{
  "best_metric": 0.08658882975578308,
  "best_model_checkpoint": "./outputs/checkpoint-1405",
  "epoch": 5.0,
  "global_step": 1405,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 1.9857651245551603e-05,
      "loss": 0.6493,
      "step": 10
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.9715302491103204e-05,
      "loss": 0.5006,
      "step": 20
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.9572953736654805e-05,
      "loss": 0.4339,
      "step": 30
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.9430604982206406e-05,
      "loss": 0.3415,
      "step": 40
    },
    {
      "epoch": 0.18,
      "learning_rate": 1.9288256227758007e-05,
      "loss": 0.3085,
      "step": 50
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.9145907473309612e-05,
      "loss": 0.2519,
      "step": 60
    },
    {
      "epoch": 0.25,
      "learning_rate": 1.9003558718861213e-05,
      "loss": 0.2041,
      "step": 70
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.886120996441281e-05,
      "loss": 0.267,
      "step": 80
    },
    {
      "epoch": 0.32,
      "learning_rate": 1.8718861209964415e-05,
      "loss": 0.1472,
      "step": 90
    },
    {
      "epoch": 0.36,
      "learning_rate": 1.8576512455516017e-05,
      "loss": 0.1345,
      "step": 100
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.8434163701067618e-05,
      "loss": 0.1452,
      "step": 110
    },
    {
      "epoch": 0.43,
      "learning_rate": 1.829181494661922e-05,
      "loss": 0.148,
      "step": 120
    },
    {
      "epoch": 0.46,
      "learning_rate": 1.814946619217082e-05,
      "loss": 0.0543,
      "step": 130
    },
    {
      "epoch": 0.5,
      "learning_rate": 1.800711743772242e-05,
      "loss": 0.118,
      "step": 140
    },
    {
      "epoch": 0.53,
      "learning_rate": 1.7864768683274022e-05,
      "loss": 0.1558,
      "step": 150
    },
    {
      "epoch": 0.57,
      "learning_rate": 1.7722419928825624e-05,
      "loss": 0.0911,
      "step": 160
    },
    {
      "epoch": 0.6,
      "learning_rate": 1.7580071174377225e-05,
      "loss": 0.1144,
      "step": 170
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7437722419928826e-05,
      "loss": 0.1203,
      "step": 180
    },
    {
      "epoch": 0.68,
      "learning_rate": 1.7295373665480427e-05,
      "loss": 0.0681,
      "step": 190
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.715302491103203e-05,
      "loss": 0.0949,
      "step": 200
    },
    {
      "epoch": 0.75,
      "learning_rate": 1.7010676156583633e-05,
      "loss": 0.0766,
      "step": 210
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.686832740213523e-05,
      "loss": 0.1094,
      "step": 220
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.6725978647686835e-05,
      "loss": 0.2055,
      "step": 230
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.6583629893238436e-05,
      "loss": 0.1068,
      "step": 240
    },
    {
      "epoch": 0.89,
      "learning_rate": 1.6441281138790037e-05,
      "loss": 0.0884,
      "step": 250
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.629893238434164e-05,
      "loss": 0.0776,
      "step": 260
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.615658362989324e-05,
      "loss": 0.0423,
      "step": 270
    },
    {
      "epoch": 1.0,
      "learning_rate": 1.601423487544484e-05,
      "loss": 0.0246,
      "step": 280
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9773299748110831,
      "eval_loss": 0.10174643248319626,
      "eval_runtime": 74.1,
      "eval_samples_per_second": 5.358,
      "eval_steps_per_second": 0.675,
      "step": 281
    },
    {
      "epoch": 1.03,
