{
  "best_metric": 0.9842381786339754,
  "best_model_checkpoint": "teacher-status-van-tiny-256-2/checkpoint-420",
  "epoch": 29.714285714285715,
  "eval_steps": 500,
  "global_step": 780,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.38,
      "learning_rate": 6.41025641025641e-06,
      "loss": 0.6928,
      "step": 10
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.282051282051282e-05,
      "loss": 0.6896,
      "step": 20
    },
    {
      "epoch": 0.99,
      "eval_accuracy": 0.7700534759358288,
      "eval_f1_score": 0.8700906344410876,
      "eval_loss": 0.670672595500946,
      "eval_precision": 0.7700534759358288,
      "eval_recall": 1.0,
      "eval_runtime": 2.6151,
      "eval_samples_per_second": 143.014,
      "eval_steps_per_second": 4.589,
      "step": 26
    },
    {
      "epoch": 1.14,
      "learning_rate": 1.923076923076923e-05,
      "loss": 0.6763,
      "step": 30
    },
    {
      "epoch": 1.52,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.6376,
      "step": 40
    },
    {
      "epoch": 1.9,
      "learning_rate": 3.205128205128206e-05,
      "loss": 0.5438,
      "step": 50
    },
    {
      "epoch": 1.98,
      "eval_accuracy": 0.7700534759358288,
      "eval_f1_score": 0.8700906344410876,
      "eval_loss": 0.43023183941841125,
      "eval_precision": 0.7700534759358288,
      "eval_recall": 1.0,
      "eval_runtime": 2.5825,
      "eval_samples_per_second": 144.822,
      "eval_steps_per_second": 4.647,
      "step": 52
    },
    {
      "epoch": 2.29,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.4252,
      "step": 60
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.4871794871794874e-05,
      "loss": 0.3756,
      "step": 70
    },
    {
      "epoch": 2.97,
      "eval_accuracy": 0.8850267379679144,
      "eval_f1_score": 0.9284525790349417,
      "eval_loss": 0.2762328088283539,
      "eval_precision": 0.8913738019169329,
      "eval_recall": 0.96875,
      "eval_runtime": 2.711,
      "eval_samples_per_second": 137.957,
      "eval_steps_per_second": 4.426,
      "step": 78
    },
    {
      "epoch": 3.05,
      "learning_rate": 4.985754985754986e-05,
      "loss": 0.3572,
      "step": 80
    },
    {
      "epoch": 3.43,
      "learning_rate": 4.9145299145299147e-05,
      "loss": 0.3217,
      "step": 90
    },
    {
      "epoch": 3.81,
      "learning_rate": 4.8433048433048433e-05,
      "loss": 0.3017,
      "step": 100
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9224598930481284,
      "eval_f1_score": 0.9502572898799314,
      "eval_loss": 0.20020952820777893,
      "eval_precision": 0.9389830508474576,
      "eval_recall": 0.9618055555555556,
      "eval_runtime": 2.7006,
      "eval_samples_per_second": 138.487,
      "eval_steps_per_second": 4.443,
      "step": 105
    },
    {
      "epoch": 4.19,
      "learning_rate": 4.772079772079772e-05,
      "loss": 0.2563,
      "step": 110
    },
    {
      "epoch": 4.57,
      "learning_rate": 4.700854700854701e-05,
      "loss": 0.2548,
      "step": 120
    },
    {
      "epoch": 4.95,
      "learning_rate": 4.62962962962963e-05,
      "loss": 0.257,
      "step": 130
    },
    {
      "epoch": 4.99,
      "eval_accuracy": 0.9385026737967914,
      "eval_f1_score": 0.9605488850771869,
      "eval_loss": 0.17939455807209015,
      "eval_precision": 0.9491525423728814,
      "eval_recall": 0.9722222222222222,
      "eval_runtime": 2.4102,
      "eval_samples_per_second": 155.173,
      "eval_steps_per_second": 4.979,
