norm": 0.04501425102353096,
      "learning_rate": 8.682703321878581e-05,
      "epoch": 71.32,
      "step": 285
    },
    {
      "loss": 0.0082,
      "grad_norm": 0.029004130512475967,
      "learning_rate": 8.654066437571594e-05,
      "epoch": 72.64,
      "step": 290
    },
    {
      "loss": 0.0084,
      "grad_norm": 0.03226853162050247,
      "learning_rate": 8.625429553264606e-05,
      "epoch": 73.96,
      "step": 295
    },
    {
      "loss": 0.0082,
      "grad_norm": 0.08804097026586533,
      "learning_rate": 8.596792668957619e-05,
      "epoch": 75.0,
      "step": 300
    },
    {
      "loss": 0.008,
      "grad_norm": 0.021941116079688072,
      "learning_rate": 8.56815578465063e-05,
      "epoch": 76.32,
      "step": 305
    },
    {
      "loss": 0.0085,
      "grad_norm": 0.026565641164779663,
      "learning_rate": 8.539518900343643e-05,
      "epoch": 77.64,
      "step": 310
    },
    {
      "loss": 0.0085,
      "grad_norm": 0.02847815677523613,
      "learning_rate": 8.510882016036655e-05,
      "epoch": 78.96,
      "step": 315
    },
    {
      "loss": 0.0083,
      "grad_norm": 0.08784179389476776,
      "learning_rate": 8.482245131729668e-05,
      "epoch": 80.0,
      "step": 320
    },
    {
      "loss": 0.0082,
      "grad_norm": 0.025265933945775032,
      "learning_rate": 8.453608247422681e-05,
      "epoch": 81.32,
      "step": 325
    },
    {
      "loss": 0.0081,
      "grad_norm": 0.025272194296121597,
      "learning_rate": 8.424971363115693e-05,
      "epoch": 82.64,
      "step": 330
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.028765657916665077,
      "learning_rate": 8.396334478808706e-05,
      "epoch": 83.96,
      "step": 335
    },
    {
      "loss": 0.0081,
      "grad_norm": 0.07430175691843033,
      "learning_rate": 8.367697594501719e-05,
      "epoch": 85.0,
      "step": 340
    },
    {
      "loss": 0.008,
      "grad_norm": 0.026086682453751564,
      "learning_rate": 8.339060710194731e-05,
      "epoch": 86.32,
      "step": 345
    },
    {
      "loss": 0.0081,
      "grad_norm": 0.0352594219148159,
      "learning_rate": 8.310423825887744e-05,
      "epoch": 87.64,
      "step": 350
    },
    {
      "loss": 0.0082,
      "grad_norm": 0.022339310497045517,
      "learning_rate": 8.281786941580757e-05,
      "epoch": 88.96,
      "step": 355
    },
    {
      "loss": 0.0081,
      "grad_norm": 0.05659110099077225,
      "learning_rate": 8.253150057273768e-05,
      "epoch": 90.0,
      "step": 360
    },
    {
      "loss": 0.0081,
      "grad_norm": 0.02188769169151783,
      "learning_rate": 8.224513172966782e-05,
      "epoch": 91.32,
      "step": 365
    },
    {
      "loss": 0.008,
      "grad_norm": 0.020206712186336517,
      "learning_rate": 8.195876288659795e-05,
      "epoch": 92.64,
      "step": 370
    },
    {
      "loss": 0.0083,
      "grad_norm": 0.028029531240463257,
      "learning_rate": 8.167239404352807e-05,
      "epoch": 93.96,
      "step": 375
    },
    {
      "loss": 0.008,
      "grad_norm": 0.08095509558916092,
      "learning_rate": 8.13860252004582e-05,
      "epoch": 95.0,
      "step": 380
    },
    {
      "loss": 0.0082,
      "grad_norm": 0.022908154875040054,
      "learning_rate": 8.109965635738833e-05,
      "epoch": 96.32,
      "step": 385
    },
    {
      "loss": 0.0087,
      "grad_norm": 0.026914075016975403,
      "learning_rate": 8.081328751431845e-05,
      "epoch": 97.64,
      "step": 390
    },
    {
      "loss": 0.0079,
      "grad_norm": 0.0204644575715065,
      "learning_rate": 8.052691867124858e-05,
      "epoch": 98.96,
      "step": 395
    },
    {
      "loss": 0.0089,
      "grad_norm": 0.09402702003717422,
      "learning_rate": 8.02405498281787e-05,
      "epoch": 100.0,
      "step": 400
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.018987316638231277,
      "learning_rate": 7.995418098510883e-05,
      "epoch": 101.32,
      "step": 405
    },
    {
      "loss": 0.0082,
      "grad_norm": 0.025304267182946205,
      "learning_rate": 7.966781214203894e-05,
      "epoch": 102.64,
      "step": 410
    },
    {
      "loss": 0.0081,
      "grad_norm": 0.028645571321249008,
      "learning_rate": 7.938144329896907e-05,
      "epoch": 103.96,
      "step": 415
    },
    {
      "loss": 0.0082,
      "grad_norm": 0.08248745650053024,
      "learning_rate": 7.90950744558992e-05,
      "epoch": 105.0,
      "step": 420
    },
    {
      "loss": 0.008,
      "grad_norm": 0.01870081014931202,
      "learning_rate": 7.880870561282932e-05,
      "epoch": 106.32,
      "step": 425
    },
    {
      "loss": 0.008,
      "grad_norm": 0.023562012240290642,
      "learning_rate": 7.852233676975945e-05,
      "epoch": 107.64,
      "step": 430
    },
    {
      "loss": 0.0079,
      "grad_norm": 0.025380929931998253,
      "learning_rate": 7.823596792668958e-05,
      "epoch": 108.96,
      "step": 435
    },
    {
      "loss": 0.0079,
      "grad_norm": 0.053522542119026184,
      "learning_rate": 7.79495990836197e-05,
      "epoch": 110.0,
      "step": 440
    },
    {
      "loss": 0.008,
      "grad_norm": 0.015515265986323357,
      "learning_rate": 7.766323024054983e-05,
      "epoch": 111.32,
      "step": 445
    },
    {
      "loss": 0.0082,
      "grad_norm": 0.023496465757489204,
      "learning_rate": 7.737686139747996e-05,
      "epoch": 112.64,
      "step": 450
    },
    {
      "loss": 0.008,
      "grad_norm": 0.018937699496746063,
      "learning_rate": 7.709049255441008e-05,
      "epoch": 113.96,
      "step": 455
    },
    {
      "loss": 0.0081,
      "grad_norm": 0.0531855970621109,
      "learning_rate": 7.680412371134021e-05,
      "epoch": 115.0,
      "step": 460
    },
    {
      "loss": 0.0079,
      "grad_norm": 0.01843051239848137,
      "learning_rate": 7.651775486827034e-05,
      "epoch": 116.32,
      "step": 465
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.024867741391062737,
      "learning_rate": 7.623138602520046e-05,
      "epoch": 117.64,
      "step": 470
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.019503291696310043,
      "learning_rate": 7.594501718213059e-05,
      "epoch": 118.96,
      "step": 475
    },
    {
      "loss": 0.0079,
      "grad_norm": 0.0843457356095314,
      "learning_rate": 7.565864833906071e-05,
      "epoch": 120.0,
      "step": 480
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.020036857575178146,
      "learning_rate": 7.537227949599084e-05,
      "epoch": 121.32,
      "step": 485
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.018207866698503494,
      "learning_rate": 7.508591065292097e-05,
      "epoch": 122.64,
      "step": 490
    },
    {
      "loss": 0.0079,
