{
  "best_metric": 0.8550169467926025,
  "best_model_checkpoint": "./beans_outputs/checkpoint-1950",
  "epoch": 15.0,
  "eval_steps": 500,
  "global_step": 1950,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07692307692307693,
      "grad_norm": 2.0955963134765625,
      "learning_rate": 1.9692307692307696e-05,
      "loss": 1.1239,
      "step": 10
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 1.8236017227172852,
      "learning_rate": 1.9384615384615386e-05,
      "loss": 1.1222,
      "step": 20
    },
    {
      "epoch": 0.23076923076923078,
      "grad_norm": 1.9866633415222168,
      "learning_rate": 1.907692307692308e-05,
      "loss": 1.1165,
      "step": 30
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 2.4556338787078857,
      "learning_rate": 1.876923076923077e-05,
      "loss": 1.1047,
      "step": 40
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 1.6562598943710327,
      "learning_rate": 1.8461538461538465e-05,
      "loss": 1.1085,
      "step": 50
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 1.501336932182312,
      "learning_rate": 1.8153846153846155e-05,
      "loss": 1.1048,
      "step": 60
    },
    {
      "epoch": 0.5384615384615384,
      "grad_norm": 2.606107234954834,
      "learning_rate": 1.784615384615385e-05,
      "loss": 1.1031,
      "step": 70
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 2.634584426879883,
      "learning_rate": 1.753846153846154e-05,
      "loss": 1.0916,
      "step": 80
    },
    {
      "epoch": 0.6923076923076923,
      "grad_norm": 2.039886236190796,
      "learning_rate": 1.7230769230769234e-05,
      "loss": 1.095,
      "step": 90
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 2.4800949096679688,
      "learning_rate": 1.6923076923076924e-05,
      "loss": 1.0916,
      "step": 100
    },
    {
      "epoch": 0.8461538461538461,
      "grad_norm": 1.6766040325164795,
      "learning_rate": 1.6615384615384618e-05,
      "loss": 1.0959,
      "step": 110
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.2848353385925293,
      "learning_rate": 1.630769230769231e-05,
      "loss": 1.1005,
      "step": 120
    },
    {
      "epoch": 1.0,
      "grad_norm": 4.07199239730835,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.0881,
      "step": 130
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.41353383458646614,
      "eval_loss": 1.0901767015457153,
      "eval_runtime": 0.7894,
      "eval_samples_per_second": 168.48,
      "eval_steps_per_second": 21.535,
      "step": 130
    },
    {
      "epoch": 1.0769230769230769,
      "grad_norm": 2.0711560249328613,
      "learning_rate": 1.5692307692307693e-05,
      "loss": 1.0794,
      "step": 140
    },
    {
      "epoch": 1.1538461538461537,
      "grad_norm": 2.499732494354248,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 1.0755,
      "step": 150
    },
    {
      "epoch": 1.2307692307692308,
      "grad_norm": 1.4569323062896729,
      "learning_rate": 1.5076923076923078e-05,
      "loss": 1.0807,
      "step": 160
    },
    {
      "epoch": 1.3076923076923077,
      "grad_norm": 2.351478338241577,
      "learning_rate": 1.4769230769230772e-05,
      "loss": 1.0965,
      "step": 170
    },
    {
      "epoch": 1.3846153846153846,
      "grad_norm": 2.1514322757720947,
      "learning_rate": 1.4461538461538462e-05,
      "loss": 1.0839,
      "step": 180
    },
    {
      "epoch": 1.4615384615384617,
      "grad_norm": 2.151601791381836,
      "learning_rate": 1.4153846153846156e-05,
      "loss": 1.0837,
      "step": 190
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 1.797500729560852,
      "learning_rate": 1.3846153846153847e-05,
      "loss": 1.0807,
      "step": 200
    },
    {
      "epoch": 1.6153846153846154,
      "grad_norm": 1.7583892345428467,
      "learning_rate": 1.353846153846154e-05,
      "loss": 1.073,
      "step": 210
    },
    {
