{
  "best_metric": 0.23128265142440796,
  "best_model_checkpoint": "finetuned-indian-food/checkpoint-1100",
  "epoch": 4.0,
  "eval_steps": 100,
  "global_step": 1332,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 0.0001984984984984985,
      "loss": 0.1875,
      "step": 10
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019699699699699701,
      "loss": 0.3897,
      "step": 20
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001954954954954955,
      "loss": 0.4291,
      "step": 30
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019399399399399402,
      "loss": 0.5206,
      "step": 40
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001924924924924925,
      "loss": 0.2651,
      "step": 50
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.000190990990990991,
      "loss": 0.5324,
      "step": 60
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001894894894894895,
      "loss": 0.4103,
      "step": 70
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.000187987987987988,
      "loss": 0.5292,
      "step": 80
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001864864864864865,
      "loss": 0.4953,
      "step": 90
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.000184984984984985,
      "loss": 0.4578,
      "step": 100
    },
    {
      "epoch": 0.3,
      "eval_accuracy": 0.8841657810839533,
      "eval_loss": 0.3982384204864502,
      "eval_runtime": 30.6948,
      "eval_samples_per_second": 30.657,
      "eval_steps_per_second": 3.844,
      "step": 100
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001834834834834835,
      "loss": 0.274,
      "step": 110
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.000181981981981982,
      "loss": 0.3558,
      "step": 120
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001804804804804805,
      "loss": 0.5567,
      "step": 130
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00017897897897897898,
      "loss": 0.4687,
      "step": 140
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001774774774774775,
      "loss": 0.4152,
      "step": 150
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.000175975975975976,
      "loss": 0.3191,
      "step": 160
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001744744744744745,
      "loss": 0.2954,
      "step": 170
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.000172972972972973,
      "loss": 0.3248,
      "step": 180
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00017147147147147148,
      "loss": 0.3865,
      "step": 190
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00016996996996997,
      "loss": 0.3823,
      "step": 200
    },
    {
      "epoch": 0.6,
      "eval_accuracy": 0.8862911795961743,
      "eval_loss": 0.4436054527759552,
      "eval_runtime": 30.5998,
      "eval_samples_per_second": 30.752,
      "eval_steps_per_second": 3.856,
      "step": 200
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00016846846846846846,
      "loss": 0.4138,
      "step": 210
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00016696696696696697,
      "loss": 0.4523,
      "step": 220
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00016546546546546546,
      "loss": 0.4075,
      "step": 230
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00016396396396396395,
      "loss": 0.3963,
      "step": 240
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00016246246246246247,
      "loss": 0.3831,
      "step": 250
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00016096096096096096,
      "loss": 0.4096,
      "step": 260
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00015945945945945947,
      "loss": 0.306,
      "step": 270
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00015795795795795796,
      "loss": 0.357,
      "step": 280
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00015645645645645645,
      "loss": 0.3535,
      "step": 290
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00015495495495495496,
      "loss": 0.4317,
      "step": 300
    },
    {
      "epoch": 0.9,
      "eval_accuracy": 0.8820403825717322,
      "eval_loss": 0.40273454785346985,
      "eval_runtime": 29.8704,
      "eval_samples_per_second": 31.503,
      "eval_steps_per_second": 3.95,
      "step": 300
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00015345345345345345,
      "loss": 0.3328,
      "step": 310
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00015210210210210212,
      "loss": 0.4023,
      "step": 320
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001506006006006006,
      "loss": 0.3135,
      "step": 330
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00014909909909909913,
      "loss": 0.3583,
      "step": 340
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00014759759759759762,