      "grad_norm": 0.019459234550595284,
      "learning_rate": 7.47995418098511e-05,
      "epoch": 123.96,
      "step": 495
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.05320321023464203,
      "learning_rate": 7.451317296678122e-05,
      "epoch": 125.0,
      "step": 500
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.022196603938937187,
      "learning_rate": 7.422680412371135e-05,
      "epoch": 126.32,
      "step": 505
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.023695651441812515,
      "learning_rate": 7.394043528064147e-05,
      "epoch": 127.64,
      "step": 510
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.01924101635813713,
      "learning_rate": 7.36540664375716e-05,
      "epoch": 128.96,
      "step": 515
    },
    {
      "loss": 0.008,
      "grad_norm": 0.045008257031440735,
      "learning_rate": 7.336769759450171e-05,
      "epoch": 130.0,
      "step": 520
    },
    {
      "loss": 0.0081,
      "grad_norm": 0.02046283893287182,
      "learning_rate": 7.308132875143184e-05,
      "epoch": 131.32,
      "step": 525
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.01908033899962902,
      "learning_rate": 7.279495990836197e-05,
      "epoch": 132.64,
      "step": 530
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.02048024907708168,
      "learning_rate": 7.250859106529209e-05,
      "epoch": 133.96,
      "step": 535
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.06090843304991722,
      "learning_rate": 7.222222222222222e-05,
      "epoch": 135.0,
      "step": 540
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.019933057948946953,
      "learning_rate": 7.193585337915235e-05,
      "epoch": 136.32,
      "step": 545
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.01903274841606617,
      "learning_rate": 7.164948453608247e-05,
      "epoch": 137.64,
      "step": 550
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.017385313287377357,
      "learning_rate": 7.136311569301261e-05,
      "epoch": 138.96,
      "step": 555
    },
    {
      "loss": 0.0083,
      "grad_norm": 0.06623967736959457,
      "learning_rate": 7.107674684994274e-05,
      "epoch": 140.0,
      "step": 560
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.015628529712557793,
      "learning_rate": 7.079037800687286e-05,
      "epoch": 141.32,
      "step": 565
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.015574640594422817,
      "learning_rate": 7.050400916380299e-05,
      "epoch": 142.64,
      "step": 570
    },
    {
      "loss": 0.008,
      "grad_norm": 0.01933234930038452,
      "learning_rate": 7.02176403207331e-05,
      "epoch": 143.96,
      "step": 575
    },
    {
      "loss": 0.0069,
      "grad_norm": 0.05281882733106613,
      "learning_rate": 6.993127147766323e-05,
      "epoch": 145.0,
      "step": 580
    },
    {
      "loss": 0.008,
      "grad_norm": 0.01630941592156887,
      "learning_rate": 6.964490263459336e-05,
      "epoch": 146.32,
      "step": 585
    },
    {
      "loss": 0.0079,
      "grad_norm": 0.0179227776825428,
      "learning_rate": 6.935853379152348e-05,
      "epoch": 147.64,
      "step": 590
    },
    {
      "loss": 0.0083,
      "grad_norm": 0.018087850883603096,
      "learning_rate": 6.907216494845361e-05,
      "epoch": 148.96,
      "step": 595
    },
    {
      "loss": 0.0081,
      "grad_norm": 0.06578338891267776,
      "learning_rate": 6.878579610538374e-05,
      "epoch": 150.0,
      "step": 600
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.01650572009384632,
      "learning_rate": 6.849942726231386e-05,
      "epoch": 151.32,
      "step": 605
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.018793148919939995,
      "learning_rate": 6.821305841924399e-05,
      "epoch": 152.64,
      "step": 610
    },
    {
      "loss": 0.0079,
      "grad_norm": 0.01930132694542408,
      "learning_rate": 6.792668957617412e-05,
      "epoch": 153.96,
      "step": 615
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.06265976279973984,
      "learning_rate": 6.764032073310424e-05,
      "epoch": 155.0,
      "step": 620
    },
    {
      "loss": 0.008,
      "grad_norm": 0.018820270895957947,
      "learning_rate": 6.735395189003437e-05,
      "epoch": 156.32,
      "step": 625
    },
    {
      "loss": 0.0079,
      "grad_norm": 0.01635253243148327,
      "learning_rate": 6.706758304696448e-05,
      "epoch": 157.64,
      "step": 630
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.021151645109057426,
      "learning_rate": 6.678121420389462e-05,
      "epoch": 158.96,
      "step": 635
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.04630030691623688,
      "learning_rate": 6.649484536082475e-05,
      "epoch": 160.0,
      "step": 640
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.018470125272870064,
      "learning_rate": 6.620847651775487e-05,
      "epoch": 161.32,
      "step": 645
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.018053382635116577,
      "learning_rate": 6.5922107674685e-05,
      "epoch": 162.64,
      "step": 650
    },
    {
      "loss": 0.0082,
      "grad_norm": 0.01694164238870144,
      "learning_rate": 6.563573883161513e-05,
      "epoch": 163.96,
      "step": 655
    },
    {
      "loss": 0.0068,
      "grad_norm": 0.052025895565748215,
      "learning_rate": 6.534936998854525e-05,
      "epoch": 165.0,
      "step": 660
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.017838332802057266,
      "learning_rate": 6.506300114547538e-05,
      "epoch": 166.32,
      "step": 665
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.017433341592550278,
      "learning_rate": 6.477663230240551e-05,
      "epoch": 167.64,
      "step": 670
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.01773149147629738,
      "learning_rate": 6.449026345933563e-05,
      "epoch": 168.96,
      "step": 675
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.035425540059804916,
      "learning_rate": 6.420389461626576e-05,
      "epoch": 170.0,
      "step": 680
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.016005905345082283,
      "learning_rate": 6.391752577319587e-05,
      "epoch": 171.32,
      "step": 685
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.01620178110897541,
      "learning_rate": 6.3631156930126e-05,
      "epoch": 172.64,
      "step": 690
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.01678958721458912,
      "learning_rate": 6.334478808705613e-05,
      "epoch": 173.96,
      "step": 695
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.05086919292807579,
      "learning_rate": 6.305841924398625e-05,
      "epoch": 175.0,
      "step": 700
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.018727965652942657,
