{
  "best_metric": 0.7603686635944701,
  "best_model_checkpoint": "videomae-base-finetuned-subset\\checkpoint-2006",
  "epoch": 24.031794871794872,
  "eval_steps": 500,
  "global_step": 2925,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1.7064846416382253e-06,
      "loss": 1.5975,
      "step": 10
    },
    {
      "epoch": 0.01,
      "learning_rate": 3.4129692832764506e-06,
      "loss": 1.6186,
      "step": 20
    },
    {
      "epoch": 0.01,
      "learning_rate": 5.119453924914676e-06,
      "loss": 1.6342,
      "step": 30
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.825938566552901e-06,
      "loss": 1.623,
      "step": 40
    },
    {
      "epoch": 0.02,
      "learning_rate": 8.532423208191128e-06,
      "loss": 1.6342,
      "step": 50
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.0238907849829352e-05,
      "loss": 1.6521,
      "step": 60
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.1945392491467578e-05,
      "loss": 1.6252,
      "step": 70
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.3651877133105803e-05,
      "loss": 1.6544,
      "step": 80
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.535836177474403e-05,
      "loss": 1.598,
      "step": 90
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.7064846416382256e-05,
      "loss": 1.6387,
      "step": 100
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.877133105802048e-05,
      "loss": 1.6183,
      "step": 110
    },
    {
      "epoch": 0.04,
      "eval_accuracy": 0.41935483870967744,
      "eval_loss": 1.5575686693191528,
      "eval_runtime": 222.132,
      "eval_samples_per_second": 0.977,
      "eval_steps_per_second": 0.248,
      "step": 118
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.0477815699658705e-05,
      "loss": 1.6468,
      "step": 120
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.218430034129693e-05,
      "loss": 1.6418,
      "step": 130
    },
    {
      "epoch": 1.01,
      "learning_rate": 2.3890784982935157e-05,
      "loss": 1.5639,
      "step": 140
    },
    {
      "epoch": 1.01,
      "learning_rate": 2.559726962457338e-05,
      "loss": 1.5755,
      "step": 150
    },
    {
      "epoch": 1.01,
      "learning_rate": 2.7303754266211605e-05,
      "loss": 1.5607,
      "step": 160
    },
    {
      "epoch": 1.02,
      "learning_rate": 2.901023890784983e-05,
      "loss": 1.6052,
      "step": 170
    },
    {
      "epoch": 1.02,
      "learning_rate": 3.071672354948806e-05,
      "loss": 1.6605,
      "step": 180
    },
    {
      "epoch": 1.02,
      "learning_rate": 3.242320819112628e-05,
      "loss": 1.5728,
      "step": 190
    },
    {
      "epoch": 1.03,
      "learning_rate": 3.412969283276451e-05,
      "loss": 1.4505,
      "step": 200
    },
    {
      "epoch": 1.03,
      "learning_rate": 3.5836177474402735e-05,
      "loss": 1.3527,
      "step": 210
    },
    {
      "epoch": 1.03,
      "learning_rate": 3.754266211604096e-05,
      "loss": 1.3427,
      "step": 220
    },
    {
      "epoch": 1.04,
      "learning_rate": 3.924914675767919e-05,
      "loss": 1.4874,
      "step": 230
    },
    {
      "epoch": 1.04,
      "eval_accuracy": 0.423963133640553,
      "eval_loss": 1.3517247438430786,
      "eval_runtime": 217.4616,
      "eval_samples_per_second": 0.998,
      "eval_steps_per_second": 0.253,
      "step": 236
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.095563139931741e-05,
      "loss": 1.3605,
      "step": 240
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.266211604095564e-05,
      "loss": 1.4474,
      "step": 250
    },
    {
      "epoch": 2.01,
      "learning_rate": 4.436860068259386e-05,
      "loss": 1.5998,
      "step": 260
    },
    {
      "epoch": 2.01,
      "learning_rate": 4.6075085324232084e-05,
      "loss": 1.4064,
      "step": 270
    },
    {
      "epoch": 2.02,
      "learning_rate": 4.778156996587031e-05,
      "loss": 1.4201,
      "step": 280
    },
    {
      "epoch": 2.02,
      "learning_rate": 4.9488054607508536e-05,
      "loss": 1.252,
      "step": 290
    },
    {
      "epoch": 2.02,
      "learning_rate": 4.986702127659575e-05,
      "loss": 1.3703,
      "step": 300
    },
    {
      "epoch": 2.03,
      "learning_rate": 4.9677051671732525e-05,
      "loss": 1.1052,
      "step": 310
    },
    {
      "epoch": 2.03,
      "learning_rate": 4.948708206686931e-05,
      "loss": 1.4309,
      "step": 320
    },
    {
      "epoch": 2.03,
      "learning_rate": 4.9297112462006076e-05,
      "loss": 1.4591,
      "step": 330
    },
    {
      "epoch": 2.04,
      "learning_rate": 4.910714285714286e-05,
      "loss": 1.3331,
      "step": 340
    },
    {
      "epoch": 2.04,
      "learning_rate": 4.8917173252279635e-05,
      "loss": 1.6523,
      "step": 350
    },
    {
      "epoch": 2.04,
      "eval_accuracy": 0.33640552995391704,
      "eval_loss": 1.300092101097107,
      "eval_runtime": 225.2135,
      "eval_samples_per_second": 0.964,
      "eval_steps_per_second": 0.244,
      "step": 354
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.872720364741641e-05,
      "loss": 1.4045,
      "step": 360
    },
    {
      "epoch": 3.01,
      "learning_rate": 4.8537234042553194e-05,
      "loss": 1.441,
      "step": 370
    },
    {