      "loss": 0.347,
      "step": 350
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.0001460960960960961,
      "loss": 0.1753,
      "step": 360
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00014459459459459462,
      "loss": 0.3504,
      "step": 370
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00014309309309309308,
      "loss": 0.4404,
      "step": 380
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001415915915915916,
      "loss": 0.2829,
      "step": 390
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001400900900900901,
      "loss": 0.3051,
      "step": 400
    },
    {
      "epoch": 1.2,
      "eval_accuracy": 0.8894792773645058,
      "eval_loss": 0.3761645257472992,
      "eval_runtime": 30.8085,
      "eval_samples_per_second": 30.543,
      "eval_steps_per_second": 3.83,
      "step": 400
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00013858858858858858,
      "loss": 0.2375,
      "step": 410
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001370870870870871,
      "loss": 0.3325,
      "step": 420
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00013558558558558558,
      "loss": 0.3024,
      "step": 430
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001340840840840841,
      "loss": 0.3907,
      "step": 440
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00013258258258258258,
      "loss": 0.3148,
      "step": 450
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00013108108108108107,
      "loss": 0.4637,
      "step": 460
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001295795795795796,
      "loss": 0.3781,
      "step": 470
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00012807807807807808,
      "loss": 0.3198,
      "step": 480
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00012657657657657657,
      "loss": 0.238,
      "step": 490
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00012507507507507508,
      "loss": 0.1829,
      "step": 500
    },
    {
      "epoch": 1.5,
      "eval_accuracy": 0.9086078639744952,
      "eval_loss": 0.36793917417526245,
      "eval_runtime": 32.3497,
      "eval_samples_per_second": 29.088,
      "eval_steps_per_second": 3.648,
      "step": 500
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00012357357357357357,
      "loss": 0.3644,
      "step": 510
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00012207207207207209,
      "loss": 0.3944,
      "step": 520
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00012057057057057057,
      "loss": 0.2006,
      "step": 530
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00011906906906906908,
      "loss": 0.3097,
      "step": 540
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00011756756756756758,
      "loss": 0.3852,
      "step": 550
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00011606606606606607,
      "loss": 0.2087,
      "step": 560
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00011456456456456457,
      "loss": 0.1634,
      "step": 570
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00011306306306306307,
      "loss": 0.3495,
      "step": 580
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011156156156156157,
      "loss": 0.2408,
      "step": 590
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011006006006006006,
      "loss": 0.2193,
      "step": 600
    },
    {
      "epoch": 1.8,
      "eval_accuracy": 0.9234856535600425,
      "eval_loss": 0.30461981892585754,
      "eval_runtime": 29.8181,
      "eval_samples_per_second": 31.558,
      "eval_steps_per_second": 3.957,
      "step": 600
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00010855855855855856,
      "loss": 0.2662,
      "step": 610
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00010705705705705707,
      "loss": 0.2284,
      "step": 620
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00010555555555555557,
      "loss": 0.2535,
      "step": 630
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00010405405405405406,
      "loss": 0.1944,
      "step": 640
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00010255255255255256,
      "loss": 0.3346,
      "step": 650
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00010105105105105106,
      "loss": 0.3154,
      "step": 660
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.954954954954956e-05,
      "loss": 0.2668,
      "step": 670
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.804804804804806e-05,
      "loss": 0.308,
      "step": 680
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.654654654654654e-05,
      "loss": 0.2409,
      "step": 690
    },
    {
      "epoch": 2.1,
      "learning_rate": 9.504504504504504e-05,
      "loss": 0.1673,
      "step": 700
    },
    {