      "epoch": 1.6923076923076923,
      "grad_norm": 3.5006496906280518,
      "learning_rate": 1.3230769230769231e-05,
      "loss": 1.0741,
      "step": 220
    },
    {
      "epoch": 1.7692307692307692,
      "grad_norm": 2.0179672241210938,
      "learning_rate": 1.2923076923076925e-05,
      "loss": 1.077,
      "step": 230
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 2.057086944580078,
      "learning_rate": 1.2615384615384616e-05,
      "loss": 1.0925,
      "step": 240
    },
    {
      "epoch": 1.9230769230769231,
      "grad_norm": 1.9710103273391724,
      "learning_rate": 1.230769230769231e-05,
      "loss": 1.0757,
      "step": 250
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.9315028190612793,
      "learning_rate": 1.2e-05,
      "loss": 1.0716,
      "step": 260
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5037593984962406,
      "eval_loss": 1.0685255527496338,
      "eval_runtime": 0.7711,
      "eval_samples_per_second": 172.471,
      "eval_steps_per_second": 22.045,
      "step": 260
    },
    {
      "epoch": 2.076923076923077,
      "grad_norm": 2.183527946472168,
      "learning_rate": 1.1692307692307694e-05,
      "loss": 1.0786,
      "step": 270
    },
    {
      "epoch": 2.1538461538461537,
      "grad_norm": 2.379652500152588,
      "learning_rate": 1.1384615384615385e-05,
      "loss": 1.0676,
      "step": 280
    },
    {
      "epoch": 2.230769230769231,
      "grad_norm": 2.387296438217163,
      "learning_rate": 1.1076923076923079e-05,
      "loss": 1.064,
      "step": 290
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 2.8164947032928467,
      "learning_rate": 1.076923076923077e-05,
      "loss": 1.0548,
      "step": 300
    },
    {
      "epoch": 2.3846153846153846,
      "grad_norm": 1.849363088607788,
      "learning_rate": 1.0461538461538463e-05,
      "loss": 1.0693,
      "step": 310
    },
    {
      "epoch": 2.4615384615384617,
      "grad_norm": 2.0274300575256348,
      "learning_rate": 1.0153846153846154e-05,
      "loss": 1.0604,
      "step": 320
    },
    {
      "epoch": 2.5384615384615383,
      "grad_norm": 1.4544223546981812,
      "learning_rate": 9.846153846153848e-06,
      "loss": 1.0458,
      "step": 330
    },
    {
      "epoch": 2.6153846153846154,
      "grad_norm": 2.1712183952331543,
      "learning_rate": 9.53846153846154e-06,
      "loss": 1.0609,
      "step": 340
    },
    {
      "epoch": 2.6923076923076925,
      "grad_norm": 1.922677993774414,
      "learning_rate": 9.230769230769232e-06,
      "loss": 1.0646,
      "step": 350
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 1.8288860321044922,
      "learning_rate": 8.923076923076925e-06,
      "loss": 1.0597,
      "step": 360
    },
    {
      "epoch": 2.8461538461538463,
      "grad_norm": 2.123480796813965,
      "learning_rate": 8.615384615384617e-06,
      "loss": 1.052,
      "step": 370
    },
    {
      "epoch": 2.9230769230769234,
      "grad_norm": 1.820168137550354,
      "learning_rate": 8.307692307692309e-06,
      "loss": 1.0434,
      "step": 380
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.973505020141602,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.061,
      "step": 390
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6240601503759399,
      "eval_loss": 1.0459144115447998,
      "eval_runtime": 0.7618,
      "eval_samples_per_second": 174.588,
      "eval_steps_per_second": 22.316,
      "step": 390
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 2.1673223972320557,
      "learning_rate": 7.692307692307694e-06,
      "loss": 1.0616,
      "step": 400
    },
    {
      "epoch": 3.1538461538461537,
      "grad_norm": 1.8567888736724854,
      "learning_rate": 7.384615384615386e-06,
      "loss": 1.0501,
      "step": 410
    },
    {
      "epoch": 3.230769230769231,
      "grad_norm": 2.0640571117401123,
      "learning_rate": 7.076923076923078e-06,