      "epoch": 3.01,
      "learning_rate": 4.834726443768997e-05,
      "loss": 1.167,
      "step": 380
    },
    {
      "epoch": 3.01,
      "learning_rate": 4.815729483282675e-05,
      "loss": 1.1586,
      "step": 390
    },
    {
      "epoch": 3.02,
      "learning_rate": 4.796732522796353e-05,
      "loss": 1.5054,
      "step": 400
    },
    {
      "epoch": 3.02,
      "learning_rate": 4.7777355623100304e-05,
      "loss": 1.4918,
      "step": 410
    },
    {
      "epoch": 3.02,
      "learning_rate": 4.7587386018237086e-05,
      "loss": 1.3132,
      "step": 420
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.739741641337386e-05,
      "loss": 1.2849,
      "step": 430
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.720744680851064e-05,
      "loss": 1.4326,
      "step": 440
    },
    {
      "epoch": 3.03,
      "learning_rate": 4.701747720364742e-05,
      "loss": 1.2572,
      "step": 450
    },
    {
      "epoch": 3.04,
      "learning_rate": 4.68275075987842e-05,
      "loss": 1.2612,
      "step": 460
    },
    {
      "epoch": 3.04,
      "learning_rate": 4.663753799392098e-05,
      "loss": 1.2763,
      "step": 470
    },
    {
      "epoch": 3.04,
      "eval_accuracy": 0.391705069124424,
      "eval_loss": 1.189802646636963,
      "eval_runtime": 216.5418,
      "eval_samples_per_second": 1.002,
      "eval_steps_per_second": 0.254,
      "step": 472
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.6447568389057755e-05,
      "loss": 1.2211,
      "step": 480
    },
    {
      "epoch": 4.01,
      "learning_rate": 4.625759878419453e-05,
      "loss": 1.2479,
      "step": 490
    },
    {
      "epoch": 4.01,
      "learning_rate": 4.6067629179331314e-05,
      "loss": 1.4419,
      "step": 500
    },
    {
      "epoch": 4.01,
      "learning_rate": 4.587765957446808e-05,
      "loss": 1.3921,
      "step": 510
    },
    {
      "epoch": 4.02,
      "learning_rate": 4.5687689969604865e-05,
      "loss": 1.3209,
      "step": 520
    },
    {
      "epoch": 4.02,
      "learning_rate": 4.549772036474164e-05,
      "loss": 1.2095,
      "step": 530
    },
    {
      "epoch": 4.02,
      "learning_rate": 4.530775075987842e-05,
      "loss": 1.1979,
      "step": 540
    },
    {
      "epoch": 4.03,
      "learning_rate": 4.51177811550152e-05,
      "loss": 1.2475,
      "step": 550
    },
    {
      "epoch": 4.03,
      "learning_rate": 4.4927811550151976e-05,
      "loss": 1.1093,
      "step": 560
    },
    {
      "epoch": 4.03,
      "learning_rate": 4.473784194528876e-05,
      "loss": 1.0525,
      "step": 570
    },
    {
      "epoch": 4.04,
      "learning_rate": 4.4547872340425534e-05,
      "loss": 1.2237,
      "step": 580
    },
    {
      "epoch": 4.04,
      "learning_rate": 4.435790273556231e-05,
      "loss": 1.2819,
      "step": 590
    },
    {
      "epoch": 4.04,
      "eval_accuracy": 0.5161290322580645,
      "eval_loss": 1.1944913864135742,
      "eval_runtime": 228.4885,
      "eval_samples_per_second": 0.95,
      "eval_steps_per_second": 0.241,
      "step": 590
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.416793313069909e-05,
      "loss": 1.3334,
      "step": 600
    },
    {
      "epoch": 5.01,
      "learning_rate": 4.397796352583587e-05,
      "loss": 0.9267,
      "step": 610
    },
    {
      "epoch": 5.01,
      "learning_rate": 4.3787993920972644e-05,
      "loss": 1.0805,
      "step": 620
    },
    {
      "epoch": 5.01,
      "learning_rate": 4.359802431610943e-05,
      "loss": 1.0312,
      "step": 630
    },
    {
      "epoch": 5.02,
      "learning_rate": 4.34080547112462e-05,
      "loss": 1.1775,
      "step": 640
    },
    {
      "epoch": 5.02,
      "learning_rate": 4.3218085106382986e-05,
      "loss": 1.2365,
      "step": 650
    },
    {
      "epoch": 5.02,
      "learning_rate": 4.302811550151976e-05,
      "loss": 1.3273,
      "step": 660
    },
    {
      "epoch": 5.03,
      "learning_rate": 4.283814589665654e-05,
      "loss": 0.975,
      "step": 670
    },
    {
      "epoch": 5.03,
      "learning_rate": 4.264817629179331e-05,
      "loss": 1.2531,
      "step": 680
    },
    {
      "epoch": 5.03,
      "learning_rate": 4.245820668693009e-05,
      "loss": 1.0818,
      "step": 690
    },
    {
      "epoch": 5.04,
      "learning_rate": 4.226823708206687e-05,
      "loss": 1.1196,
      "step": 700
    },
    {
      "epoch": 5.04,
      "eval_accuracy": 0.5668202764976958,
      "eval_loss": 0.97713303565979,
      "eval_runtime": 247.078,
      "eval_samples_per_second": 0.878,
      "eval_steps_per_second": 0.223,
      "step": 708
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.207826747720365e-05,
      "loss": 1.0335,
      "step": 710
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.188829787234042e-05,
      "loss": 1.2908,
      "step": 720
    },
    {
      "epoch": 6.01,
      "learning_rate": 4.1698328267477206e-05,
      "loss": 0.9995,
      "step": 730
    },
    {
      "epoch": 6.01,
      "learning_rate": 4.150835866261398e-05,
      "loss": 1.1156,
      "step": 740
    },
    {
      "epoch": 6.01,
      "learning_rate": 4.1318389057750765e-05,
      "loss": 1.0464,
      "step": 750
    },
    {
      "epoch": 6.02,
      "learning_rate": 4.112841945288754e-05,