      "learning_rate": 6.277205040091638e-05,
      "epoch": 176.32,
      "step": 705
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.01865091733634472,
      "learning_rate": 6.24856815578465e-05,
      "epoch": 177.64,
      "step": 710
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.017102951183915138,
      "learning_rate": 6.219931271477663e-05,
      "epoch": 178.96,
      "step": 715
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.04637826979160309,
      "learning_rate": 6.191294387170676e-05,
      "epoch": 180.0,
      "step": 720
    },
    {
      "loss": 0.0069,
      "grad_norm": 0.017209526151418686,
      "learning_rate": 6.162657502863689e-05,
      "epoch": 181.32,
      "step": 725
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.015485143288969994,
      "learning_rate": 6.134020618556701e-05,
      "epoch": 182.64,
      "step": 730
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.016986800357699394,
      "learning_rate": 6.105383734249714e-05,
      "epoch": 183.96,
      "step": 735
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.041255682706832886,
      "learning_rate": 6.076746849942726e-05,
      "epoch": 185.0,
      "step": 740
    },
    {
      "loss": 0.0079,
      "grad_norm": 0.01544552855193615,
      "learning_rate": 6.0481099656357384e-05,
      "epoch": 186.32,
      "step": 745
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.014815689995884895,
      "learning_rate": 6.019473081328752e-05,
      "epoch": 187.64,
      "step": 750
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.01709260232746601,
      "learning_rate": 5.9908361970217644e-05,
      "epoch": 188.96,
      "step": 755
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.05759681761264801,
      "learning_rate": 5.962199312714777e-05,
      "epoch": 190.0,
      "step": 760
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.01603686809539795,
      "learning_rate": 5.93356242840779e-05,
      "epoch": 191.32,
      "step": 765
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.016008200123906136,
      "learning_rate": 5.904925544100802e-05,
      "epoch": 192.64,
      "step": 770
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.017906546592712402,
      "learning_rate": 5.876288659793815e-05,
      "epoch": 193.96,
      "step": 775
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.05671071633696556,
      "learning_rate": 5.8476517754868276e-05,
      "epoch": 195.0,
      "step": 780
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014617575332522392,
      "learning_rate": 5.81901489117984e-05,
      "epoch": 196.32,
      "step": 785
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.016560440883040428,
      "learning_rate": 5.790378006872853e-05,
      "epoch": 197.64,
      "step": 790
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.014147703535854816,
      "learning_rate": 5.761741122565865e-05,
      "epoch": 198.96,
      "step": 795
    },
    {
      "loss": 0.008,
      "grad_norm": 0.060914311558008194,
      "learning_rate": 5.7331042382588775e-05,
      "epoch": 200.0,
      "step": 800
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.016522547230124474,
      "learning_rate": 5.70446735395189e-05,
      "epoch": 201.32,
      "step": 805
    },
    {
      "loss": 0.008,
      "grad_norm": 0.017009710893034935,
      "learning_rate": 5.675830469644903e-05,
      "epoch": 202.64,
      "step": 810
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.01857658289372921,
      "learning_rate": 5.6471935853379155e-05,
      "epoch": 203.96,
      "step": 815
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.059511855244636536,
      "learning_rate": 5.618556701030928e-05,
      "epoch": 205.0,
      "step": 820
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.013280014507472515,
      "learning_rate": 5.589919816723941e-05,
      "epoch": 206.32,
      "step": 825
    },
    {
      "loss": 0.0079,
      "grad_norm": 0.01620969921350479,
      "learning_rate": 5.5612829324169534e-05,
      "epoch": 207.64,
      "step": 830
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.017914501950144768,
      "learning_rate": 5.532646048109966e-05,
      "epoch": 208.96,
      "step": 835
    },
    {
      "loss": 0.0083,
      "grad_norm": 0.07273726910352707,
      "learning_rate": 5.504009163802979e-05,
      "epoch": 210.0,
      "step": 840
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.018732460215687752,
      "learning_rate": 5.4753722794959914e-05,
      "epoch": 211.32,
      "step": 845
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.017509156838059425,
      "learning_rate": 5.4467353951890033e-05,
      "epoch": 212.64,
      "step": 850
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.017160458490252495,
      "learning_rate": 5.418098510882016e-05,
      "epoch": 213.96,
      "step": 855
    },
    {
      "loss": 0.008,
      "grad_norm": 0.041403669863939285,
      "learning_rate": 5.3894616265750286e-05,
      "epoch": 215.0,
      "step": 860
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.01642041653394699,
      "learning_rate": 5.360824742268041e-05,
      "epoch": 216.32,
      "step": 865
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.015811027958989143,
      "learning_rate": 5.332187857961054e-05,
      "epoch": 217.64,
      "step": 870
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.01576993055641651,
      "learning_rate": 5.3035509736540666e-05,
      "epoch": 218.96,
      "step": 875
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.04965505003929138,
      "learning_rate": 5.274914089347079e-05,
      "epoch": 220.0,
      "step": 880
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.014361195266246796,
      "learning_rate": 5.246277205040092e-05,
      "epoch": 221.32,
      "step": 885
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.01668650470674038,
      "learning_rate": 5.2176403207331045e-05,
      "epoch": 222.64,
      "step": 890
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.014330503530800343,
      "learning_rate": 5.189003436426118e-05,
      "epoch": 223.96,
      "step": 895
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.06589684635400772,
      "learning_rate": 5.1603665521191305e-05,
      "epoch": 225.0,
      "step": 900
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.016238626092672348,
      "learning_rate": 5.131729667812142e-05,
      "epoch": 226.32,
      "step": 905
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.015328693203628063,
      "learning_rate": 5.1030927835051544e-05,
      "epoch": 227.64,
      "step": 910
    },
    {
      "loss": 0.008,
      "grad_norm": 0.0158688984811306,