      "loss": 1.0611,
      "step": 420
    },
    {
      "epoch": 3.3076923076923075,
      "grad_norm": 2.215384006500244,
      "learning_rate": 6.76923076923077e-06,
      "loss": 1.0469,
      "step": 430
    },
    {
      "epoch": 3.3846153846153846,
      "grad_norm": 2.1000049114227295,
      "learning_rate": 6.461538461538463e-06,
      "loss": 1.0499,
      "step": 440
    },
    {
      "epoch": 3.4615384615384617,
      "grad_norm": 1.7218382358551025,
      "learning_rate": 6.153846153846155e-06,
      "loss": 1.0564,
      "step": 450
    },
    {
      "epoch": 3.5384615384615383,
      "grad_norm": 2.3569300174713135,
      "learning_rate": 5.846153846153847e-06,
      "loss": 1.0599,
      "step": 460
    },
    {
      "epoch": 3.6153846153846154,
      "grad_norm": 1.5210909843444824,
      "learning_rate": 5.538461538461539e-06,
      "loss": 1.0367,
      "step": 470
    },
    {
      "epoch": 3.6923076923076925,
      "grad_norm": 2.7621657848358154,
      "learning_rate": 5.230769230769232e-06,
      "loss": 1.0421,
      "step": 480
    },
    {
      "epoch": 3.769230769230769,
      "grad_norm": 1.5097808837890625,
      "learning_rate": 4.923076923076924e-06,
      "loss": 1.0362,
      "step": 490
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 1.5118447542190552,
      "learning_rate": 4.615384615384616e-06,
      "loss": 1.0572,
      "step": 500
    },
    {
      "epoch": 3.9230769230769234,
      "grad_norm": 1.7513490915298462,
      "learning_rate": 4.307692307692308e-06,
      "loss": 1.0361,
      "step": 510
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.398025035858154,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.0514,
      "step": 520
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6015037593984962,
      "eval_loss": 1.0407124757766724,
      "eval_runtime": 0.7726,
      "eval_samples_per_second": 172.155,
      "eval_steps_per_second": 22.005,
      "step": 520
    },
    {
      "epoch": 4.076923076923077,
      "grad_norm": 2.5516345500946045,
      "learning_rate": 3.692307692307693e-06,
      "loss": 1.0529,
      "step": 530
    },
    {
      "epoch": 4.153846153846154,
      "grad_norm": 1.6976008415222168,
      "learning_rate": 3.384615384615385e-06,
      "loss": 1.0472,
      "step": 540
    },
    {
      "epoch": 4.230769230769231,
      "grad_norm": 2.5672519207000732,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 1.0565,
      "step": 550
    },
    {
      "epoch": 4.3076923076923075,
      "grad_norm": 2.166529655456543,
      "learning_rate": 2.7692307692307697e-06,
      "loss": 1.0619,
      "step": 560
    },
    {
      "epoch": 4.384615384615385,
      "grad_norm": 1.961472511291504,
      "learning_rate": 2.461538461538462e-06,
      "loss": 1.0322,
      "step": 570
    },
    {
      "epoch": 4.461538461538462,
      "grad_norm": 2.392319440841675,
      "learning_rate": 2.153846153846154e-06,
      "loss": 1.0388,
      "step": 580
    },
    {
      "epoch": 4.538461538461538,
      "grad_norm": 2.3034205436706543,
      "learning_rate": 1.8461538461538465e-06,
      "loss": 1.0358,
      "step": 590
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 2.037050247192383,
      "learning_rate": 1.5384615384615387e-06,
      "loss": 1.0334,
      "step": 600
    },
    {
      "epoch": 4.6923076923076925,
      "grad_norm": 3.0737335681915283,
      "learning_rate": 1.230769230769231e-06,
      "loss": 1.0506,
      "step": 610
    },
    {
      "epoch": 4.769230769230769,
      "grad_norm": 2.1824796199798584,
      "learning_rate": 9.230769230769232e-07,
      "loss": 1.0516,
      "step": 620
    },
    {
      "epoch": 4.846153846153846,
      "grad_norm": 1.9239214658737183,
      "learning_rate": 6.153846153846155e-07,
      "loss": 1.0399,
      "step": 630
    },
    {
      "epoch": 4.923076923076923,
      "grad_norm": 2.267302989959717,