      "loss": 1.1695,
      "step": 760
    },
    {
      "epoch": 6.02,
      "learning_rate": 4.0938449848024316e-05,
      "loss": 1.2852,
      "step": 770
    },
    {
      "epoch": 6.02,
      "learning_rate": 4.07484802431611e-05,
      "loss": 0.8937,
      "step": 780
    },
    {
      "epoch": 6.03,
      "learning_rate": 4.0558510638297875e-05,
      "loss": 1.4134,
      "step": 790
    },
    {
      "epoch": 6.03,
      "learning_rate": 4.036854103343465e-05,
      "loss": 0.973,
      "step": 800
    },
    {
      "epoch": 6.03,
      "learning_rate": 4.017857142857143e-05,
      "loss": 0.8961,
      "step": 810
    },
    {
      "epoch": 6.04,
      "learning_rate": 3.998860182370821e-05,
      "loss": 1.1976,
      "step": 820
    },
    {
      "epoch": 6.04,
      "eval_accuracy": 0.47465437788018433,
      "eval_loss": 1.1078805923461914,
      "eval_runtime": 222.1834,
      "eval_samples_per_second": 0.977,
      "eval_steps_per_second": 0.248,
      "step": 826
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.979863221884499e-05,
      "loss": 1.4089,
      "step": 830
    },
    {
      "epoch": 7.0,
      "learning_rate": 3.960866261398177e-05,
      "loss": 1.0279,
      "step": 840
    },
    {
      "epoch": 7.01,
      "learning_rate": 3.9418693009118544e-05,
      "loss": 0.9245,
      "step": 850
    },
    {
      "epoch": 7.01,
      "learning_rate": 3.922872340425532e-05,
      "loss": 1.0762,
      "step": 860
    },
    {
      "epoch": 7.02,
      "learning_rate": 3.9038753799392095e-05,
      "loss": 1.2435,
      "step": 870
    },
    {
      "epoch": 7.02,
      "learning_rate": 3.884878419452888e-05,
      "loss": 1.136,
      "step": 880
    },
    {
      "epoch": 7.02,
      "learning_rate": 3.8658814589665654e-05,
      "loss": 1.1807,
      "step": 890
    },
    {
      "epoch": 7.03,
      "learning_rate": 3.846884498480243e-05,
      "loss": 1.2093,
      "step": 900
    },
    {
      "epoch": 7.03,
      "learning_rate": 3.827887537993921e-05,
      "loss": 0.9918,
      "step": 910
    },
    {
      "epoch": 7.03,
      "learning_rate": 3.808890577507599e-05,
      "loss": 1.0212,
      "step": 920
    },
    {
      "epoch": 7.04,
      "learning_rate": 3.789893617021277e-05,
      "loss": 0.8492,
      "step": 930
    },
    {
      "epoch": 7.04,
      "learning_rate": 3.770896656534955e-05,
      "loss": 0.8174,
      "step": 940
    },
    {
      "epoch": 7.04,
      "eval_accuracy": 0.663594470046083,
      "eval_loss": 0.8704583644866943,
      "eval_runtime": 239.4279,
      "eval_samples_per_second": 0.906,
      "eval_steps_per_second": 0.23,
      "step": 944
    },
    {
      "epoch": 8.0,
      "learning_rate": 3.751899696048632e-05,
      "loss": 0.6433,
      "step": 950
    },
    {
      "epoch": 8.01,
      "learning_rate": 3.7329027355623105e-05,
      "loss": 1.1821,
      "step": 960
    },
    {
      "epoch": 8.01,
      "learning_rate": 3.713905775075988e-05,
      "loss": 1.0545,
      "step": 970
    },
    {
      "epoch": 8.01,
      "learning_rate": 3.694908814589666e-05,
      "loss": 1.0039,
      "step": 980
    },
    {
      "epoch": 8.02,
      "learning_rate": 3.675911854103344e-05,
      "loss": 1.2187,
      "step": 990
    },
    {
      "epoch": 8.02,
      "learning_rate": 3.6569148936170215e-05,
      "loss": 0.876,
      "step": 1000
    },
    {
      "epoch": 8.02,
      "learning_rate": 3.6379179331307e-05,
      "loss": 1.0129,
      "step": 1010
    },
    {
      "epoch": 8.03,
      "learning_rate": 3.6189209726443774e-05,
      "loss": 0.8301,
      "step": 1020
    },
    {
      "epoch": 8.03,
      "learning_rate": 3.599924012158054e-05,
      "loss": 0.9005,
      "step": 1030
    },
    {
      "epoch": 8.03,
      "learning_rate": 3.5809270516717326e-05,
      "loss": 0.9794,
      "step": 1040
    },
    {
      "epoch": 8.04,
      "learning_rate": 3.56193009118541e-05,
      "loss": 0.8015,
      "step": 1050
    },
    {
      "epoch": 8.04,
      "learning_rate": 3.5429331306990884e-05,
      "loss": 0.7923,
      "step": 1060
    },
    {
      "epoch": 8.04,
      "eval_accuracy": 0.663594470046083,
      "eval_loss": 0.782666027545929,
      "eval_runtime": 238.0014,
      "eval_samples_per_second": 0.912,
      "eval_steps_per_second": 0.231,
      "step": 1062
    },
    {
      "epoch": 9.0,
      "learning_rate": 3.523936170212766e-05,
      "loss": 1.0872,
      "step": 1070
    },
    {
      "epoch": 9.01,
      "learning_rate": 3.5049392097264436e-05,
      "loss": 0.9785,
      "step": 1080
    },
    {
      "epoch": 9.01,
      "learning_rate": 3.485942249240122e-05,
      "loss": 1.1311,
      "step": 1090
    },
    {
      "epoch": 9.01,
      "learning_rate": 3.4669452887537994e-05,
      "loss": 1.1006,
      "step": 1100
    },
    {
      "epoch": 9.02,
      "learning_rate": 3.447948328267478e-05,
      "loss": 1.1369,
      "step": 1110
    },
    {
      "epoch": 9.02,
      "learning_rate": 3.428951367781155e-05,
      "loss": 0.8852,
      "step": 1120
    },
    {
      "epoch": 9.02,
      "learning_rate": 3.409954407294833e-05,
      "loss": 0.7969,
      "step": 1130
    },
    {
      "epoch": 9.03,
      "learning_rate": 3.390957446808511e-05,
      "loss": 1.0656,