      "learning_rate": 1.5871886120996442e-05,
      "loss": 0.0418,
      "step": 290
    },
    {
      "epoch": 1.07,
      "learning_rate": 1.5729537366548043e-05,
      "loss": 0.0408,
      "step": 300
    },
    {
      "epoch": 1.1,
      "learning_rate": 1.5587188612099648e-05,
      "loss": 0.0672,
      "step": 310
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.5444839857651245e-05,
      "loss": 0.1029,
      "step": 320
    },
    {
      "epoch": 1.17,
      "learning_rate": 1.5302491103202847e-05,
      "loss": 0.1104,
      "step": 330
    },
    {
      "epoch": 1.21,
      "learning_rate": 1.516014234875445e-05,
      "loss": 0.1299,
      "step": 340
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.5017793594306052e-05,
      "loss": 0.016,
      "step": 350
    },
    {
      "epoch": 1.28,
      "learning_rate": 1.4875444839857654e-05,
      "loss": 0.0661,
      "step": 360
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.4733096085409253e-05,
      "loss": 0.0437,
      "step": 370
    },
    {
      "epoch": 1.35,
      "learning_rate": 1.4590747330960854e-05,
      "loss": 0.0358,
      "step": 380
    },
    {
      "epoch": 1.39,
      "learning_rate": 1.4448398576512457e-05,
      "loss": 0.0846,
      "step": 390
    },
    {
      "epoch": 1.42,
      "learning_rate": 1.4306049822064058e-05,
      "loss": 0.1405,
      "step": 400
    },
    {
      "epoch": 1.46,
      "learning_rate": 1.4163701067615661e-05,
      "loss": 0.0524,
      "step": 410
    },
    {
      "epoch": 1.49,
      "learning_rate": 1.402135231316726e-05,
      "loss": 0.0691,
      "step": 420
    },
    {
      "epoch": 1.53,
      "learning_rate": 1.3879003558718862e-05,
      "loss": 0.0404,
      "step": 430
    },
    {
      "epoch": 1.57,
      "learning_rate": 1.3736654804270464e-05,
      "loss": 0.0239,
      "step": 440
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.3594306049822066e-05,
      "loss": 0.092,
      "step": 450
    },
    {
      "epoch": 1.64,
      "learning_rate": 1.3451957295373668e-05,
      "loss": 0.0501,
      "step": 460
    },
    {
      "epoch": 1.67,
      "learning_rate": 1.3309608540925268e-05,
      "loss": 0.0662,
      "step": 470
    },
    {
      "epoch": 1.71,
      "learning_rate": 1.3167259786476869e-05,
      "loss": 0.097,
      "step": 480
    },
    {
      "epoch": 1.74,
      "learning_rate": 1.302491103202847e-05,
      "loss": 0.0362,
      "step": 490
    },
    {
      "epoch": 1.78,
      "learning_rate": 1.2882562277580073e-05,
      "loss": 0.048,
      "step": 500
    },
    {
      "epoch": 1.81,
      "learning_rate": 1.2740213523131673e-05,
      "loss": 0.0757,
      "step": 510
    },
    {
      "epoch": 1.85,
      "learning_rate": 1.2597864768683274e-05,
      "loss": 0.0647,
      "step": 520
    },
    {
      "epoch": 1.89,
      "learning_rate": 1.2455516014234877e-05,
      "loss": 0.0318,
      "step": 530
    },
    {
      "epoch": 1.92,
      "learning_rate": 1.2313167259786478e-05,
      "loss": 0.0468,
      "step": 540
    },
    {
      "epoch": 1.96,
      "learning_rate": 1.217081850533808e-05,
      "loss": 0.0147,
      "step": 550
    },
    {
      "epoch": 1.99,
      "learning_rate": 1.202846975088968e-05,
      "loss": 0.0223,
      "step": 560
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9722921914357683,
      "eval_loss": 0.11001072078943253,
      "eval_runtime": 73.9606,
      "eval_samples_per_second": 5.368,
      "eval_steps_per_second": 0.676,
      "step": 562
    },
    {
      "epoch": 2.03,
      "learning_rate": 1.1886120996441281e-05,
      "loss": 0.1322,
      "step": 570
    },
    {