      "learning_rate": 3.0769230769230774e-07,
      "loss": 1.0374,
      "step": 640
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.4957404136657715,
      "learning_rate": 0.0,
      "loss": 1.05,
      "step": 650
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6766917293233082,
      "eval_loss": 1.0332472324371338,
      "eval_runtime": 0.8208,
      "eval_samples_per_second": 162.035,
      "eval_steps_per_second": 20.711,
      "step": 650
    },
    {
      "epoch": 5.076923076923077,
      "grad_norm": 1.9310355186462402,
      "learning_rate": 1.3230769230769231e-05,
      "loss": 1.0281,
      "step": 660
    },
    {
      "epoch": 5.153846153846154,
      "grad_norm": 2.36603045463562,
      "learning_rate": 1.312820512820513e-05,
      "loss": 1.0587,
      "step": 670
    },
    {
      "epoch": 5.230769230769231,
      "grad_norm": 2.0758917331695557,
      "learning_rate": 1.3025641025641027e-05,
      "loss": 1.0534,
      "step": 680
    },
    {
      "epoch": 5.3076923076923075,
      "grad_norm": 2.549725294113159,
      "learning_rate": 1.2923076923076925e-05,
      "loss": 1.0421,
      "step": 690
    },
    {
      "epoch": 5.384615384615385,
      "grad_norm": 2.7183680534362793,
      "learning_rate": 1.2820512820512823e-05,
      "loss": 1.035,
      "step": 700
    },
    {
      "epoch": 5.461538461538462,
      "grad_norm": 1.7176955938339233,
      "learning_rate": 1.2717948717948718e-05,
      "loss": 1.0268,
      "step": 710
    },
    {
      "epoch": 5.538461538461538,
      "grad_norm": 1.8257861137390137,
      "learning_rate": 1.2615384615384616e-05,
      "loss": 1.0302,
      "step": 720
    },
    {
      "epoch": 5.615384615384615,
      "grad_norm": 2.50368595123291,
      "learning_rate": 1.2512820512820514e-05,
      "loss": 1.0331,
      "step": 730
    },
    {
      "epoch": 5.6923076923076925,
      "grad_norm": 2.4315121173858643,
      "learning_rate": 1.2410256410256412e-05,
      "loss": 1.0208,
      "step": 740
    },
    {
      "epoch": 5.769230769230769,
      "grad_norm": 2.043854236602783,
      "learning_rate": 1.230769230769231e-05,
      "loss": 1.0241,
      "step": 750
    },
    {
      "epoch": 5.846153846153846,
      "grad_norm": 2.0800740718841553,
      "learning_rate": 1.2205128205128208e-05,
      "loss": 1.0251,
      "step": 760
    },
    {
      "epoch": 5.923076923076923,
      "grad_norm": 3.4356396198272705,
      "learning_rate": 1.2102564102564102e-05,
      "loss": 1.0104,
      "step": 770
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.49416971206665,
      "learning_rate": 1.2e-05,
      "loss": 1.0357,
      "step": 780
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6541353383458647,
      "eval_loss": 1.0109117031097412,
      "eval_runtime": 0.7801,
      "eval_samples_per_second": 170.483,
      "eval_steps_per_second": 21.791,
      "step": 780
    },
    {
      "epoch": 6.076923076923077,
      "grad_norm": 2.650513172149658,
      "learning_rate": 1.1897435897435898e-05,
      "loss": 1.0148,
      "step": 790
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 2.3682632446289062,
      "learning_rate": 1.1794871794871796e-05,
      "loss": 1.0094,
      "step": 800
    },
    {
      "epoch": 6.230769230769231,
      "grad_norm": 1.6716077327728271,
      "learning_rate": 1.1692307692307694e-05,
      "loss": 1.0097,
      "step": 810
    },
    {
      "epoch": 6.3076923076923075,
      "grad_norm": 2.4839890003204346,
      "learning_rate": 1.1589743589743592e-05,
      "loss": 1.01,
      "step": 820
    },
    {
      "epoch": 6.384615384615385,
      "grad_norm": 2.4004769325256348,
      "learning_rate": 1.1487179487179487e-05,
      "loss": 1.0104,
      "step": 830
    },
    {
      "epoch": 6.461538461538462,
      "grad_norm": 2.9597084522247314,
      "learning_rate": 1.1384615384615385e-05,
      "loss": 1.0137,
      "step": 840
    },
    {