      "step": 1140
    },
    {
      "epoch": 9.03,
      "learning_rate": 3.371960486322189e-05,
      "loss": 0.6855,
      "step": 1150
    },
    {
      "epoch": 9.03,
      "learning_rate": 3.352963525835866e-05,
      "loss": 0.9424,
      "step": 1160
    },
    {
      "epoch": 9.04,
      "learning_rate": 3.3339665653495446e-05,
      "loss": 0.9198,
      "step": 1170
    },
    {
      "epoch": 9.04,
      "learning_rate": 3.314969604863222e-05,
      "loss": 0.6052,
      "step": 1180
    },
    {
      "epoch": 9.04,
      "eval_accuracy": 0.631336405529954,
      "eval_loss": 0.945479691028595,
      "eval_runtime": 243.8739,
      "eval_samples_per_second": 0.89,
      "eval_steps_per_second": 0.226,
      "step": 1180
    },
    {
      "epoch": 10.0,
      "learning_rate": 3.2959726443769004e-05,
      "loss": 0.748,
      "step": 1190
    },
    {
      "epoch": 10.01,
      "learning_rate": 3.2769756838905773e-05,
      "loss": 0.9443,
      "step": 1200
    },
    {
      "epoch": 10.01,
      "learning_rate": 3.257978723404255e-05,
      "loss": 0.8901,
      "step": 1210
    },
    {
      "epoch": 10.01,
      "learning_rate": 3.238981762917933e-05,
      "loss": 0.8665,
      "step": 1220
    },
    {
      "epoch": 10.02,
      "learning_rate": 3.219984802431611e-05,
      "loss": 0.8145,
      "step": 1230
    },
    {
      "epoch": 10.02,
      "learning_rate": 3.200987841945289e-05,
      "loss": 0.9171,
      "step": 1240
    },
    {
      "epoch": 10.02,
      "learning_rate": 3.1819908814589666e-05,
      "loss": 1.0532,
      "step": 1250
    },
    {
      "epoch": 10.03,
      "learning_rate": 3.162993920972644e-05,
      "loss": 0.8078,
      "step": 1260
    },
    {
      "epoch": 10.03,
      "learning_rate": 3.1439969604863225e-05,
      "loss": 0.6475,
      "step": 1270
    },
    {
      "epoch": 10.03,
      "learning_rate": 3.125e-05,
      "loss": 0.8414,
      "step": 1280
    },
    {
      "epoch": 10.04,
      "learning_rate": 3.106003039513678e-05,
      "loss": 0.7902,
      "step": 1290
    },
    {
      "epoch": 10.04,
      "eval_accuracy": 0.6175115207373272,
      "eval_loss": 0.9512102603912354,
      "eval_runtime": 253.9642,
      "eval_samples_per_second": 0.854,
      "eval_steps_per_second": 0.217,
      "step": 1298
    },
    {
      "epoch": 11.0,
      "learning_rate": 3.087006079027356e-05,
      "loss": 0.7779,
      "step": 1300
    },
    {
      "epoch": 11.0,
      "learning_rate": 3.0680091185410335e-05,
      "loss": 0.6305,
      "step": 1310
    },
    {
      "epoch": 11.01,
      "learning_rate": 3.0490121580547114e-05,
      "loss": 1.0614,
      "step": 1320
    },
    {
      "epoch": 11.01,
      "learning_rate": 3.0300151975683894e-05,
      "loss": 0.609,
      "step": 1330
    },
    {
      "epoch": 11.01,
      "learning_rate": 3.0110182370820673e-05,
      "loss": 0.8034,
      "step": 1340
    },
    {
      "epoch": 11.02,
      "learning_rate": 2.9920212765957452e-05,
      "loss": 1.1648,
      "step": 1350
    },
    {
      "epoch": 11.02,
      "learning_rate": 2.9730243161094228e-05,
      "loss": 0.8944,
      "step": 1360
    },
    {
      "epoch": 11.02,
      "learning_rate": 2.9540273556231e-05,
      "loss": 0.7993,
      "step": 1370
    },
    {
      "epoch": 11.03,
      "learning_rate": 2.935030395136778e-05,
      "loss": 0.9185,
      "step": 1380
    },
    {
      "epoch": 11.03,
      "learning_rate": 2.916033434650456e-05,
      "loss": 0.824,
      "step": 1390
    },
    {
      "epoch": 11.03,
      "learning_rate": 2.8970364741641338e-05,
      "loss": 0.8603,
      "step": 1400
    },
    {
      "epoch": 11.04,
      "learning_rate": 2.8780395136778114e-05,
      "loss": 0.9501,
      "step": 1410
    },
    {
      "epoch": 11.04,
      "eval_accuracy": 0.7373271889400922,
      "eval_loss": 0.8057688474655151,
      "eval_runtime": 244.5172,
      "eval_samples_per_second": 0.887,
      "eval_steps_per_second": 0.225,
      "step": 1416
    },
    {
      "epoch": 12.0,
      "learning_rate": 2.8590425531914893e-05,
      "loss": 0.8682,
      "step": 1420
    },
    {
      "epoch": 12.0,
      "learning_rate": 2.8400455927051673e-05,
      "loss": 0.688,
      "step": 1430
    },
    {
      "epoch": 12.01,
      "learning_rate": 2.8210486322188452e-05,
      "loss": 0.755,
      "step": 1440
    },
    {
      "epoch": 12.01,
      "learning_rate": 2.8020516717325228e-05,
      "loss": 1.0344,
      "step": 1450
    },
    {
      "epoch": 12.02,
      "learning_rate": 2.7830547112462007e-05,
      "loss": 1.0774,
      "step": 1460
    },
    {
      "epoch": 12.02,
      "learning_rate": 2.7640577507598786e-05,
      "loss": 0.9952,
      "step": 1470
    },
    {
      "epoch": 12.02,
      "learning_rate": 2.7450607902735565e-05,
      "loss": 1.011,
      "step": 1480
    },
    {
      "epoch": 12.03,
      "learning_rate": 2.726063829787234e-05,
      "loss": 0.8582,
      "step": 1490
    },
    {
      "epoch": 12.03,
      "learning_rate": 2.707066869300912e-05,
      "loss": 0.8557,
      "step": 1500
    },
    {
      "epoch": 12.03,
      "learning_rate": 2.68806990881459e-05,
      "loss": 0.9247,
      "step": 1510
    },
    {
      "epoch": 12.04,
      "learning_rate": 2.669072948328268e-05,