      "learning_rate": 5.074455899198167e-05,
      "epoch": 228.96,
      "step": 915
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.043009161949157715,
      "learning_rate": 5.04581901489118e-05,
      "epoch": 230.0,
      "step": 920
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014415422454476357,
      "learning_rate": 5.0171821305841924e-05,
      "epoch": 231.32,
      "step": 925
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.014107465744018555,
      "learning_rate": 4.988545246277205e-05,
      "epoch": 232.64,
      "step": 930
    },
    {
      "loss": 0.0083,
      "grad_norm": 0.01650586724281311,
      "learning_rate": 4.9599083619702184e-05,
      "epoch": 233.96,
      "step": 935
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.05056240037083626,
      "learning_rate": 4.931271477663231e-05,
      "epoch": 235.0,
      "step": 940
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.015558524057269096,
      "learning_rate": 4.902634593356243e-05,
      "epoch": 236.32,
      "step": 945
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.01564890705049038,
      "learning_rate": 4.8739977090492556e-05,
      "epoch": 237.64,
      "step": 950
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.016325442120432854,
      "learning_rate": 4.845360824742268e-05,
      "epoch": 238.96,
      "step": 955
    },
    {
      "loss": 0.0085,
      "grad_norm": 0.06062578037381172,
      "learning_rate": 4.816723940435281e-05,
      "epoch": 240.0,
      "step": 960
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.015442856587469578,
      "learning_rate": 4.7880870561282936e-05,
      "epoch": 241.32,
      "step": 965
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.014774948358535767,
      "learning_rate": 4.7594501718213055e-05,
      "epoch": 242.64,
      "step": 970
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.012394092977046967,
      "learning_rate": 4.730813287514318e-05,
      "epoch": 243.96,
      "step": 975
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.049357641488313675,
      "learning_rate": 4.7021764032073315e-05,
      "epoch": 245.0,
      "step": 980
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.014757816679775715,
      "learning_rate": 4.673539518900344e-05,
      "epoch": 246.32,
      "step": 985
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.013863503001630306,
      "learning_rate": 4.644902634593357e-05,
      "epoch": 247.64,
      "step": 990
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.01559313666075468,
      "learning_rate": 4.6162657502863694e-05,
      "epoch": 248.96,
      "step": 995
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.056271184235811234,
      "learning_rate": 4.5876288659793814e-05,
      "epoch": 250.0,
      "step": 1000
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.01591244712471962,
      "learning_rate": 4.558991981672394e-05,
      "epoch": 251.32,
      "step": 1005
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.016048980876803398,
      "learning_rate": 4.530355097365407e-05,
      "epoch": 252.64,
      "step": 1010
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.014479824341833591,
      "learning_rate": 4.5017182130584194e-05,
      "epoch": 253.96,
      "step": 1015
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.03930358588695526,
      "learning_rate": 4.473081328751432e-05,
      "epoch": 255.0,
      "step": 1020
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014157247729599476,
      "learning_rate": 4.4444444444444447e-05,
      "epoch": 256.32,
      "step": 1025
    },
    {
      "loss": 0.0069,
      "grad_norm": 0.01348467543721199,
      "learning_rate": 4.415807560137457e-05,
      "epoch": 257.64,
      "step": 1030
    },
    {
      "loss": 0.0068,
      "grad_norm": 0.014039754867553711,
      "learning_rate": 4.38717067583047e-05,
      "epoch": 258.96,
      "step": 1035
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.05152149498462677,
      "learning_rate": 4.3585337915234826e-05,
      "epoch": 260.0,
      "step": 1040
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.014613005332648754,
      "learning_rate": 4.329896907216495e-05,
      "epoch": 261.32,
      "step": 1045
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.01624435745179653,
      "learning_rate": 4.301260022909508e-05,
      "epoch": 262.64,
      "step": 1050
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.013759251683950424,
      "learning_rate": 4.27262313860252e-05,
      "epoch": 263.96,
      "step": 1055
    },
    {
      "loss": 0.0083,
      "grad_norm": 0.04224378615617752,
      "learning_rate": 4.2439862542955325e-05,
      "epoch": 265.0,
      "step": 1060
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.014585004188120365,
      "learning_rate": 4.215349369988545e-05,
      "epoch": 266.32,
      "step": 1065
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.015146151185035706,
      "learning_rate": 4.1867124856815585e-05,
      "epoch": 267.64,
      "step": 1070
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.01387553196400404,
      "learning_rate": 4.158075601374571e-05,
      "epoch": 268.96,
      "step": 1075
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.04610530287027359,
      "learning_rate": 4.129438717067583e-05,
      "epoch": 270.0,
      "step": 1080
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.015416707843542099,
      "learning_rate": 4.100801832760596e-05,
      "epoch": 271.32,
      "step": 1085
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.015935108065605164,
      "learning_rate": 4.0721649484536084e-05,
      "epoch": 272.64,
      "step": 1090
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.015026471577584743,
      "learning_rate": 4.043528064146621e-05,
      "epoch": 273.96,
      "step": 1095
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.05057341605424881,
      "learning_rate": 4.014891179839634e-05,
      "epoch": 275.0,
      "step": 1100
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.015247948467731476,
      "learning_rate": 3.9862542955326463e-05,
      "epoch": 276.32,
      "step": 1105
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.01620655134320259,
      "learning_rate": 3.957617411225659e-05,
      "epoch": 277.64,
      "step": 1110
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.015572886914014816,
      "learning_rate": 3.9289805269186716e-05,
      "epoch": 278.96,
      "step": 1115
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.06242557615041733,
      "learning_rate": 3.900343642611684e-05,
      "epoch": 280.0,
      "step": 1120
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.014887187629938126,