      "step": 131
    },
    {
      "epoch": 5.33,
      "learning_rate": 4.558404558404559e-05,
      "loss": 0.2484,
      "step": 140
    },
    {
      "epoch": 5.71,
      "learning_rate": 4.4871794871794874e-05,
      "loss": 0.2345,
      "step": 150
    },
    {
      "epoch": 5.98,
      "eval_accuracy": 0.9358288770053476,
      "eval_f1_score": 0.9581881533101045,
      "eval_loss": 0.14852212369441986,
      "eval_precision": 0.9615384615384616,
      "eval_recall": 0.9548611111111112,
      "eval_runtime": 2.3968,
      "eval_samples_per_second": 156.039,
      "eval_steps_per_second": 5.007,
      "step": 157
    },
    {
      "epoch": 6.1,
      "learning_rate": 4.415954415954416e-05,
      "loss": 0.2514,
      "step": 160
    },
    {
      "epoch": 6.48,
      "learning_rate": 4.344729344729345e-05,
      "loss": 0.2179,
      "step": 170
    },
    {
      "epoch": 6.86,
      "learning_rate": 4.2735042735042735e-05,
      "loss": 0.2318,
      "step": 180
    },
    {
      "epoch": 6.97,
      "eval_accuracy": 0.9438502673796791,
      "eval_f1_score": 0.9630931458699472,
      "eval_loss": 0.13021136820316315,
      "eval_precision": 0.9750889679715302,
      "eval_recall": 0.9513888888888888,
      "eval_runtime": 2.3767,
      "eval_samples_per_second": 157.361,
      "eval_steps_per_second": 5.049,
      "step": 183
    },
    {
      "epoch": 7.24,
      "learning_rate": 4.202279202279202e-05,
      "loss": 0.2151,
      "step": 190
    },
    {
      "epoch": 7.62,
      "learning_rate": 4.131054131054131e-05,
      "loss": 0.2311,
      "step": 200
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.05982905982906e-05,
      "loss": 0.2173,
      "step": 210
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9518716577540107,
      "eval_f1_score": 0.9688581314878894,
      "eval_loss": 0.12773053348064423,
      "eval_precision": 0.9655172413793104,
      "eval_recall": 0.9722222222222222,
      "eval_runtime": 2.3674,
      "eval_samples_per_second": 157.977,
      "eval_steps_per_second": 5.069,
      "step": 210
    },
    {
      "epoch": 8.38,
      "learning_rate": 3.988603988603989e-05,
      "loss": 0.1934,
      "step": 220
    },
    {
      "epoch": 8.76,
      "learning_rate": 3.9173789173789176e-05,
      "loss": 0.2058,
      "step": 230
    },
    {
      "epoch": 8.99,
      "eval_accuracy": 0.9572192513368984,
      "eval_f1_score": 0.9722222222222222,
      "eval_loss": 0.1269279271364212,
      "eval_precision": 0.9722222222222222,
      "eval_recall": 0.9722222222222222,
      "eval_runtime": 2.2694,
      "eval_samples_per_second": 164.802,
      "eval_steps_per_second": 5.288,
      "step": 236
    },
    {
      "epoch": 9.14,
      "learning_rate": 3.846153846153846e-05,
      "loss": 0.1959,
      "step": 240
    },
    {
      "epoch": 9.52,
      "learning_rate": 3.774928774928775e-05,
      "loss": 0.2016,
      "step": 250
    },
    {
      "epoch": 9.9,
      "learning_rate": 3.7037037037037037e-05,
      "loss": 0.1955,
      "step": 260
    },
    {
      "epoch": 9.98,
      "eval_accuracy": 0.9572192513368984,
      "eval_f1_score": 0.9724137931034483,
      "eval_loss": 0.11462008953094482,
      "eval_precision": 0.9657534246575342,
      "eval_recall": 0.9791666666666666,
      "eval_runtime": 2.2006,
      "eval_samples_per_second": 169.954,
      "eval_steps_per_second": 5.453,
      "step": 262
    },
    {
      "epoch": 10.29,
      "learning_rate": 3.6324786324786323e-05,