      "loss": 0.7037,
      "step": 1520
    },
    {
      "epoch": 12.04,
      "learning_rate": 2.6500759878419458e-05,
      "loss": 1.1756,
      "step": 1530
    },
    {
      "epoch": 12.04,
      "eval_accuracy": 0.7327188940092166,
      "eval_loss": 0.7674494385719299,
      "eval_runtime": 255.0659,
      "eval_samples_per_second": 0.851,
      "eval_steps_per_second": 0.216,
      "step": 1534
    },
    {
      "epoch": 13.0,
      "learning_rate": 2.631079027355623e-05,
      "loss": 0.7381,
      "step": 1540
    },
    {
      "epoch": 13.01,
      "learning_rate": 2.6120820668693007e-05,
      "loss": 0.7239,
      "step": 1550
    },
    {
      "epoch": 13.01,
      "learning_rate": 2.5930851063829786e-05,
      "loss": 0.8164,
      "step": 1560
    },
    {
      "epoch": 13.01,
      "learning_rate": 2.5740881458966565e-05,
      "loss": 0.7785,
      "step": 1570
    },
    {
      "epoch": 13.02,
      "learning_rate": 2.5550911854103344e-05,
      "loss": 0.554,
      "step": 1580
    },
    {
      "epoch": 13.02,
      "learning_rate": 2.536094224924012e-05,
      "loss": 1.0316,
      "step": 1590
    },
    {
      "epoch": 13.02,
      "learning_rate": 2.51709726443769e-05,
      "loss": 0.7692,
      "step": 1600
    },
    {
      "epoch": 13.03,
      "learning_rate": 2.498100303951368e-05,
      "loss": 0.8983,
      "step": 1610
    },
    {
      "epoch": 13.03,
      "learning_rate": 2.4791033434650458e-05,
      "loss": 0.7436,
      "step": 1620
    },
    {
      "epoch": 13.03,
      "learning_rate": 2.4601063829787234e-05,
      "loss": 0.6312,
      "step": 1630
    },
    {
      "epoch": 13.04,
      "learning_rate": 2.4411094224924013e-05,
      "loss": 0.7844,
      "step": 1640
    },
    {
      "epoch": 13.04,
      "learning_rate": 2.4221124620060792e-05,
      "loss": 0.8482,
      "step": 1650
    },
    {
      "epoch": 13.04,
      "eval_accuracy": 0.6866359447004609,
      "eval_loss": 0.8454816341400146,
      "eval_runtime": 252.0585,
      "eval_samples_per_second": 0.861,
      "eval_steps_per_second": 0.218,
      "step": 1652
    },
    {
      "epoch": 14.0,
      "learning_rate": 2.403115501519757e-05,
      "loss": 0.7618,
      "step": 1660
    },
    {
      "epoch": 14.01,
      "learning_rate": 2.3841185410334348e-05,
      "loss": 0.6308,
      "step": 1670
    },
    {
      "epoch": 14.01,
      "learning_rate": 2.3651215805471123e-05,
      "loss": 0.6756,
      "step": 1680
    },
    {
      "epoch": 14.01,
      "learning_rate": 2.3461246200607903e-05,
      "loss": 0.6555,
      "step": 1690
    },
    {
      "epoch": 14.02,
      "learning_rate": 2.3271276595744682e-05,
      "loss": 0.8249,
      "step": 1700
    },
    {
      "epoch": 14.02,
      "learning_rate": 2.308130699088146e-05,
      "loss": 0.6899,
      "step": 1710
    },
    {
      "epoch": 14.02,
      "learning_rate": 2.2891337386018237e-05,
      "loss": 0.7601,
      "step": 1720
    },
    {
      "epoch": 14.03,
      "learning_rate": 2.2701367781155016e-05,
      "loss": 0.7638,
      "step": 1730
    },
    {
      "epoch": 14.03,
      "learning_rate": 2.2511398176291796e-05,
      "loss": 0.4925,
      "step": 1740
    },
    {
      "epoch": 14.03,
      "learning_rate": 2.2321428571428575e-05,
      "loss": 0.6234,
      "step": 1750
    },
    {
      "epoch": 14.04,
      "learning_rate": 2.213145896656535e-05,
      "loss": 1.3981,
      "step": 1760
    },
    {
      "epoch": 14.04,
      "learning_rate": 2.1941489361702127e-05,
      "loss": 0.7657,
      "step": 1770
    },
    {
      "epoch": 14.04,
      "eval_accuracy": 0.6820276497695853,
      "eval_loss": 0.912429928779602,
      "eval_runtime": 258.9167,
      "eval_samples_per_second": 0.838,
      "eval_steps_per_second": 0.212,
      "step": 1770
    },
    {
      "epoch": 15.0,
      "learning_rate": 2.1751519756838906e-05,
      "loss": 0.5498,
      "step": 1780
    },
    {
      "epoch": 15.01,
      "learning_rate": 2.1561550151975685e-05,
      "loss": 0.6559,
      "step": 1790
    },
    {
      "epoch": 15.01,
      "learning_rate": 2.1371580547112464e-05,
      "loss": 0.9217,
      "step": 1800
    },
    {
      "epoch": 15.01,
      "learning_rate": 2.118161094224924e-05,
      "loss": 0.7298,
      "step": 1810
    },
    {
      "epoch": 15.02,
      "learning_rate": 2.099164133738602e-05,
      "loss": 0.9773,
      "step": 1820
    },
    {
      "epoch": 15.02,
      "learning_rate": 2.08016717325228e-05,
      "loss": 0.5767,
      "step": 1830
    },
    {
      "epoch": 15.02,
      "learning_rate": 2.0611702127659578e-05,
      "loss": 0.8968,
      "step": 1840
    },
    {
      "epoch": 15.03,
      "learning_rate": 2.0421732522796354e-05,
      "loss": 0.7559,
      "step": 1850
    },
    {
      "epoch": 15.03,
      "learning_rate": 2.023176291793313e-05,
      "loss": 0.6842,
      "step": 1860
    },
    {
      "epoch": 15.03,
      "learning_rate": 2.004179331306991e-05,
      "loss": 0.7599,
      "step": 1870
    },
    {
      "epoch": 15.04,
      "learning_rate": 1.9851823708206688e-05,
      "loss": 0.9752,
      "step": 1880
    },
    {
      "epoch": 15.04,
      "eval_accuracy": 0.631336405529954,
      "eval_loss": 0.9787760376930237,