      "learning_rate": 3.871706758304697e-05,
      "epoch": 281.32,
      "step": 1125
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.014611662365496159,
      "learning_rate": 3.8430698739977096e-05,
      "epoch": 282.64,
      "step": 1130
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014667696319520473,
      "learning_rate": 3.8144329896907216e-05,
      "epoch": 283.96,
      "step": 1135
    },
    {
      "loss": 0.008,
      "grad_norm": 0.05566182732582092,
      "learning_rate": 3.785796105383734e-05,
      "epoch": 285.0,
      "step": 1140
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.014292476698756218,
      "learning_rate": 3.757159221076747e-05,
      "epoch": 286.32,
      "step": 1145
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.018137851729989052,
      "learning_rate": 3.7285223367697595e-05,
      "epoch": 287.64,
      "step": 1150
    },
    {
      "loss": 0.007,
      "grad_norm": 0.020582007244229317,
      "learning_rate": 3.699885452462772e-05,
      "epoch": 288.96,
      "step": 1155
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.04622265696525574,
      "learning_rate": 3.671248568155785e-05,
      "epoch": 290.0,
      "step": 1160
    },
    {
      "loss": 0.0067,
      "grad_norm": 0.013806179165840149,
      "learning_rate": 3.6426116838487974e-05,
      "epoch": 291.32,
      "step": 1165
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.014332096092402935,
      "learning_rate": 3.61397479954181e-05,
      "epoch": 292.64,
      "step": 1170
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014601546339690685,
      "learning_rate": 3.585337915234823e-05,
      "epoch": 293.96,
      "step": 1175
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.04430964216589928,
      "learning_rate": 3.5567010309278354e-05,
      "epoch": 295.0,
      "step": 1180
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.013638140633702278,
      "learning_rate": 3.528064146620848e-05,
      "epoch": 296.32,
      "step": 1185
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.016876142472028732,
      "learning_rate": 3.49942726231386e-05,
      "epoch": 297.64,
      "step": 1190
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.017564833164215088,
      "learning_rate": 3.4707903780068726e-05,
      "epoch": 298.96,
      "step": 1195
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.057826753705739975,
      "learning_rate": 3.442153493699885e-05,
      "epoch": 300.0,
      "step": 1200
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.01532206404954195,
      "learning_rate": 3.4135166093928986e-05,
      "epoch": 301.32,
      "step": 1205
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.012946589849889278,
      "learning_rate": 3.384879725085911e-05,
      "epoch": 302.64,
      "step": 1210
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.015184017829596996,
      "learning_rate": 3.356242840778923e-05,
      "epoch": 303.96,
      "step": 1215
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.05617891252040863,
      "learning_rate": 3.327605956471936e-05,
      "epoch": 305.0,
      "step": 1220
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.01357415970414877,
      "learning_rate": 3.2989690721649485e-05,
      "epoch": 306.32,
      "step": 1225
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.014721889980137348,
      "learning_rate": 3.270332187857961e-05,
      "epoch": 307.64,
      "step": 1230
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.013552107848227024,
      "learning_rate": 3.241695303550974e-05,
      "epoch": 308.96,
      "step": 1235
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.04199804365634918,
      "learning_rate": 3.2130584192439865e-05,
      "epoch": 310.0,
      "step": 1240
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.014133174903690815,
      "learning_rate": 3.184421534936999e-05,
      "epoch": 311.32,
      "step": 1245
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.013671276159584522,
      "learning_rate": 3.155784650630012e-05,
      "epoch": 312.64,
      "step": 1250
    },
    {
      "loss": 0.0081,
      "grad_norm": 0.014025907963514328,
      "learning_rate": 3.1271477663230244e-05,
      "epoch": 313.96,
      "step": 1255
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.04884964972734451,
      "learning_rate": 3.098510882016037e-05,
      "epoch": 315.0,
      "step": 1260
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.014592539519071579,
      "learning_rate": 3.06987399770905e-05,
      "epoch": 316.32,
      "step": 1265
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.013585659675300121,
      "learning_rate": 3.0412371134020617e-05,
      "epoch": 317.64,
      "step": 1270
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.017339883372187614,
      "learning_rate": 3.0126002290950743e-05,
      "epoch": 318.96,
      "step": 1275
    },
    {
      "loss": 0.007,
      "grad_norm": 0.02973540499806404,
      "learning_rate": 2.983963344788087e-05,
      "epoch": 320.0,
      "step": 1280
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.013446581549942493,
      "learning_rate": 2.9553264604811e-05,
      "epoch": 321.32,
      "step": 1285
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.016039373353123665,
      "learning_rate": 2.9266895761741126e-05,
      "epoch": 322.64,
      "step": 1290
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.016038790345191956,
      "learning_rate": 2.8980526918671253e-05,
      "epoch": 323.96,
      "step": 1295
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.0627412423491478,
      "learning_rate": 2.8694158075601372e-05,
      "epoch": 325.0,
      "step": 1300
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.014775470830500126,
      "learning_rate": 2.8407789232531502e-05,
      "epoch": 326.32,
      "step": 1305
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.015030876733362675,
      "learning_rate": 2.812142038946163e-05,
      "epoch": 327.64,
      "step": 1310
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014904695563018322,
      "learning_rate": 2.7835051546391755e-05,
      "epoch": 328.96,
      "step": 1315
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.04180411621928215,
      "learning_rate": 2.754868270332188e-05,
      "epoch": 330.0,
      "step": 1320
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.014243628829717636,
      "learning_rate": 2.7262313860252005e-05,
      "epoch": 331.32,
      "step": 1325
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.013925555162131786,
      "learning_rate": 2.697594501718213e-05,
      "epoch": 332.64,
      "step": 1330
    },
    {
      "loss": 0.0068,
      "grad_norm": 0.015809444710612297,