      "loss": 0.2038,
      "step": 270
    },
    {
      "epoch": 10.67,
      "learning_rate": 3.561253561253561e-05,
      "loss": 0.2083,
      "step": 280
    },
    {
      "epoch": 10.97,
      "eval_accuracy": 0.9652406417112299,
      "eval_f1_score": 0.9772329246935202,
      "eval_loss": 0.1083158627152443,
      "eval_precision": 0.9858657243816255,
      "eval_recall": 0.96875,
      "eval_runtime": 2.2496,
      "eval_samples_per_second": 166.25,
      "eval_steps_per_second": 5.334,
      "step": 288
    },
    {
      "epoch": 11.05,
      "learning_rate": 3.4900284900284904e-05,
      "loss": 0.2107,
      "step": 290
    },
    {
      "epoch": 11.43,
      "learning_rate": 3.418803418803419e-05,
      "loss": 0.1725,
      "step": 300
    },
    {
      "epoch": 11.81,
      "learning_rate": 3.347578347578348e-05,
      "loss": 0.1886,
      "step": 310
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9598930481283422,
      "eval_f1_score": 0.9740932642487047,
      "eval_loss": 0.10481037944555283,
      "eval_precision": 0.9690721649484536,
      "eval_recall": 0.9791666666666666,
      "eval_runtime": 2.1714,
      "eval_samples_per_second": 172.242,
      "eval_steps_per_second": 5.526,
      "step": 315
    },
    {
      "epoch": 12.19,
      "learning_rate": 3.2763532763532764e-05,
      "loss": 0.1703,
      "step": 320
    },
    {
      "epoch": 12.57,
      "learning_rate": 3.205128205128206e-05,
      "loss": 0.1817,
      "step": 330
    },
    {
      "epoch": 12.95,
      "learning_rate": 3.133903133903134e-05,
      "loss": 0.1618,
      "step": 340
    },
    {
      "epoch": 12.99,
      "eval_accuracy": 0.9625668449197861,
      "eval_f1_score": 0.9756944444444444,
      "eval_loss": 0.10334747284650803,
      "eval_precision": 0.9756944444444444,
      "eval_recall": 0.9756944444444444,
      "eval_runtime": 2.1514,
      "eval_samples_per_second": 173.842,
      "eval_steps_per_second": 5.578,
      "step": 341
    },
    {
      "epoch": 13.33,
      "learning_rate": 3.0626780626780625e-05,
      "loss": 0.1733,
      "step": 350
    },
    {
      "epoch": 13.71,
      "learning_rate": 2.9914529914529915e-05,
      "loss": 0.1908,
      "step": 360
    },
    {
      "epoch": 13.98,
      "eval_accuracy": 0.9598930481283422,
      "eval_f1_score": 0.9739130434782608,
      "eval_loss": 0.10439594089984894,
      "eval_precision": 0.975609756097561,
      "eval_recall": 0.9722222222222222,
      "eval_runtime": 2.2063,
      "eval_samples_per_second": 169.517,
      "eval_steps_per_second": 5.439,
      "step": 367
    },
    {
      "epoch": 14.1,
      "learning_rate": 2.9202279202279202e-05,
      "loss": 0.1677,
      "step": 370
    },
    {
      "epoch": 14.48,
      "learning_rate": 2.8490028490028492e-05,
      "loss": 0.1641,
      "step": 380
    },
    {
      "epoch": 14.86,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1594,
      "step": 390
    },
    {
      "epoch": 14.97,
      "eval_accuracy": 0.9625668449197861,
      "eval_f1_score": 0.9757785467128027,
      "eval_loss": 0.09152617305517197,
      "eval_precision": 0.9724137931034482,
      "eval_recall": 0.9791666666666666,
      "eval_runtime": 2.146,
      "eval_samples_per_second": 174.274,
      "eval_steps_per_second": 5.592,
      "step": 393
    },
    {
      "epoch": 15.24,
      "learning_rate": 2.706552706552707e-05,
      "loss": 0.1621,
      "step": 400
    },
    {
      "epoch": 15.62,