      "epoch": 2.06,
      "learning_rate": 1.1743772241992884e-05,
      "loss": 0.0655,
      "step": 580
    },
    {
      "epoch": 2.1,
      "learning_rate": 1.1601423487544485e-05,
      "loss": 0.0678,
      "step": 590
    },
    {
      "epoch": 2.14,
      "learning_rate": 1.1459074733096086e-05,
      "loss": 0.1083,
      "step": 600
    },
    {
      "epoch": 2.17,
      "learning_rate": 1.1316725978647688e-05,
      "loss": 0.036,
      "step": 610
    },
    {
      "epoch": 2.21,
      "learning_rate": 1.1174377224199289e-05,
      "loss": 0.0662,
      "step": 620
    },
    {
      "epoch": 2.24,
      "learning_rate": 1.103202846975089e-05,
      "loss": 0.078,
      "step": 630
    },
    {
      "epoch": 2.28,
      "learning_rate": 1.0889679715302493e-05,
      "loss": 0.0356,
      "step": 640
    },
    {
      "epoch": 2.31,
      "learning_rate": 1.0747330960854094e-05,
      "loss": 0.061,
      "step": 650
    },
    {
      "epoch": 2.35,
      "learning_rate": 1.0604982206405693e-05,
      "loss": 0.0275,
      "step": 660
    },
    {
      "epoch": 2.38,
      "learning_rate": 1.0462633451957296e-05,
      "loss": 0.0323,
      "step": 670
    },
    {
      "epoch": 2.42,
      "learning_rate": 1.0320284697508897e-05,
      "loss": 0.0808,
      "step": 680
    },
    {
      "epoch": 2.46,
      "learning_rate": 1.01779359430605e-05,
      "loss": 0.0395,
      "step": 690
    },
    {
      "epoch": 2.49,
      "learning_rate": 1.0035587188612101e-05,
      "loss": 0.0222,
      "step": 700
    },
    {
      "epoch": 2.53,
      "learning_rate": 9.893238434163703e-06,
      "loss": 0.0586,
      "step": 710
    },
    {
      "epoch": 2.56,
      "learning_rate": 9.750889679715304e-06,
      "loss": 0.0218,
      "step": 720
    },
    {
      "epoch": 2.6,
      "learning_rate": 9.608540925266905e-06,
      "loss": 0.0297,
      "step": 730
    },
    {
      "epoch": 2.63,
      "learning_rate": 9.466192170818506e-06,
      "loss": 0.0207,
      "step": 740
    },
    {
      "epoch": 2.67,
      "learning_rate": 9.323843416370107e-06,
      "loss": 0.1084,
      "step": 750
    },
    {
      "epoch": 2.7,
      "learning_rate": 9.18149466192171e-06,
      "loss": 0.0212,
      "step": 760
    },
    {
      "epoch": 2.74,
      "learning_rate": 9.03914590747331e-06,
      "loss": 0.0168,
      "step": 770
    },
    {
      "epoch": 2.78,
      "learning_rate": 8.896797153024912e-06,
      "loss": 0.0218,
      "step": 780
    },
    {
      "epoch": 2.81,
      "learning_rate": 8.754448398576513e-06,
      "loss": 0.0823,
      "step": 790
    },
    {
      "epoch": 2.85,
      "learning_rate": 8.612099644128115e-06,
      "loss": 0.0081,
      "step": 800
    },
    {
      "epoch": 2.88,
      "learning_rate": 8.469750889679716e-06,
      "loss": 0.0318,
      "step": 810
    },
    {
      "epoch": 2.92,
      "learning_rate": 8.327402135231317e-06,
      "loss": 0.0198,
      "step": 820
    },
    {
      "epoch": 2.95,
      "learning_rate": 8.18505338078292e-06,
      "loss": 0.0221,
      "step": 830
    },
    {
      "epoch": 2.99,
      "learning_rate": 8.04270462633452e-06,
      "loss": 0.061,
      "step": 840
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9798488664987406,
      "eval_loss": 0.08890923857688904,
      "eval_runtime": 69.7679,
      "eval_samples_per_second": 5.69,
      "eval_steps_per_second": 0.717,
      "step": 843
    },
    {
      "epoch": 3.02,
      "learning_rate": 7.900355871886122e-06,
      "loss": 0.0123,
      "step": 850
    },
    {
      "epoch": 3.06,
      "learning_rate": 7.758007117437723e-06,
      "loss": 0.0852,
      "step": 860