      "eval_runtime": 270.6691,
      "eval_samples_per_second": 0.802,
      "eval_steps_per_second": 0.203,
      "step": 1888
    },
    {
      "epoch": 16.0,
      "learning_rate": 1.9661854103343467e-05,
      "loss": 0.5757,
      "step": 1890
    },
    {
      "epoch": 16.0,
      "learning_rate": 1.9471884498480243e-05,
      "loss": 0.8474,
      "step": 1900
    },
    {
      "epoch": 16.01,
      "learning_rate": 1.9281914893617023e-05,
      "loss": 0.6007,
      "step": 1910
    },
    {
      "epoch": 16.01,
      "learning_rate": 1.9091945288753802e-05,
      "loss": 0.7523,
      "step": 1920
    },
    {
      "epoch": 16.01,
      "learning_rate": 1.8901975683890578e-05,
      "loss": 0.6769,
      "step": 1930
    },
    {
      "epoch": 16.02,
      "learning_rate": 1.8712006079027357e-05,
      "loss": 0.9608,
      "step": 1940
    },
    {
      "epoch": 16.02,
      "learning_rate": 1.8522036474164133e-05,
      "loss": 0.6158,
      "step": 1950
    },
    {
      "epoch": 16.02,
      "learning_rate": 1.8332066869300912e-05,
      "loss": 0.8995,
      "step": 1960
    },
    {
      "epoch": 16.03,
      "learning_rate": 1.814209726443769e-05,
      "loss": 0.67,
      "step": 1970
    },
    {
      "epoch": 16.03,
      "learning_rate": 1.795212765957447e-05,
      "loss": 0.7389,
      "step": 1980
    },
    {
      "epoch": 16.03,
      "learning_rate": 1.7762158054711246e-05,
      "loss": 0.5213,
      "step": 1990
    },
    {
      "epoch": 16.04,
      "learning_rate": 1.7572188449848026e-05,
      "loss": 0.7176,
      "step": 2000
    },
    {
      "epoch": 16.04,
      "eval_accuracy": 0.7603686635944701,
      "eval_loss": 0.6860308051109314,
      "eval_runtime": 242.3496,
      "eval_samples_per_second": 0.895,
      "eval_steps_per_second": 0.227,
      "step": 2006
    },
    {
      "epoch": 17.0,
      "learning_rate": 1.7382218844984805e-05,
      "loss": 0.6261,
      "step": 2010
    },
    {
      "epoch": 17.0,
      "learning_rate": 1.719224924012158e-05,
      "loss": 0.7283,
      "step": 2020
    },
    {
      "epoch": 17.01,
      "learning_rate": 1.700227963525836e-05,
      "loss": 0.8489,
      "step": 2030
    },
    {
      "epoch": 17.01,
      "learning_rate": 1.6812310030395136e-05,
      "loss": 0.6356,
      "step": 2040
    },
    {
      "epoch": 17.02,
      "learning_rate": 1.6622340425531915e-05,
      "loss": 0.7721,
      "step": 2050
    },
    {
      "epoch": 17.02,
      "learning_rate": 1.6432370820668694e-05,
      "loss": 0.914,
      "step": 2060
    },
    {
      "epoch": 17.02,
      "learning_rate": 1.6242401215805474e-05,
      "loss": 0.5385,
      "step": 2070
    },
    {
      "epoch": 17.03,
      "learning_rate": 1.605243161094225e-05,
      "loss": 0.6783,
      "step": 2080
    },
    {
      "epoch": 17.03,
      "learning_rate": 1.586246200607903e-05,
      "loss": 0.6296,
      "step": 2090
    },
    {
      "epoch": 17.03,
      "learning_rate": 1.5672492401215808e-05,
      "loss": 0.4384,
      "step": 2100
    },
    {
      "epoch": 17.04,
      "learning_rate": 1.5482522796352584e-05,
      "loss": 0.7417,
      "step": 2110
    },
    {
      "epoch": 17.04,
      "learning_rate": 1.5292553191489363e-05,
      "loss": 0.7682,
      "step": 2120
    },
    {
      "epoch": 17.04,
      "eval_accuracy": 0.7142857142857143,
      "eval_loss": 0.8052204847335815,
      "eval_runtime": 263.0911,
      "eval_samples_per_second": 0.825,
      "eval_steps_per_second": 0.209,
      "step": 2124
    },
    {
      "epoch": 18.0,
      "learning_rate": 1.510258358662614e-05,
      "loss": 0.9385,
      "step": 2130
    },
    {
      "epoch": 18.01,
      "learning_rate": 1.4912613981762918e-05,
      "loss": 0.639,
      "step": 2140
    },
    {
      "epoch": 18.01,
      "learning_rate": 1.4722644376899698e-05,
      "loss": 0.379,
      "step": 2150
    },
    {
      "epoch": 18.01,
      "learning_rate": 1.4532674772036475e-05,
      "loss": 0.9416,
      "step": 2160
    },
    {
      "epoch": 18.02,
      "learning_rate": 1.4342705167173254e-05,
      "loss": 0.7913,
      "step": 2170
    },
    {
      "epoch": 18.02,
      "learning_rate": 1.4152735562310032e-05,
      "loss": 0.6423,
      "step": 2180
    },
    {
      "epoch": 18.02,
      "learning_rate": 1.3962765957446808e-05,
      "loss": 0.7556,
      "step": 2190
    },
    {
      "epoch": 18.03,
      "learning_rate": 1.3772796352583587e-05,
      "loss": 0.3175,
      "step": 2200
    },
    {
      "epoch": 18.03,
      "learning_rate": 1.3582826747720365e-05,
      "loss": 0.9512,
      "step": 2210
    },
    {
      "epoch": 18.03,
      "learning_rate": 1.3392857142857144e-05,
      "loss": 0.6495,
      "step": 2220
    },
    {
      "epoch": 18.04,
      "learning_rate": 1.3202887537993921e-05,
      "loss": 0.8305,
      "step": 2230
    },
    {
      "epoch": 18.04,
      "learning_rate": 1.30129179331307e-05,
      "loss": 0.8862,
      "step": 2240
    },
    {
      "epoch": 18.04,
      "eval_accuracy": 0.6820276497695853,
      "eval_loss": 0.9101675152778625,
      "eval_runtime": 259.2768,
      "eval_samples_per_second": 0.837,
      "eval_steps_per_second": 0.212,
      "step": 2242
    },
    {
      "epoch": 19.0,
      "learning_rate": 1.2822948328267478e-05,