      "learning_rate": 2.6353276353276356e-05,
      "loss": 0.1698,
      "step": 410
    },
    {
      "epoch": 16.0,
      "learning_rate": 2.564102564102564e-05,
      "loss": 0.1474,
      "step": 420
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9759358288770054,
      "eval_f1_score": 0.9842381786339754,
      "eval_loss": 0.09159436821937561,
      "eval_precision": 0.9929328621908127,
      "eval_recall": 0.9756944444444444,
      "eval_runtime": 2.1573,
      "eval_samples_per_second": 173.367,
      "eval_steps_per_second": 5.563,
      "step": 420
    },
    {
      "epoch": 16.38,
      "learning_rate": 2.492877492877493e-05,
      "loss": 0.1326,
      "step": 430
    },
    {
      "epoch": 16.76,
      "learning_rate": 2.4216524216524217e-05,
      "loss": 0.1734,
      "step": 440
    },
    {
      "epoch": 16.99,
      "eval_accuracy": 0.9652406417112299,
      "eval_f1_score": 0.9773123909249563,
      "eval_loss": 0.09513744711875916,
      "eval_precision": 0.9824561403508771,
      "eval_recall": 0.9722222222222222,
      "eval_runtime": 2.1734,
      "eval_samples_per_second": 172.078,
      "eval_steps_per_second": 5.521,
      "step": 446
    },
    {
      "epoch": 17.14,
      "learning_rate": 2.3504273504273504e-05,
      "loss": 0.1827,
      "step": 450
    },
    {
      "epoch": 17.52,
      "learning_rate": 2.2792022792022794e-05,
      "loss": 0.1419,
      "step": 460
    },
    {
      "epoch": 17.9,
      "learning_rate": 2.207977207977208e-05,
      "loss": 0.1484,
      "step": 470
    },
    {
      "epoch": 17.98,
      "eval_accuracy": 0.9705882352941176,
      "eval_f1_score": 0.9808695652173912,
      "eval_loss": 0.10494749993085861,
      "eval_precision": 0.9825783972125436,
      "eval_recall": 0.9791666666666666,
      "eval_runtime": 2.155,
      "eval_samples_per_second": 173.55,
      "eval_steps_per_second": 5.568,
      "step": 472
    },
    {
      "epoch": 18.29,
      "learning_rate": 2.1367521367521368e-05,
      "loss": 0.1515,
      "step": 480
    },
    {
      "epoch": 18.67,
      "learning_rate": 2.0655270655270654e-05,
      "loss": 0.1495,
      "step": 490
    },
    {
      "epoch": 18.97,
      "eval_accuracy": 0.9679144385026738,
      "eval_f1_score": 0.9790940766550522,
      "eval_loss": 0.09304243326187134,
      "eval_precision": 0.9825174825174825,
      "eval_recall": 0.9756944444444444,
      "eval_runtime": 2.1676,
      "eval_samples_per_second": 172.538,
      "eval_steps_per_second": 5.536,
      "step": 498
    },
    {
      "epoch": 19.05,
      "learning_rate": 1.9943019943019945e-05,
      "loss": 0.1525,
      "step": 500
    },
    {
      "epoch": 19.43,
      "learning_rate": 1.923076923076923e-05,
      "loss": 0.1575,
      "step": 510
    },
    {
      "epoch": 19.81,
      "learning_rate": 1.8518518518518518e-05,
      "loss": 0.1385,
      "step": 520
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.9625668449197861,
      "eval_f1_score": 0.9758620689655172,
      "eval_loss": 0.095456063747406,
      "eval_precision": 0.9691780821917808,
      "eval_recall": 0.9826388888888888,
      "eval_runtime": 2.217,
      "eval_samples_per_second": 168.695,
      "eval_steps_per_second": 5.413,
      "step": 525
    },
    {
      "epoch": 20.19,
      "learning_rate": 1.7806267806267805e-05,
      "loss": 0.139,
      "step": 530
    },
    {
      "epoch": 20.57,
      "learning_rate": 1.7094017094017095e-05,
      "loss": 0.1567,
      "step": 540
    },
    {