      "epoch": 6.538461538461538,
      "grad_norm": 2.680335760116577,
      "learning_rate": 1.1282051282051283e-05,
      "loss": 1.0147,
      "step": 850
    },
    {
      "epoch": 6.615384615384615,
      "grad_norm": 1.7677160501480103,
      "learning_rate": 1.117948717948718e-05,
      "loss": 0.9947,
      "step": 860
    },
    {
      "epoch": 6.6923076923076925,
      "grad_norm": 2.0484132766723633,
      "learning_rate": 1.1076923076923079e-05,
      "loss": 1.0169,
      "step": 870
    },
    {
      "epoch": 6.769230769230769,
      "grad_norm": 2.1910479068756104,
      "learning_rate": 1.0974358974358977e-05,
      "loss": 1.0024,
      "step": 880
    },
    {
      "epoch": 6.846153846153846,
      "grad_norm": 2.181236743927002,
      "learning_rate": 1.0871794871794871e-05,
      "loss": 0.9962,
      "step": 890
    },
    {
      "epoch": 6.923076923076923,
      "grad_norm": 2.898885488510132,
      "learning_rate": 1.076923076923077e-05,
      "loss": 1.012,
      "step": 900
    },
    {
      "epoch": 7.0,
      "grad_norm": 4.950052738189697,
      "learning_rate": 1.0666666666666667e-05,
      "loss": 1.0012,
      "step": 910
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7368421052631579,
      "eval_loss": 0.981479287147522,
      "eval_runtime": 0.7693,
      "eval_samples_per_second": 172.886,
      "eval_steps_per_second": 22.098,
      "step": 910
    },
    {
      "epoch": 7.076923076923077,
      "grad_norm": 2.692753553390503,
      "learning_rate": 1.0564102564102565e-05,
      "loss": 0.9889,
      "step": 920
    },
    {
      "epoch": 7.153846153846154,
      "grad_norm": 2.9175124168395996,
      "learning_rate": 1.0461538461538463e-05,
      "loss": 0.9911,
      "step": 930
    },
    {
      "epoch": 7.230769230769231,
      "grad_norm": 3.221527099609375,
      "learning_rate": 1.0358974358974361e-05,
      "loss": 0.9827,
      "step": 940
    },
    {
      "epoch": 7.3076923076923075,
      "grad_norm": 2.507923126220703,
      "learning_rate": 1.0256410256410256e-05,
      "loss": 0.9919,
      "step": 950
    },
    {
      "epoch": 7.384615384615385,
      "grad_norm": 2.4533870220184326,
      "learning_rate": 1.0153846153846154e-05,
      "loss": 0.9962,
      "step": 960
    },
    {
      "epoch": 7.461538461538462,
      "grad_norm": 2.1032631397247314,
      "learning_rate": 1.0051282051282052e-05,
      "loss": 0.9751,
      "step": 970
    },
    {
      "epoch": 7.538461538461538,
      "grad_norm": 2.5848186016082764,
      "learning_rate": 9.94871794871795e-06,
      "loss": 0.9939,
      "step": 980
    },
    {
      "epoch": 7.615384615384615,
      "grad_norm": 2.17742919921875,
      "learning_rate": 9.846153846153848e-06,
      "loss": 0.9745,
      "step": 990
    },
    {
      "epoch": 7.6923076923076925,
      "grad_norm": 1.9953967332839966,
      "learning_rate": 9.743589743589744e-06,
      "loss": 0.9665,
      "step": 1000
    },
    {
      "epoch": 7.769230769230769,
      "grad_norm": 3.0263218879699707,
      "learning_rate": 9.641025641025642e-06,
      "loss": 0.9646,
      "step": 1010
    },
    {
      "epoch": 7.846153846153846,
      "grad_norm": 2.3735406398773193,
      "learning_rate": 9.53846153846154e-06,
      "loss": 0.9836,
      "step": 1020
    },
    {
      "epoch": 7.923076923076923,
      "grad_norm": 2.548480272293091,
      "learning_rate": 9.435897435897436e-06,
      "loss": 0.9546,
      "step": 1030
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.8450028896331787,
      "learning_rate": 9.333333333333334e-06,
      "loss": 0.9932,
      "step": 1040
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7669172932330827,
      "eval_loss": 0.9549766778945923,
      "eval_runtime": 0.7536,
      "eval_samples_per_second": 176.486,
      "eval_steps_per_second": 22.558,
      "step": 1040
    },
    {
      "epoch": 8.076923076923077,