      "loss": 0.5608,
      "step": 2250
    },
    {
      "epoch": 19.01,
      "learning_rate": 1.2632978723404257e-05,
      "loss": 0.6882,
      "step": 2260
    },
    {
      "epoch": 19.01,
      "learning_rate": 1.2443009118541033e-05,
      "loss": 1.2312,
      "step": 2270
    },
    {
      "epoch": 19.01,
      "learning_rate": 1.2253039513677813e-05,
      "loss": 0.6054,
      "step": 2280
    },
    {
      "epoch": 19.02,
      "learning_rate": 1.206306990881459e-05,
      "loss": 0.5243,
      "step": 2290
    },
    {
      "epoch": 19.02,
      "learning_rate": 1.1873100303951368e-05,
      "loss": 0.6013,
      "step": 2300
    },
    {
      "epoch": 19.02,
      "learning_rate": 1.1683130699088147e-05,
      "loss": 0.7646,
      "step": 2310
    },
    {
      "epoch": 19.03,
      "learning_rate": 1.1493161094224925e-05,
      "loss": 0.6475,
      "step": 2320
    },
    {
      "epoch": 19.03,
      "learning_rate": 1.1303191489361704e-05,
      "loss": 0.7699,
      "step": 2330
    },
    {
      "epoch": 19.03,
      "learning_rate": 1.1113221884498481e-05,
      "loss": 0.5824,
      "step": 2340
    },
    {
      "epoch": 19.04,
      "learning_rate": 1.0923252279635259e-05,
      "loss": 0.6196,
      "step": 2350
    },
    {
      "epoch": 19.04,
      "learning_rate": 1.0733282674772036e-05,
      "loss": 0.6609,
      "step": 2360
    },
    {
      "epoch": 19.04,
      "eval_accuracy": 0.6866359447004609,
      "eval_loss": 0.9359898567199707,
      "eval_runtime": 268.8932,
      "eval_samples_per_second": 0.807,
      "eval_steps_per_second": 0.205,
      "step": 2360
    },
    {
      "epoch": 20.0,
      "learning_rate": 1.0543313069908816e-05,
      "loss": 0.6174,
      "step": 2370
    },
    {
      "epoch": 20.01,
      "learning_rate": 1.0353343465045593e-05,
      "loss": 0.7386,
      "step": 2380
    },
    {
      "epoch": 20.01,
      "learning_rate": 1.0163373860182371e-05,
      "loss": 0.6358,
      "step": 2390
    },
    {
      "epoch": 20.01,
      "learning_rate": 9.973404255319148e-06,
      "loss": 0.4331,
      "step": 2400
    },
    {
      "epoch": 20.02,
      "learning_rate": 9.783434650455928e-06,
      "loss": 0.5209,
      "step": 2410
    },
    {
      "epoch": 20.02,
      "learning_rate": 9.593465045592707e-06,
      "loss": 0.4354,
      "step": 2420
    },
    {
      "epoch": 20.02,
      "learning_rate": 9.403495440729483e-06,
      "loss": 0.8506,
      "step": 2430
    },
    {
      "epoch": 20.03,
      "learning_rate": 9.213525835866262e-06,
      "loss": 1.0944,
      "step": 2440
    },
    {
      "epoch": 20.03,
      "learning_rate": 9.02355623100304e-06,
      "loss": 0.6736,
      "step": 2450
    },
    {
      "epoch": 20.03,
      "learning_rate": 8.833586626139819e-06,
      "loss": 0.5064,
      "step": 2460
    },
    {
      "epoch": 20.04,
      "learning_rate": 8.643617021276596e-06,
      "loss": 0.9678,
      "step": 2470
    },
    {
      "epoch": 20.04,
      "eval_accuracy": 0.6866359447004609,
      "eval_loss": 0.844132125377655,
      "eval_runtime": 266.4679,
      "eval_samples_per_second": 0.814,
      "eval_steps_per_second": 0.206,
      "step": 2478
    },
    {
      "epoch": 21.0,
      "learning_rate": 8.453647416413374e-06,
      "loss": 0.6142,
      "step": 2480
    },
    {
      "epoch": 21.0,
      "learning_rate": 8.263677811550152e-06,
      "loss": 0.769,
      "step": 2490
    },
    {
      "epoch": 21.01,
      "learning_rate": 8.07370820668693e-06,
      "loss": 0.4844,
      "step": 2500
    },
    {
      "epoch": 21.01,
      "learning_rate": 7.883738601823708e-06,
      "loss": 0.6551,
      "step": 2510
    },
    {
      "epoch": 21.01,
      "learning_rate": 7.693768996960486e-06,
      "loss": 0.8922,
      "step": 2520
    },
    {
      "epoch": 21.02,
      "learning_rate": 7.503799392097264e-06,
      "loss": 0.6909,
      "step": 2530
    },
    {
      "epoch": 21.02,
      "learning_rate": 7.313829787234043e-06,
      "loss": 0.6353,
      "step": 2540
    },
    {
      "epoch": 21.02,
      "learning_rate": 7.123860182370821e-06,
      "loss": 0.3498,
      "step": 2550
    },
    {
      "epoch": 21.03,
      "learning_rate": 6.933890577507599e-06,
      "loss": 0.4762,
      "step": 2560
    },
    {
      "epoch": 21.03,
      "learning_rate": 6.743920972644377e-06,
      "loss": 0.3909,
      "step": 2570
    },
    {
      "epoch": 21.03,
      "learning_rate": 6.5539513677811555e-06,
      "loss": 0.9675,
      "step": 2580
    },
    {
      "epoch": 21.04,
      "learning_rate": 6.363981762917934e-06,
      "loss": 0.8617,
      "step": 2590
    },
    {
      "epoch": 21.04,
      "eval_accuracy": 0.7096774193548387,
      "eval_loss": 0.8135567307472229,
      "eval_runtime": 251.1165,
      "eval_samples_per_second": 0.864,
      "eval_steps_per_second": 0.219,
      "step": 2596
    },
    {
      "epoch": 22.0,
      "learning_rate": 6.1740121580547115e-06,
      "loss": 0.3984,
      "step": 2600
    },
    {
      "epoch": 22.0,
      "learning_rate": 5.98404255319149e-06,
      "loss": 0.7669,
      "step": 2610
    },
    {
      "epoch": 22.01,
      "learning_rate": 5.7940729483282675e-06,
      "loss": 0.5996,
      "step": 2620