      "epoch": 20.95,
      "learning_rate": 1.6381766381766382e-05,
      "loss": 0.1492,
      "step": 550
    },
    {
      "epoch": 20.99,
      "eval_accuracy": 0.9598930481283422,
      "eval_f1_score": 0.9740932642487047,
      "eval_loss": 0.09114021807909012,
      "eval_precision": 0.9690721649484536,
      "eval_recall": 0.9791666666666666,
      "eval_runtime": 2.1474,
      "eval_samples_per_second": 174.163,
      "eval_steps_per_second": 5.588,
      "step": 551
    },
    {
      "epoch": 21.33,
      "learning_rate": 1.566951566951567e-05,
      "loss": 0.1664,
      "step": 560
    },
    {
      "epoch": 21.71,
      "learning_rate": 1.4957264957264958e-05,
      "loss": 0.1401,
      "step": 570
    },
    {
      "epoch": 21.98,
      "eval_accuracy": 0.9705882352941176,
      "eval_f1_score": 0.9808695652173912,
      "eval_loss": 0.09274759143590927,
      "eval_precision": 0.9825783972125436,
      "eval_recall": 0.9791666666666666,
      "eval_runtime": 2.1665,
      "eval_samples_per_second": 172.63,
      "eval_steps_per_second": 5.539,
      "step": 577
    },
    {
      "epoch": 22.1,
      "learning_rate": 1.4245014245014246e-05,
      "loss": 0.1538,
      "step": 580
    },
    {
      "epoch": 22.48,
      "learning_rate": 1.3532763532763535e-05,
      "loss": 0.1245,
      "step": 590
    },
    {
      "epoch": 22.86,
      "learning_rate": 1.282051282051282e-05,
      "loss": 0.1288,
      "step": 600
    },
    {
      "epoch": 22.97,
      "eval_accuracy": 0.9705882352941176,
      "eval_f1_score": 0.9808695652173912,
      "eval_loss": 0.09401033818721771,
      "eval_precision": 0.9825783972125436,
      "eval_recall": 0.9791666666666666,
      "eval_runtime": 2.1911,
      "eval_samples_per_second": 170.692,
      "eval_steps_per_second": 5.477,
      "step": 603
    },
    {
      "epoch": 23.24,
      "learning_rate": 1.2108262108262108e-05,
      "loss": 0.1422,
      "step": 610
    },
    {
      "epoch": 23.62,
      "learning_rate": 1.1396011396011397e-05,
      "loss": 0.1262,
      "step": 620
    },
    {
      "epoch": 24.0,
      "learning_rate": 1.0683760683760684e-05,
      "loss": 0.1304,
      "step": 630
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.9652406417112299,
      "eval_f1_score": 0.9775474956822107,
      "eval_loss": 0.09132420271635056,
      "eval_precision": 0.9725085910652921,
      "eval_recall": 0.9826388888888888,
      "eval_runtime": 2.1685,
      "eval_samples_per_second": 172.467,
      "eval_steps_per_second": 5.534,
      "step": 630
    },
    {
      "epoch": 24.38,
      "learning_rate": 9.971509971509972e-06,
      "loss": 0.1542,
      "step": 640
    },
    {
      "epoch": 24.76,
      "learning_rate": 9.259259259259259e-06,
      "loss": 0.14,
      "step": 650
    },
    {
      "epoch": 24.99,
      "eval_accuracy": 0.9652406417112299,
      "eval_f1_score": 0.9776247848537005,
      "eval_loss": 0.09787322580814362,
      "eval_precision": 0.9692832764505119,
      "eval_recall": 0.9861111111111112,
      "eval_runtime": 2.1719,
      "eval_samples_per_second": 172.201,
      "eval_steps_per_second": 5.525,
      "step": 656
    },
    {
      "epoch": 25.14,
      "learning_rate": 8.547008547008548e-06,
      "loss": 0.1529,
      "step": 660
    },
    {
      "epoch": 25.52,
      "learning_rate": 7.834757834757835e-06,
      "loss": 0.1367,
      "step": 670
    },
    {
      "epoch": 25.9,
      "learning_rate": 7.122507122507123e-06,