      "learning_rate": 2.6689576174112258e-05,
      "epoch": 333.96,
      "step": 1335
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.04403695464134216,
      "learning_rate": 2.6403207331042384e-05,
      "epoch": 335.0,
      "step": 1340
    },
    {
      "loss": 0.0069,
      "grad_norm": 0.01286964863538742,
      "learning_rate": 2.611683848797251e-05,
      "epoch": 336.32,
      "step": 1345
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.013802004978060722,
      "learning_rate": 2.5830469644902637e-05,
      "epoch": 337.64,
      "step": 1350
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.013678145594894886,
      "learning_rate": 2.554410080183276e-05,
      "epoch": 338.96,
      "step": 1355
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.05216424912214279,
      "learning_rate": 2.5257731958762887e-05,
      "epoch": 340.0,
      "step": 1360
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.014664101414382458,
      "learning_rate": 2.4971363115693013e-05,
      "epoch": 341.32,
      "step": 1365
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.013857342302799225,
      "learning_rate": 2.468499427262314e-05,
      "epoch": 342.64,
      "step": 1370
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.013593857176601887,
      "learning_rate": 2.4398625429553266e-05,
      "epoch": 343.96,
      "step": 1375
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.05102638900279999,
      "learning_rate": 2.4112256586483393e-05,
      "epoch": 345.0,
      "step": 1380
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.014895925298333168,
      "learning_rate": 2.3825887743413516e-05,
      "epoch": 346.32,
      "step": 1385
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.013237119652330875,
      "learning_rate": 2.3539518900343642e-05,
      "epoch": 347.64,
      "step": 1390
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.012369737960398197,
      "learning_rate": 2.3253150057273772e-05,
      "epoch": 348.96,
      "step": 1395
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.051589202135801315,
      "learning_rate": 2.2966781214203895e-05,
      "epoch": 350.0,
      "step": 1400
    },
    {
      "loss": 0.007,
      "grad_norm": 0.013157777488231659,
      "learning_rate": 2.268041237113402e-05,
      "epoch": 351.32,
      "step": 1405
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.014956669881939888,
      "learning_rate": 2.2394043528064148e-05,
      "epoch": 352.64,
      "step": 1410
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.015596367418766022,
      "learning_rate": 2.210767468499427e-05,
      "epoch": 353.96,
      "step": 1415
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.05619361996650696,
      "learning_rate": 2.18213058419244e-05,
      "epoch": 355.0,
      "step": 1420
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.016337988898158073,
      "learning_rate": 2.1534936998854528e-05,
      "epoch": 356.32,
      "step": 1425
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014916105195879936,
      "learning_rate": 2.124856815578465e-05,
      "epoch": 357.64,
      "step": 1430
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.015843307599425316,
      "learning_rate": 2.0962199312714777e-05,
      "epoch": 358.96,
      "step": 1435
    },
    {
      "loss": 0.0068,
      "grad_norm": 0.04489946737885475,
      "learning_rate": 2.0675830469644904e-05,
      "epoch": 360.0,
      "step": 1440
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.015025758184492588,
      "learning_rate": 2.038946162657503e-05,
      "epoch": 361.32,
      "step": 1445
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.013098886236548424,
      "learning_rate": 2.0103092783505157e-05,
      "epoch": 362.64,
      "step": 1450
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.013708668760955334,
      "learning_rate": 1.981672394043528e-05,
      "epoch": 363.96,
      "step": 1455
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.042526885867118835,
      "learning_rate": 1.9530355097365406e-05,
      "epoch": 365.0,
      "step": 1460
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.013285653665661812,
      "learning_rate": 1.9243986254295536e-05,
      "epoch": 366.32,
      "step": 1465
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.013405455276370049,
      "learning_rate": 1.895761741122566e-05,
      "epoch": 367.64,
      "step": 1470
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.013635101728141308,
      "learning_rate": 1.8671248568155786e-05,
      "epoch": 368.96,
      "step": 1475
    },
    {
      "loss": 0.0068,
      "grad_norm": 0.02924210950732231,
      "learning_rate": 1.8384879725085912e-05,
      "epoch": 370.0,
      "step": 1480
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.0134768420830369,
      "learning_rate": 1.809851088201604e-05,
      "epoch": 371.32,
      "step": 1485
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.01462192740291357,
      "learning_rate": 1.7812142038946165e-05,
      "epoch": 372.64,
      "step": 1490
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.014531673863530159,
      "learning_rate": 1.7525773195876288e-05,
      "epoch": 373.96,
      "step": 1495
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.058406367897987366,
      "learning_rate": 1.7239404352806415e-05,
      "epoch": 375.0,
      "step": 1500
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.015267807058990002,
      "learning_rate": 1.695303550973654e-05,
      "epoch": 376.32,
      "step": 1505
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.014395340345799923,
      "learning_rate": 1.6666666666666667e-05,
      "epoch": 377.64,
      "step": 1510
    },
    {
      "loss": 0.007,
      "grad_norm": 0.016068171709775925,
      "learning_rate": 1.6380297823596794e-05,
      "epoch": 378.96,
      "step": 1515
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.048419468104839325,
      "learning_rate": 1.609392898052692e-05,
      "epoch": 380.0,
      "step": 1520
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.014497853815555573,
      "learning_rate": 1.5807560137457044e-05,
      "epoch": 381.32,
      "step": 1525
    },
    {
      "loss": 0.0067,
      "grad_norm": 0.015284787863492966,
      "learning_rate": 1.5521191294387173e-05,
      "epoch": 382.64,
      "step": 1530
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.012362837791442871,
      "learning_rate": 1.5234822451317298e-05,
      "epoch": 383.96,
      "step": 1535
    },
    {
      "loss": 0.0085,
      "grad_norm": 0.04469563066959381,
      "learning_rate": 1.4948453608247423e-05,
      "epoch": 385.0,
      "step": 1540