      "epoch": 2.1,
      "eval_accuracy": 0.922422954303932,
      "eval_loss": 0.3170412480831146,
      "eval_runtime": 30.9938,
      "eval_samples_per_second": 30.361,
      "eval_steps_per_second": 3.807,
      "step": 700
    },
    {
      "epoch": 2.13,
      "learning_rate": 9.354354354354354e-05,
      "loss": 0.3351,
      "step": 710
    },
    {
      "epoch": 2.16,
      "learning_rate": 9.204204204204205e-05,
      "loss": 0.2524,
      "step": 720
    },
    {
      "epoch": 2.19,
      "learning_rate": 9.054054054054055e-05,
      "loss": 0.168,
      "step": 730
    },
    {
      "epoch": 2.22,
      "learning_rate": 8.903903903903904e-05,
      "loss": 0.1613,
      "step": 740
    },
    {
      "epoch": 2.25,
      "learning_rate": 8.753753753753754e-05,
      "loss": 0.2485,
      "step": 750
    },
    {
      "epoch": 2.28,
      "learning_rate": 8.603603603603604e-05,
      "loss": 0.2042,
      "step": 760
    },
    {
      "epoch": 2.31,
      "learning_rate": 8.453453453453454e-05,
      "loss": 0.2449,
      "step": 770
    },
    {
      "epoch": 2.34,
      "learning_rate": 8.303303303303303e-05,
      "loss": 0.2349,
      "step": 780
    },
    {
      "epoch": 2.37,
      "learning_rate": 8.153153153153153e-05,
      "loss": 0.1965,
      "step": 790
    },
    {
      "epoch": 2.4,
      "learning_rate": 8.003003003003004e-05,
      "loss": 0.2694,
      "step": 800
    },
    {
      "epoch": 2.4,
      "eval_accuracy": 0.9341126461211477,
      "eval_loss": 0.2725915014743805,
      "eval_runtime": 32.8131,
      "eval_samples_per_second": 28.678,
      "eval_steps_per_second": 3.596,
      "step": 800
    },
    {
      "epoch": 2.43,
      "learning_rate": 7.852852852852854e-05,
      "loss": 0.1043,
      "step": 810
    },
    {
      "epoch": 2.46,
      "learning_rate": 7.702702702702703e-05,
      "loss": 0.2578,
      "step": 820
    },
    {
      "epoch": 2.49,
      "learning_rate": 7.552552552552553e-05,
      "loss": 0.1695,
      "step": 830
    },
    {
      "epoch": 2.52,
      "learning_rate": 7.402402402402403e-05,
      "loss": 0.2135,
      "step": 840
    },
    {
      "epoch": 2.55,
      "learning_rate": 7.252252252252253e-05,
      "loss": 0.2322,
      "step": 850
    },
    {
      "epoch": 2.58,
      "learning_rate": 7.102102102102103e-05,
      "loss": 0.318,
      "step": 860
    },
    {
      "epoch": 2.61,
      "learning_rate": 6.951951951951952e-05,
      "loss": 0.2602,
      "step": 870
    },
    {
      "epoch": 2.64,
      "learning_rate": 6.801801801801802e-05,
      "loss": 0.2277,
      "step": 880
    },
    {
      "epoch": 2.67,
      "learning_rate": 6.651651651651653e-05,
      "loss": 0.1575,
      "step": 890
    },
    {
      "epoch": 2.7,
      "learning_rate": 6.501501501501502e-05,
      "loss": 0.1209,
      "step": 900
    },
    {
      "epoch": 2.7,
      "eval_accuracy": 0.9287991498405951,
      "eval_loss": 0.2777404189109802,
      "eval_runtime": 29.9332,
      "eval_samples_per_second": 31.437,
      "eval_steps_per_second": 3.942,
      "step": 900
    },
    {
      "epoch": 2.73,
      "learning_rate": 6.351351351351352e-05,
      "loss": 0.312,
      "step": 910
    },
    {
      "epoch": 2.76,
      "learning_rate": 6.2012012012012e-05,
      "loss": 0.1441,
      "step": 920
    },
    {
      "epoch": 2.79,
      "learning_rate": 6.051051051051051e-05,
      "loss": 0.1176,
      "step": 930
    },
    {
      "epoch": 2.82,
      "learning_rate": 5.900900900900901e-05,
      "loss": 0.1827,
      "step": 940
    },
    {
      "epoch": 2.85,
      "learning_rate": 5.7507507507507506e-05,
      "loss": 0.1594,
      "step": 950
    },
    {
      "epoch": 2.88,
      "learning_rate": 5.600600600600601e-05,
      "loss": 0.1515,
      "step": 960
    },
    {
      "epoch": 2.91,
      "learning_rate": 5.45045045045045e-05,
      "loss": 0.1662,
      "step": 970
    },
    {
      "epoch": 2.94,
      "learning_rate": 5.3003003003003005e-05,
      "loss": 0.2344,
      "step": 980
    },
    {
      "epoch": 2.97,
      "learning_rate": 5.15015015015015e-05,
      "loss": 0.2594,
      "step": 990
    },
    {
      "epoch": 3.0,
      "learning_rate": 5e-05,
      "loss": 0.146,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9383634431455898,
      "eval_loss": 0.24151237308979034,
      "eval_runtime": 30.6134,
      "eval_samples_per_second": 30.738,
      "eval_steps_per_second": 3.855,
      "step": 1000
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.8498498498498504e-05,
      "loss": 0.1399,
      "step": 1010
    },
    {
      "epoch": 3.06,
      "learning_rate": 4.6996996996997e-05,
      "loss": 0.1721,
      "step": 1020
    },
    {
      "epoch": 3.09,
      "learning_rate": 4.54954954954955e-05,