      "loss": 0.1461,
      "step": 680
    },
    {
      "epoch": 25.98,
      "eval_accuracy": 0.9705882352941176,
      "eval_f1_score": 0.9810017271157168,
      "eval_loss": 0.08736571669578552,
      "eval_precision": 0.9759450171821306,
      "eval_recall": 0.9861111111111112,
      "eval_runtime": 2.1664,
      "eval_samples_per_second": 172.637,
      "eval_steps_per_second": 5.539,
      "step": 682
    },
    {
      "epoch": 26.29,
      "learning_rate": 6.41025641025641e-06,
      "loss": 0.1349,
      "step": 690
    },
    {
      "epoch": 26.67,
      "learning_rate": 5.6980056980056985e-06,
      "loss": 0.1429,
      "step": 700
    },
    {
      "epoch": 26.97,
      "eval_accuracy": 0.9705882352941176,
      "eval_f1_score": 0.9808027923211169,
      "eval_loss": 0.08370037376880646,
      "eval_precision": 0.9859649122807017,
      "eval_recall": 0.9756944444444444,
      "eval_runtime": 2.1544,
      "eval_samples_per_second": 173.601,
      "eval_steps_per_second": 5.57,
      "step": 708
    },
    {
      "epoch": 27.05,
      "learning_rate": 4.985754985754986e-06,
      "loss": 0.134,
      "step": 710
    },
    {
      "epoch": 27.43,
      "learning_rate": 4.273504273504274e-06,
      "loss": 0.1366,
      "step": 720
    },
    {
      "epoch": 27.81,
      "learning_rate": 3.5612535612535615e-06,
      "loss": 0.1444,
      "step": 730
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.9679144385026738,
      "eval_f1_score": 0.9791666666666666,
      "eval_loss": 0.08762019872665405,
      "eval_precision": 0.9791666666666666,
      "eval_recall": 0.9791666666666666,
      "eval_runtime": 2.1635,
      "eval_samples_per_second": 172.868,
      "eval_steps_per_second": 5.547,
      "step": 735
    },
    {
      "epoch": 28.19,
      "learning_rate": 2.8490028490028492e-06,
      "loss": 0.1167,
      "step": 740
    },
    {
      "epoch": 28.57,
      "learning_rate": 2.136752136752137e-06,
      "loss": 0.1129,
      "step": 750
    },
    {
      "epoch": 28.95,
      "learning_rate": 1.4245014245014246e-06,
      "loss": 0.145,
      "step": 760
    },
    {
      "epoch": 28.99,
      "eval_accuracy": 0.9705882352941176,
      "eval_f1_score": 0.9808695652173912,
      "eval_loss": 0.09031202644109726,
      "eval_precision": 0.9825783972125436,
      "eval_recall": 0.9791666666666666,
      "eval_runtime": 2.1632,
      "eval_samples_per_second": 172.893,
      "eval_steps_per_second": 5.547,
      "step": 761
    },
    {
      "epoch": 29.33,
      "learning_rate": 7.122507122507123e-07,
      "loss": 0.1334,
      "step": 770
    },
    {
      "epoch": 29.71,
      "learning_rate": 0.0,
      "loss": 0.1445,
      "step": 780
    },
    {
      "epoch": 29.71,
      "eval_accuracy": 0.9679144385026738,
      "eval_f1_score": 0.9790940766550522,
      "eval_loss": 0.08815235644578934,
      "eval_precision": 0.9825174825174825,
      "eval_recall": 0.9756944444444444,
      "eval_runtime": 3.4052,
      "eval_samples_per_second": 109.832,
      "eval_steps_per_second": 3.524,
      "step": 780
    },
    {
      "epoch": 29.71,
      "step": 780,
      "total_flos": 4.5315392030480794e+17,
      "train_loss": 0.2131147768252935,
      "train_runtime": 1561.1697,
      "train_samples_per_second": 64.529,
      "train_steps_per_second": 0.5
    }
  ],
  "logging_steps": 10,
  "max_steps": 780,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 4.5315392030480794e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}