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.01514203380793333,
      "learning_rate": 1.466208476517755e-05,
      "epoch": 386.32,
      "step": 1545
    },
    {
      "loss": 0.007,
      "grad_norm": 0.014873849228024483,
      "learning_rate": 1.4375715922107674e-05,
      "epoch": 387.64,
      "step": 1550
    },
    {
      "loss": 0.0077,
      "grad_norm": 0.014835209585726261,
      "learning_rate": 1.40893470790378e-05,
      "epoch": 388.96,
      "step": 1555
    },
    {
      "loss": 0.0067,
      "grad_norm": 0.05331091210246086,
      "learning_rate": 1.3802978235967929e-05,
      "epoch": 390.0,
      "step": 1560
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014174208045005798,
      "learning_rate": 1.3516609392898052e-05,
      "epoch": 391.32,
      "step": 1565
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.013322395272552967,
      "learning_rate": 1.323024054982818e-05,
      "epoch": 392.64,
      "step": 1570
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.01326716784387827,
      "learning_rate": 1.2943871706758307e-05,
      "epoch": 393.96,
      "step": 1575
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.04614840820431709,
      "learning_rate": 1.2657502863688431e-05,
      "epoch": 395.0,
      "step": 1580
    },
    {
      "loss": 0.007,
      "grad_norm": 0.013342869468033314,
      "learning_rate": 1.2371134020618558e-05,
      "epoch": 396.32,
      "step": 1585
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.013926559127867222,
      "learning_rate": 1.2084765177548683e-05,
      "epoch": 397.64,
      "step": 1590
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.014695955440402031,
      "learning_rate": 1.1798396334478809e-05,
      "epoch": 398.96,
      "step": 1595
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.04952579736709595,
      "learning_rate": 1.1512027491408934e-05,
      "epoch": 400.0,
      "step": 1600
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014569703489542007,
      "learning_rate": 1.1225658648339062e-05,
      "epoch": 401.32,
      "step": 1605
    },
    {
      "loss": 0.0069,
      "grad_norm": 0.013354274444282055,
      "learning_rate": 1.0939289805269187e-05,
      "epoch": 402.64,
      "step": 1610
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.013735768385231495,
      "learning_rate": 1.0652920962199313e-05,
      "epoch": 403.96,
      "step": 1615
    },
    {
      "loss": 0.007,
      "grad_norm": 0.049133505672216415,
      "learning_rate": 1.036655211912944e-05,
      "epoch": 405.0,
      "step": 1620
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.018285108730196953,
      "learning_rate": 1.0080183276059566e-05,
      "epoch": 406.32,
      "step": 1625
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.015425886027514935,
      "learning_rate": 9.793814432989691e-06,
      "epoch": 407.64,
      "step": 1630
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.0143016017973423,
      "learning_rate": 9.507445589919818e-06,
      "epoch": 408.96,
      "step": 1635
    },
    {
      "loss": 0.007,
      "grad_norm": 0.044388387352228165,
      "learning_rate": 9.221076746849944e-06,
      "epoch": 410.0,
      "step": 1640
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.013605360873043537,
      "learning_rate": 8.934707903780069e-06,
      "epoch": 411.32,
      "step": 1645
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.01632440648972988,
      "learning_rate": 8.648339060710195e-06,
      "epoch": 412.64,
      "step": 1650
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.01337454468011856,
      "learning_rate": 8.36197021764032e-06,
      "epoch": 413.96,
      "step": 1655
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.04445042833685875,
      "learning_rate": 8.075601374570448e-06,
      "epoch": 415.0,
      "step": 1660
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.014583917334675789,
      "learning_rate": 7.789232531500573e-06,
      "epoch": 416.32,
      "step": 1665
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.01495777815580368,
      "learning_rate": 7.502863688430699e-06,
      "epoch": 417.64,
      "step": 1670
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.015282702632248402,
      "learning_rate": 7.216494845360824e-06,
      "epoch": 418.96,
      "step": 1675
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.04799186438322067,
      "learning_rate": 6.930126002290952e-06,
      "epoch": 420.0,
      "step": 1680
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014717133715748787,
      "learning_rate": 6.643757159221077e-06,
      "epoch": 421.32,
      "step": 1685
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014286785386502743,
      "learning_rate": 6.357388316151203e-06,
      "epoch": 422.64,
      "step": 1690
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.01473111379891634,
      "learning_rate": 6.071019473081329e-06,
      "epoch": 423.96,
      "step": 1695
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.0494658537209034,
      "learning_rate": 5.784650630011455e-06,
      "epoch": 425.0,
      "step": 1700
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.013659783639013767,
      "learning_rate": 5.498281786941581e-06,
      "epoch": 426.32,
      "step": 1705
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.014057163149118423,
      "learning_rate": 5.211912943871707e-06,
      "epoch": 427.64,
      "step": 1710
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.015153934247791767,
      "learning_rate": 4.925544100801833e-06,
      "epoch": 428.96,
      "step": 1715
    },
    {
      "loss": 0.0066,
      "grad_norm": 0.03869548439979553,
      "learning_rate": 4.639175257731959e-06,
      "epoch": 430.0,
      "step": 1720
    },
    {
      "loss": 0.0075,
      "grad_norm": 0.015271749347448349,
      "learning_rate": 4.352806414662085e-06,
      "epoch": 431.32,
      "step": 1725
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014637422747910023,
      "learning_rate": 4.066437571592211e-06,
      "epoch": 432.64,
      "step": 1730
    },
    {
      "loss": 0.007,
      "grad_norm": 0.0150294816121459,
      "learning_rate": 3.7800687285223365e-06,
      "epoch": 433.96,
      "step": 1735
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.042534444481134415,
      "learning_rate": 3.493699885452463e-06,
      "epoch": 435.0,
      "step": 1740
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.015038814395666122,
      "learning_rate": 3.2073310423825886e-06,
      "epoch": 436.32,
      "step": 1745
    },
    {
      "loss": 0.0073,
      "grad_norm": 0.014388471841812134,
      "learning_rate": 2.920962199312715e-06,
      "epoch": 437.64,