      "grad_norm": 1.6870847940444946,
      "learning_rate": 9.230769230769232e-06,
      "loss": 0.9752,
      "step": 1050
    },
    {
      "epoch": 8.153846153846153,
      "grad_norm": 2.1122217178344727,
      "learning_rate": 9.128205128205129e-06,
      "loss": 0.9573,
      "step": 1060
    },
    {
      "epoch": 8.23076923076923,
      "grad_norm": 4.1552886962890625,
      "learning_rate": 9.025641025641027e-06,
      "loss": 0.9764,
      "step": 1070
    },
    {
      "epoch": 8.307692307692308,
      "grad_norm": 1.7864203453063965,
      "learning_rate": 8.923076923076925e-06,
      "loss": 0.9434,
      "step": 1080
    },
    {
      "epoch": 8.384615384615385,
      "grad_norm": 2.2091946601867676,
      "learning_rate": 8.820512820512821e-06,
      "loss": 0.974,
      "step": 1090
    },
    {
      "epoch": 8.461538461538462,
      "grad_norm": 2.4063644409179688,
      "learning_rate": 8.717948717948719e-06,
      "loss": 0.9576,
      "step": 1100
    },
    {
      "epoch": 8.538461538461538,
      "grad_norm": 1.6061931848526,
      "learning_rate": 8.615384615384617e-06,
      "loss": 0.9588,
      "step": 1110
    },
    {
      "epoch": 8.615384615384615,
      "grad_norm": 2.8999595642089844,
      "learning_rate": 8.512820512820513e-06,
      "loss": 0.9791,
      "step": 1120
    },
    {
      "epoch": 8.692307692307692,
      "grad_norm": 3.6554131507873535,
      "learning_rate": 8.410256410256411e-06,
      "loss": 0.9629,
      "step": 1130
    },
    {
      "epoch": 8.76923076923077,
      "grad_norm": 1.7246966361999512,
      "learning_rate": 8.307692307692309e-06,
      "loss": 0.9707,
      "step": 1140
    },
    {
      "epoch": 8.846153846153847,
      "grad_norm": 2.3160033226013184,
      "learning_rate": 8.205128205128205e-06,
      "loss": 0.9578,
      "step": 1150
    },
    {
      "epoch": 8.923076923076923,
      "grad_norm": 2.685718059539795,
      "learning_rate": 8.102564102564103e-06,
      "loss": 0.972,
      "step": 1160
    },
    {
      "epoch": 9.0,
      "grad_norm": 3.6465442180633545,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.9748,
      "step": 1170
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7669172932330827,
      "eval_loss": 0.9408761858940125,
      "eval_runtime": 0.7615,
      "eval_samples_per_second": 174.644,
      "eval_steps_per_second": 22.323,
      "step": 1170
    },
    {
      "epoch": 9.076923076923077,
      "grad_norm": 1.8668149709701538,
      "learning_rate": 7.897435897435898e-06,
      "loss": 0.9712,
      "step": 1180
    },
    {
      "epoch": 9.153846153846153,
      "grad_norm": 2.042644739151001,
      "learning_rate": 7.794871794871796e-06,
      "loss": 0.9407,
      "step": 1190
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 1.967020869255066,
      "learning_rate": 7.692307692307694e-06,
      "loss": 0.9457,
      "step": 1200
    },
    {
      "epoch": 9.307692307692308,
      "grad_norm": 2.147862672805786,
      "learning_rate": 7.58974358974359e-06,
      "loss": 0.9442,
      "step": 1210
    },
    {
      "epoch": 9.384615384615385,
      "grad_norm": 1.8528053760528564,
      "learning_rate": 7.487179487179488e-06,
      "loss": 0.9526,
      "step": 1220
    },
    {
      "epoch": 9.461538461538462,
      "grad_norm": 3.2000551223754883,
      "learning_rate": 7.384615384615386e-06,
      "loss": 0.9465,
      "step": 1230
    },
    {
      "epoch": 9.538461538461538,
      "grad_norm": 2.259323835372925,
      "learning_rate": 7.282051282051282e-06,
      "loss": 0.9503,
      "step": 1240
    },
    {
      "epoch": 9.615384615384615,
      "grad_norm": 2.4054858684539795,
      "learning_rate": 7.17948717948718e-06,
      "loss": 0.9274,
      "step": 1250
    },
    {
      "epoch": 9.692307692307692,
      "grad_norm": 3.4811408519744873,
      "learning_rate": 7.076923076923078e-06,
      "loss": 0.943,
      "step": 1260
    },