    },
    {
      "epoch": 22.01,
      "learning_rate": 5.604103343465046e-06,
      "loss": 0.6738,
      "step": 2630
    },
    {
      "epoch": 22.02,
      "learning_rate": 5.4141337386018234e-06,
      "loss": 0.774,
      "step": 2640
    },
    {
      "epoch": 22.02,
      "learning_rate": 5.224164133738603e-06,
      "loss": 0.7791,
      "step": 2650
    },
    {
      "epoch": 22.02,
      "learning_rate": 5.03419452887538e-06,
      "loss": 0.7715,
      "step": 2660
    },
    {
      "epoch": 22.03,
      "learning_rate": 4.844224924012159e-06,
      "loss": 0.5902,
      "step": 2670
    },
    {
      "epoch": 22.03,
      "learning_rate": 4.654255319148936e-06,
      "loss": 0.719,
      "step": 2680
    },
    {
      "epoch": 22.03,
      "learning_rate": 4.464285714285715e-06,
      "loss": 0.8216,
      "step": 2690
    },
    {
      "epoch": 22.04,
      "learning_rate": 4.274316109422492e-06,
      "loss": 0.3712,
      "step": 2700
    },
    {
      "epoch": 22.04,
      "learning_rate": 4.084346504559271e-06,
      "loss": 0.7097,
      "step": 2710
    },
    {
      "epoch": 22.04,
      "eval_accuracy": 0.7004608294930875,
      "eval_loss": 0.8615413308143616,
      "eval_runtime": 261.5204,
      "eval_samples_per_second": 0.83,
      "eval_steps_per_second": 0.21,
      "step": 2714
    },
    {
      "epoch": 23.0,
      "learning_rate": 3.894376899696048e-06,
      "loss": 0.7647,
      "step": 2720
    },
    {
      "epoch": 23.01,
      "learning_rate": 3.704407294832827e-06,
      "loss": 0.6997,
      "step": 2730
    },
    {
      "epoch": 23.01,
      "learning_rate": 3.5144376899696054e-06,
      "loss": 0.594,
      "step": 2740
    },
    {
      "epoch": 23.01,
      "learning_rate": 3.324468085106383e-06,
      "loss": 0.648,
      "step": 2750
    },
    {
      "epoch": 23.02,
      "learning_rate": 3.1344984802431614e-06,
      "loss": 0.9131,
      "step": 2760
    },
    {
      "epoch": 23.02,
      "learning_rate": 2.9445288753799393e-06,
      "loss": 0.417,
      "step": 2770
    },
    {
      "epoch": 23.02,
      "learning_rate": 2.7545592705167173e-06,
      "loss": 0.5632,
      "step": 2780
    },
    {
      "epoch": 23.03,
      "learning_rate": 2.5645896656534953e-06,
      "loss": 0.8107,
      "step": 2790
    },
    {
      "epoch": 23.03,
      "learning_rate": 2.3746200607902737e-06,
      "loss": 1.1321,
      "step": 2800
    },
    {
      "epoch": 23.03,
      "learning_rate": 2.1846504559270517e-06,
      "loss": 1.0339,
      "step": 2810
    },
    {
      "epoch": 23.04,
      "learning_rate": 1.99468085106383e-06,
      "loss": 0.62,
      "step": 2820
    },
    {
      "epoch": 23.04,
      "learning_rate": 1.804711246200608e-06,
      "loss": 0.5089,
      "step": 2830
    },
    {
      "epoch": 23.04,
      "eval_accuracy": 0.7004608294930875,
      "eval_loss": 0.7796175479888916,
      "eval_runtime": 273.6817,
      "eval_samples_per_second": 0.793,
      "eval_steps_per_second": 0.201,
      "step": 2832
    },
    {
      "epoch": 24.0,
      "learning_rate": 1.614741641337386e-06,
      "loss": 0.3957,
      "step": 2840
    },
    {
      "epoch": 24.01,
      "learning_rate": 1.4247720364741643e-06,
      "loss": 0.6099,
      "step": 2850
    },
    {
      "epoch": 24.01,
      "learning_rate": 1.2348024316109423e-06,
      "loss": 0.5205,
      "step": 2860
    },
    {
      "epoch": 24.01,
      "learning_rate": 1.0448328267477205e-06,
      "loss": 0.4855,
      "step": 2870
    },
    {
      "epoch": 24.02,
      "learning_rate": 8.548632218844984e-07,
      "loss": 0.6337,
      "step": 2880
    },
    {
      "epoch": 24.02,
      "learning_rate": 6.648936170212766e-07,
      "loss": 0.4915,
      "step": 2890
    },
    {
      "epoch": 24.02,
      "learning_rate": 4.749240121580547e-07,
      "loss": 0.4598,
      "step": 2900
    },
    {
      "epoch": 24.03,
      "learning_rate": 2.849544072948328e-07,
      "loss": 0.4414,
      "step": 2910
    },
    {
      "epoch": 24.03,
      "learning_rate": 9.498480243161096e-08,
      "loss": 0.715,
      "step": 2920
    },
    {
      "epoch": 24.03,
      "eval_accuracy": 0.7235023041474654,
      "eval_loss": 0.7684889435768127,
      "eval_runtime": 267.0096,
      "eval_samples_per_second": 0.813,
      "eval_steps_per_second": 0.206,
      "step": 2925
    },
    {
      "epoch": 24.03,
      "step": 2925,
      "total_flos": 1.4489626697585787e+19,
      "train_loss": 0.9356751281379635,
      "train_runtime": 22099.9881,
      "train_samples_per_second": 0.529,
      "train_steps_per_second": 0.132
    },
    {
      "epoch": 24.03,
      "eval_accuracy": 0.6898148148148148,
      "eval_loss": 0.7174022793769836,
      "eval_runtime": 270.9966,
      "eval_samples_per_second": 0.797,
      "eval_steps_per_second": 0.199,
      "step": 2925
    },
    {
      "epoch": 24.03,
      "eval_accuracy": 0.6898148148148148,
      "eval_loss": 0.7174023389816284,
      "eval_runtime": 252.2199,
      "eval_samples_per_second": 0.856,
      "eval_steps_per_second": 0.214,
      "step": 2925
    }
  ],
  "logging_steps": 10,
  "max_steps": 2925,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 1.4489626697585787e+19,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}