      "step": 1750
    },
    {
      "loss": 0.0076,
      "grad_norm": 0.014901977963745594,
      "learning_rate": 2.6345933562428407e-06,
      "epoch": 438.96,
      "step": 1755
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.04744100198149681,
      "learning_rate": 2.3482245131729668e-06,
      "epoch": 440.0,
      "step": 1760
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.013388304971158504,
      "learning_rate": 2.061855670103093e-06,
      "epoch": 441.32,
      "step": 1765
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.01415504701435566,
      "learning_rate": 1.7754868270332189e-06,
      "epoch": 442.64,
      "step": 1770
    },
    {
      "loss": 0.0071,
      "grad_norm": 0.014000612311065197,
      "learning_rate": 1.4891179839633447e-06,
      "epoch": 443.96,
      "step": 1775
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.05067407712340355,
      "learning_rate": 1.202749140893471e-06,
      "epoch": 445.0,
      "step": 1780
    },
    {
      "loss": 0.0072,
      "grad_norm": 0.014506036415696144,
      "learning_rate": 9.163802978235968e-07,
      "epoch": 446.32,
      "step": 1785
    },
    {
      "loss": 0.0074,
      "grad_norm": 0.013609930872917175,
      "learning_rate": 6.300114547537229e-07,
      "epoch": 447.64,
      "step": 1790
    },
    {
      "loss": 0.0078,
      "grad_norm": 0.01482964213937521,
      "learning_rate": 3.436426116838488e-07,
      "epoch": 448.96,
      "step": 1795
    },
    {
      "loss": 0.0065,
      "grad_norm": 0.04897365719079971,
      "learning_rate": 5.72737686139748e-08,
      "epoch": 450.0,
      "step": 1800
    },
    {
      "train_runtime": 30821.8954,
      "train_samples_per_second": 1.869,
      "train_steps_per_second": 0.058,
      "total_flos": 1.5308141101056e+18,
      "train_loss": 0.03864684621699982,
      "epoch": 450.0,
      "step": 1800
    }
  ],
  "training_args": {
    "output_dir": "/hkfs/work/workspace/scratch/tum_fmp0582-dndworkspace/自己训练lora/train_lora/outputs/ToT-Math-V1/group_03/checkpoints",
    "overwrite_output_dir": false,
    "do_train": false,
    "do_eval": false,
    "do_predict": false,
    "eval_strategy": "no",
    "prediction_loss_only": false,
    "per_device_train_batch_size": 2,
    "per_device_eval_batch_size": 8,
    "per_gpu_train_batch_size": null,
    "per_gpu_eval_batch_size": null,
    "gradient_accumulation_steps": 16,
    "eval_accumulation_steps": null,
    "eval_delay": 0,
    "torch_empty_cache_steps": null,
    "learning_rate": 0.0001,
    "weight_decay": 0.01,
    "adam_beta1": 0.9,
    "adam_beta2": 0.999,
    "adam_epsilon": 1e-08,
    "max_grad_norm": 1.0,
    "num_train_epochs": 12,
    "max_steps": 1800,
    "lr_scheduler_type": "linear",
    "lr_scheduler_kwargs": {},
    "warmup_ratio": 0.03,
    "warmup_steps": 0,
    "log_level": "passive",
    "log_level_replica": "warning",
    "log_on_each_node": true,
    "logging_dir": "/hkfs/work/workspace/scratch/tum_fmp0582-dndworkspace/自己训练lora/train_lora/logs/ToT-Math-V1/group_03",
    "logging_strategy": "steps",
    "logging_first_step": true,
    "logging_steps": 5,
    "logging_nan_inf_filter": true,
    "save_strategy": "steps",
    "save_steps": 300,
    "save_total_limit": 6,
    "save_safetensors": true,
    "save_on_each_node": false,
    "save_only_model": false,
    "restore_callback_states_from_checkpoint": false,
    "no_cuda": false,
    "use_cpu": false,
    "use_mps_device": false,
    "seed": 42,
    "data_seed": null,
    "jit_mode_eval": false,
    "bf16": true,
    "fp16": false,
    "fp16_opt_level": "O1",
    "half_precision_backend": "auto",
    "bf16_full_eval": false,
    "fp16_full_eval": false,
    "tf32": null,
    "local_rank": 0,
    "ddp_backend": null,
    "tpu_num_cores": null,
    "tpu_metrics_debug": false,
    "debug": [],
    "dataloader_drop_last": false,
    "eval_steps": null,
    "dataloader_num_workers": 0,
    "dataloader_prefetch_factor": null,
    "past_index": -1,
    "run_name": null,
    "disable_tqdm": false,
    "remove_unused_columns": true,
    "label_names": null,
    "load_best_model_at_end": false,
    "metric_for_best_model": null,
    "greater_is_better": null,
    "ignore_data_skip": false,
    "fsdp": [],
    "fsdp_min_num_params": 0,
    "fsdp_config": {
      "min_num_params": 0,
      "xla": false,
      "xla_fsdp_v2": false,
      "xla_fsdp_grad_ckpt": false
    },
    "fsdp_transformer_layer_cls_to_wrap": null,
    "accelerator_config": {
      "split_batches": false,
      "dispatch_batches": null,
      "even_batches": true,
      "use_seedable_sampler": true,
      "non_blocking": false,
      "gradient_accumulation_kwargs": null
    },
    "parallelism_config": null,
    "deepspeed": null,
    "label_smoothing_factor": 0.0,
    "optim": "adamw_torch",
    "optim_args": null,
    "adafactor": false,
    "group_by_length": false,
    "length_column_name": "length",
    "report_to": [],
    "project": "huggingface",
    "trackio_space_id": "trackio",
    "ddp_find_unused_parameters": null,
    "ddp_bucket_cap_mb": null,
    "ddp_broadcast_buffers": null,
    "dataloader_pin_memory": true,
    "dataloader_persistent_workers": false,
    "skip_memory_metrics": true,
    "use_legacy_prediction_loop": false,
    "push_to_hub": false,
    "resume_from_checkpoint": null,
    "hub_model_id": null,
    "hub_strategy": "every_save",
    "hub_token": "<HUB_TOKEN>",
    "hub_private_repo": null,
    "hub_always_push": false,
    "hub_revision": null,
    "gradient_checkpointing": true,
    "gradient_checkpointing_kwargs": null,
    "include_inputs_for_metrics": false,
    "include_for_metrics": [],
    "eval_do_concat_batches": true,
    "fp16_backend": "auto",
    "push_to_hub_model_id": null,
    "push_to_hub_organization": null,
    "push_to_hub_token": "<PUSH_TO_HUB_TOKEN>",
    "mp_parameters": "",
    "auto_find_batch_size": false,
    "full_determinism": false,
    "torchdynamo": null,
    "ray_scope": "last",
    "ddp_timeout": 1800,
    "torch_compile": false,
    "torch_compile_backend": null,
    "torch_compile_mode": null,
    "include_tokens_per_second": false,
    "include_num_input_tokens_seen": "no",
    "neftune_noise_alpha": null,
    "optim_target_modules": null,
    "batch_eval_metrics": false,
    "eval_on_start": false,
    "use_liger_kernel": false,
    "liger_kernel_config": null,
    "eval_use_gather_object": false,
    "average_tokens_across_devices": true
  },
  "lora_config": {
    "r": 64,
    "alpha": 128,
    "dropout": 0.05,
    "target_modules": [
      "q_proj",
      "k_proj",
      "v_proj",
      "o_proj",
      "gate_proj",
      "up_proj",
      "down_proj"
    ]
  },
  "effective_batch_size": 32,
  "world_size": 1,
  "git_commit": ""
}