    {
      "epoch": 9.76923076923077,
      "grad_norm": 1.7080141305923462,
      "learning_rate": 6.974358974358974e-06,
      "loss": 0.9247,
      "step": 1270
    },
    {
      "epoch": 9.846153846153847,
      "grad_norm": 2.0476508140563965,
      "learning_rate": 6.871794871794872e-06,
      "loss": 0.9194,
      "step": 1280
    },
    {
      "epoch": 9.923076923076923,
      "grad_norm": 2.149641990661621,
      "learning_rate": 6.76923076923077e-06,
      "loss": 0.9269,
      "step": 1290
    },
    {
      "epoch": 10.0,
      "grad_norm": 5.121323108673096,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.9113,
      "step": 1300
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7819548872180451,
      "eval_loss": 0.9149269461631775,
      "eval_runtime": 0.7775,
      "eval_samples_per_second": 171.06,
      "eval_steps_per_second": 21.865,
      "step": 1300
    },
    {
      "epoch": 10.076923076923077,
      "grad_norm": 2.06109881401062,
      "learning_rate": 6.564102564102565e-06,
      "loss": 0.925,
      "step": 1310
    },
    {
      "epoch": 10.153846153846153,
      "grad_norm": 1.9137018918991089,
      "learning_rate": 6.461538461538463e-06,
      "loss": 0.9657,
      "step": 1320
    },
    {
      "epoch": 10.23076923076923,
      "grad_norm": 2.0686280727386475,
      "learning_rate": 6.358974358974359e-06,
      "loss": 0.9565,
      "step": 1330
    },
    {
      "epoch": 10.307692307692308,
      "grad_norm": 2.046623945236206,
      "learning_rate": 6.256410256410257e-06,
      "loss": 0.918,
      "step": 1340
    },
    {
      "epoch": 10.384615384615385,
      "grad_norm": 2.281343936920166,
      "learning_rate": 6.153846153846155e-06,
      "loss": 0.9118,
      "step": 1350
    },
    {
      "epoch": 10.461538461538462,
      "grad_norm": 2.694427728652954,
      "learning_rate": 6.051282051282051e-06,
      "loss": 0.9377,
      "step": 1360
    },
    {
      "epoch": 10.538461538461538,
      "grad_norm": 2.3148765563964844,
      "learning_rate": 5.948717948717949e-06,
      "loss": 0.911,
      "step": 1370
    },
    {
      "epoch": 10.615384615384615,
      "grad_norm": 2.595669746398926,
      "learning_rate": 5.846153846153847e-06,
      "loss": 0.9146,
      "step": 1380
    },
    {
      "epoch": 10.692307692307692,
      "grad_norm": 2.136301279067993,
      "learning_rate": 5.743589743589743e-06,
      "loss": 0.9061,
      "step": 1390
    },
    {
      "epoch": 10.76923076923077,
      "grad_norm": 3.0159363746643066,
      "learning_rate": 5.641025641025641e-06,
      "loss": 0.9365,
      "step": 1400
    },
    {
      "epoch": 10.846153846153847,
      "grad_norm": 2.1136507987976074,
      "learning_rate": 5.538461538461539e-06,
      "loss": 0.9232,
      "step": 1410
    },
    {
      "epoch": 10.923076923076923,
      "grad_norm": 1.713663101196289,
      "learning_rate": 5.435897435897436e-06,
      "loss": 0.9344,
      "step": 1420
    },
    {
      "epoch": 11.0,
      "grad_norm": 4.04538631439209,
      "learning_rate": 5.333333333333334e-06,
      "loss": 0.9255,
      "step": 1430
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.7894736842105263,
      "eval_loss": 0.8905543088912964,
      "eval_runtime": 0.7659,
      "eval_samples_per_second": 173.659,
      "eval_steps_per_second": 22.197,
      "step": 1430
    },
    {
      "epoch": 11.076923076923077,
      "grad_norm": 2.04194974899292,
      "learning_rate": 5.230769230769232e-06,
      "loss": 0.9333,
      "step": 1440
    },
    {
      "epoch": 11.153846153846153,
      "grad_norm": 3.108344554901123,
      "learning_rate": 5.128205128205128e-06,
      "loss": 0.9174,
      "step": 1450
    },
    {
      "epoch": 11.23076923076923,
      "grad_norm": 2.406233072280884,
      "learning_rate": 5.025641025641026e-06,
      "loss": 0.8948,
      "step": 1460
    },
    {
      "epoch": 11.307692307692308,