    },
    {
      "epoch": 3.1,
      "learning_rate": 7.6156583629893245e-06,
      "loss": 0.0074,
      "step": 870
    },
    {
      "epoch": 3.13,
      "learning_rate": 7.4733096085409265e-06,
      "loss": 0.0131,
      "step": 880
    },
    {
      "epoch": 3.17,
      "learning_rate": 7.330960854092527e-06,
      "loss": 0.0604,
      "step": 890
    },
    {
      "epoch": 3.2,
      "learning_rate": 7.188612099644129e-06,
      "loss": 0.0673,
      "step": 900
    },
    {
      "epoch": 3.24,
      "learning_rate": 7.04626334519573e-06,
      "loss": 0.0774,
      "step": 910
    },
    {
      "epoch": 3.27,
      "learning_rate": 6.903914590747331e-06,
      "loss": 0.0152,
      "step": 920
    },
    {
      "epoch": 3.31,
      "learning_rate": 6.761565836298933e-06,
      "loss": 0.0277,
      "step": 930
    },
    {
      "epoch": 3.35,
      "learning_rate": 6.619217081850534e-06,
      "loss": 0.0374,
      "step": 940
    },
    {
      "epoch": 3.38,
      "learning_rate": 6.476868327402136e-06,
      "loss": 0.0648,
      "step": 950
    },
    {
      "epoch": 3.42,
      "learning_rate": 6.334519572953737e-06,
      "loss": 0.0451,
      "step": 960
    },
    {
      "epoch": 3.45,
      "learning_rate": 6.192170818505339e-06,
      "loss": 0.027,
      "step": 970
    },
    {
      "epoch": 3.49,
      "learning_rate": 6.049822064056941e-06,
      "loss": 0.0169,
      "step": 980
    },
    {
      "epoch": 3.52,
      "learning_rate": 5.907473309608541e-06,
      "loss": 0.0394,
      "step": 990
    },
    {
      "epoch": 3.56,
      "learning_rate": 5.765124555160143e-06,
      "loss": 0.0314,
      "step": 1000
    },
    {
      "epoch": 3.59,
      "learning_rate": 5.622775800711744e-06,
      "loss": 0.0164,
      "step": 1010
    },
    {
      "epoch": 3.63,
      "learning_rate": 5.480427046263346e-06,
      "loss": 0.0481,
      "step": 1020
    },
    {
      "epoch": 3.67,
      "learning_rate": 5.338078291814946e-06,
      "loss": 0.0663,
      "step": 1030
    },
    {
      "epoch": 3.7,
      "learning_rate": 5.195729537366548e-06,
      "loss": 0.032,
      "step": 1040
    },
    {
      "epoch": 3.74,
      "learning_rate": 5.05338078291815e-06,
      "loss": 0.0067,
      "step": 1050
    },
    {
      "epoch": 3.77,
      "learning_rate": 4.911032028469751e-06,
      "loss": 0.0181,
      "step": 1060
    },
    {
      "epoch": 3.81,
      "learning_rate": 4.768683274021353e-06,
      "loss": 0.1272,
      "step": 1070
    },
    {
      "epoch": 3.84,
      "learning_rate": 4.626334519572954e-06,
      "loss": 0.0469,
      "step": 1080
    },
    {
      "epoch": 3.88,
      "learning_rate": 4.483985765124556e-06,
      "loss": 0.1137,
      "step": 1090
    },
    {
      "epoch": 3.91,
      "learning_rate": 4.341637010676157e-06,
      "loss": 0.0623,
      "step": 1100
    },
    {
      "epoch": 3.95,
      "learning_rate": 4.199288256227758e-06,
      "loss": 0.0082,
      "step": 1110
    },
    {
      "epoch": 3.99,
      "learning_rate": 4.05693950177936e-06,
      "loss": 0.0143,
      "step": 1120
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.982367758186398,
      "eval_loss": 0.08782453835010529,
      "eval_runtime": 69.8578,
      "eval_samples_per_second": 5.683,
      "eval_steps_per_second": 0.716,
      "step": 1124
    },
    {
      "epoch": 4.02,
      "learning_rate": 3.914590747330961e-06,
      "loss": 0.0131,
      "step": 1130
    },
    {
      "epoch": 4.06,
      "learning_rate": 3.7722419928825625e-06,
      "loss": 0.0115,
      "step": 1140
    },
    {
      "epoch": 4.09,
      "learning_rate": 3.629893238434164e-06,