      "loss": 0.1436,
      "step": 1030
    },
    {
      "epoch": 3.12,
      "learning_rate": 4.3993993993994e-05,
      "loss": 0.1062,
      "step": 1040
    },
    {
      "epoch": 3.15,
      "learning_rate": 4.24924924924925e-05,
      "loss": 0.2072,
      "step": 1050
    },
    {
      "epoch": 3.18,
      "learning_rate": 4.099099099099099e-05,
      "loss": 0.1283,
      "step": 1060
    },
    {
      "epoch": 3.21,
      "learning_rate": 3.948948948948949e-05,
      "loss": 0.1524,
      "step": 1070
    },
    {
      "epoch": 3.24,
      "learning_rate": 3.7987987987987985e-05,
      "loss": 0.103,
      "step": 1080
    },
    {
      "epoch": 3.27,
      "learning_rate": 3.648648648648649e-05,
      "loss": 0.0162,
      "step": 1090
    },
    {
      "epoch": 3.3,
      "learning_rate": 3.498498498498499e-05,
      "loss": 0.1515,
      "step": 1100
    },
    {
      "epoch": 3.3,
      "eval_accuracy": 0.9458023379383634,
      "eval_loss": 0.23128265142440796,
      "eval_runtime": 30.6712,
      "eval_samples_per_second": 30.68,
      "eval_steps_per_second": 3.847,
      "step": 1100
    },
    {
      "epoch": 3.33,
      "learning_rate": 3.3483483483483485e-05,
      "loss": 0.0923,
      "step": 1110
    },
    {
      "epoch": 3.36,
      "learning_rate": 3.198198198198199e-05,
      "loss": 0.1725,
      "step": 1120
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.0480480480480482e-05,
      "loss": 0.1219,
      "step": 1130
    },
    {
      "epoch": 3.42,
      "learning_rate": 2.897897897897898e-05,
      "loss": 0.1848,
      "step": 1140
    },
    {
      "epoch": 3.45,
      "learning_rate": 2.7477477477477483e-05,
      "loss": 0.1743,
      "step": 1150
    },
    {
      "epoch": 3.48,
      "learning_rate": 2.5975975975975975e-05,
      "loss": 0.1848,
      "step": 1160
    },
    {
      "epoch": 3.51,
      "learning_rate": 2.4474474474474477e-05,
      "loss": 0.0712,
      "step": 1170
    },
    {
      "epoch": 3.54,
      "learning_rate": 2.2972972972972976e-05,
      "loss": 0.0939,
      "step": 1180
    },
    {
      "epoch": 3.57,
      "learning_rate": 2.147147147147147e-05,
      "loss": 0.0743,
      "step": 1190
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.996996996996997e-05,
      "loss": 0.1645,
      "step": 1200
    },
    {
      "epoch": 3.6,
      "eval_accuracy": 0.9436769394261424,
      "eval_loss": 0.23942747712135315,
      "eval_runtime": 30.926,
      "eval_samples_per_second": 30.427,
      "eval_steps_per_second": 3.816,
      "step": 1200
    },
    {
      "epoch": 3.63,
      "learning_rate": 1.846846846846847e-05,
      "loss": 0.0818,
      "step": 1210
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.6966966966966967e-05,
      "loss": 0.1648,
      "step": 1220
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.5465465465465466e-05,
      "loss": 0.1718,
      "step": 1230
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.3963963963963963e-05,
      "loss": 0.171,
      "step": 1240
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.2462462462462463e-05,
      "loss": 0.2079,
      "step": 1250
    },
    {
      "epoch": 3.78,
      "learning_rate": 1.0960960960960962e-05,
      "loss": 0.1947,
      "step": 1260
    },
    {
      "epoch": 3.81,
      "learning_rate": 9.45945945945946e-06,
      "loss": 0.1673,
      "step": 1270
    },
    {
      "epoch": 3.84,
      "learning_rate": 7.957957957957958e-06,
      "loss": 0.1035,
      "step": 1280
    },
    {
      "epoch": 3.87,
      "learning_rate": 6.456456456456457e-06,
      "loss": 0.0789,
      "step": 1290
    },
    {
      "epoch": 3.9,
      "learning_rate": 4.954954954954955e-06,
      "loss": 0.1142,
      "step": 1300
    },
    {
      "epoch": 3.9,
      "eval_accuracy": 0.944739638682253,
      "eval_loss": 0.23253901302814484,
      "eval_runtime": 30.782,
      "eval_samples_per_second": 30.57,
      "eval_steps_per_second": 3.833,
      "step": 1300
    },
    {
      "epoch": 3.93,
      "learning_rate": 3.4534534534534537e-06,
      "loss": 0.2177,
      "step": 1310
    },
    {
      "epoch": 3.96,
      "learning_rate": 1.951951951951952e-06,
      "loss": 0.091,
      "step": 1320
    },
    {
      "epoch": 3.99,
      "learning_rate": 4.504504504504505e-07,
      "loss": 0.0961,
      "step": 1330
    },
    {
      "epoch": 4.0,
      "step": 1332,
      "total_flos": 1.651775717862015e+18,
      "train_loss": 0.25963506532144975,
      "train_runtime": 1274.7645,
      "train_samples_per_second": 16.718,
      "train_steps_per_second": 1.045
    }
  ],
  "logging_steps": 10,
  "max_steps": 1332,
  "num_train_epochs": 4,
  "save_steps": 100,
  "total_flos": 1.651775717862015e+18,
  "trial_name": null,
  "trial_params": null
}