      "loss": 0.0457,
      "step": 1150
    },
    {
      "epoch": 4.13,
      "learning_rate": 3.4875444839857653e-06,
      "loss": 0.038,
      "step": 1160
    },
    {
      "epoch": 4.16,
      "learning_rate": 3.3451957295373664e-06,
      "loss": 0.0186,
      "step": 1170
    },
    {
      "epoch": 4.2,
      "learning_rate": 3.2028469750889684e-06,
      "loss": 0.0567,
      "step": 1180
    },
    {
      "epoch": 4.23,
      "learning_rate": 3.0604982206405696e-06,
      "loss": 0.0409,
      "step": 1190
    },
    {
      "epoch": 4.27,
      "learning_rate": 2.918149466192171e-06,
      "loss": 0.0451,
      "step": 1200
    },
    {
      "epoch": 4.31,
      "learning_rate": 2.7758007117437723e-06,
      "loss": 0.0123,
      "step": 1210
    },
    {
      "epoch": 4.34,
      "learning_rate": 2.633451957295374e-06,
      "loss": 0.0663,
      "step": 1220
    },
    {
      "epoch": 4.38,
      "learning_rate": 2.4911032028469755e-06,
      "loss": 0.0075,
      "step": 1230
    },
    {
      "epoch": 4.41,
      "learning_rate": 2.3487544483985766e-06,
      "loss": 0.0208,
      "step": 1240
    },
    {
      "epoch": 4.45,
      "learning_rate": 2.2064056939501782e-06,
      "loss": 0.0115,
      "step": 1250
    },
    {
      "epoch": 4.48,
      "learning_rate": 2.0640569395017794e-06,
      "loss": 0.086,
      "step": 1260
    },
    {
      "epoch": 4.52,
      "learning_rate": 1.921708185053381e-06,
      "loss": 0.018,
      "step": 1270
    },
    {
      "epoch": 4.56,
      "learning_rate": 1.7793594306049826e-06,
      "loss": 0.0301,
      "step": 1280
    },
    {
      "epoch": 4.59,
      "learning_rate": 1.6370106761565837e-06,
      "loss": 0.0122,
      "step": 1290
    },
    {
      "epoch": 4.63,
      "learning_rate": 1.494661921708185e-06,
      "loss": 0.0349,
      "step": 1300
    },
    {
      "epoch": 4.66,
      "learning_rate": 1.3523131672597867e-06,
      "loss": 0.0491,
      "step": 1310
    },
    {
      "epoch": 4.7,
      "learning_rate": 1.209964412811388e-06,
      "loss": 0.0686,
      "step": 1320
    },
    {
      "epoch": 4.73,
      "learning_rate": 1.0676156583629894e-06,
      "loss": 0.0279,
      "step": 1330
    },
    {
      "epoch": 4.77,
      "learning_rate": 9.252669039145908e-07,
      "loss": 0.0561,
      "step": 1340
    },
    {
      "epoch": 4.8,
      "learning_rate": 7.829181494661923e-07,
      "loss": 0.0062,
      "step": 1350
    },
    {
      "epoch": 4.84,
      "learning_rate": 6.405693950177936e-07,
      "loss": 0.0567,
      "step": 1360
    },
    {
      "epoch": 4.88,
      "learning_rate": 4.982206405693951e-07,
      "loss": 0.0763,
      "step": 1370
    },
    {
      "epoch": 4.91,
      "learning_rate": 3.558718861209965e-07,
      "loss": 0.026,
      "step": 1380
    },
    {
      "epoch": 4.95,
      "learning_rate": 2.135231316725979e-07,
      "loss": 0.0063,
      "step": 1390
    },
    {
      "epoch": 4.98,
      "learning_rate": 7.117437722419929e-08,
      "loss": 0.01,
      "step": 1400
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9798488664987406,
      "eval_loss": 0.08658882975578308,
      "eval_runtime": 74.0792,
      "eval_samples_per_second": 5.359,
      "eval_steps_per_second": 0.675,
      "step": 1405
    },
    {
      "epoch": 5.0,
      "step": 1405,
      "total_flos": 8.694601234757222e+17,
      "train_loss": 0.07352266346114386,
      "train_runtime": 6401.1868,
      "train_samples_per_second": 1.753,
      "train_steps_per_second": 0.219
    }
  ],
  "max_steps": 1405,
  "num_train_epochs": 5,
  "total_flos": 8.694601234757222e+17,
  "trial_name": null,
  "trial_params": null
}