      "grad_norm": 2.4100501537323,
      "learning_rate": 4.923076923076924e-06,
      "loss": 0.9155,
      "step": 1470
    },
    {
      "epoch": 11.384615384615385,
      "grad_norm": 2.7117860317230225,
      "learning_rate": 4.820512820512821e-06,
      "loss": 0.9075,
      "step": 1480
    },
    {
      "epoch": 11.461538461538462,
      "grad_norm": 2.0159695148468018,
      "learning_rate": 4.717948717948718e-06,
      "loss": 0.9338,
      "step": 1490
    },
    {
      "epoch": 11.538461538461538,
      "grad_norm": 3.280245304107666,
      "learning_rate": 4.615384615384616e-06,
      "loss": 0.9243,
      "step": 1500
    },
    {
      "epoch": 11.615384615384615,
      "grad_norm": 3.1355690956115723,
      "learning_rate": 4.512820512820513e-06,
      "loss": 0.9185,
      "step": 1510
    },
    {
      "epoch": 11.692307692307692,
      "grad_norm": 3.0900094509124756,
      "learning_rate": 4.4102564102564104e-06,
      "loss": 0.937,
      "step": 1520
    },
    {
      "epoch": 11.76923076923077,
      "grad_norm": 1.8758033514022827,
      "learning_rate": 4.307692307692308e-06,
      "loss": 0.9052,
      "step": 1530
    },
    {
      "epoch": 11.846153846153847,
      "grad_norm": 2.0586955547332764,
      "learning_rate": 4.2051282051282055e-06,
      "loss": 0.8874,
      "step": 1540
    },
    {
      "epoch": 11.923076923076923,
      "grad_norm": 2.0720062255859375,
      "learning_rate": 4.102564102564103e-06,
      "loss": 0.9141,
      "step": 1550
    },
    {
      "epoch": 12.0,
      "grad_norm": 3.183523416519165,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.8877,
      "step": 1560
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7894736842105263,
      "eval_loss": 0.8749483823776245,
      "eval_runtime": 0.7374,
      "eval_samples_per_second": 180.372,
      "eval_steps_per_second": 23.055,
      "step": 1560
    },
    {
      "epoch": 12.076923076923077,
      "grad_norm": 2.0058720111846924,
      "learning_rate": 3.897435897435898e-06,
      "loss": 0.8829,
      "step": 1570
    },
    {
      "epoch": 12.153846153846153,
      "grad_norm": 2.2991676330566406,
      "learning_rate": 3.794871794871795e-06,
      "loss": 0.9152,
      "step": 1580
    },
    {
      "epoch": 12.23076923076923,
      "grad_norm": 1.5903538465499878,
      "learning_rate": 3.692307692307693e-06,
      "loss": 0.9149,
      "step": 1590
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 1.7883615493774414,
      "learning_rate": 3.58974358974359e-06,
      "loss": 0.9163,
      "step": 1600
    },
    {
      "epoch": 12.384615384615385,
      "grad_norm": 2.2841601371765137,
      "learning_rate": 3.487179487179487e-06,
      "loss": 0.8958,
      "step": 1610
    },
    {
      "epoch": 12.461538461538462,
      "grad_norm": 2.3814501762390137,
      "learning_rate": 3.384615384615385e-06,
      "loss": 0.8918,
      "step": 1620
    },
    {
      "epoch": 12.538461538461538,
      "grad_norm": 1.9848734140396118,
      "learning_rate": 3.2820512820512823e-06,
      "loss": 0.889,
      "step": 1630
    },
    {
      "epoch": 12.615384615384615,
      "grad_norm": 1.7236778736114502,
      "learning_rate": 3.1794871794871795e-06,
      "loss": 0.8979,
      "step": 1640
    },
    {
      "epoch": 12.692307692307692,
      "grad_norm": 3.340665102005005,
      "learning_rate": 3.0769230769230774e-06,
      "loss": 0.8695,
      "step": 1650
    },
    {
      "epoch": 12.76923076923077,
      "grad_norm": 2.127927780151367,
      "learning_rate": 2.9743589743589746e-06,
      "loss": 0.9323,
      "step": 1660
    },
    {
      "epoch": 12.846153846153847,
      "grad_norm": 1.8213707208633423,
      "learning_rate": 2.8717948717948717e-06,
      "loss": 0.9178,
      "step": 1670
    },
    {
      "epoch": 12.923076923076923,
      "grad_norm": 2.0011963844299316,
      "learning_rate": 2.7692307692307697e-06,
      "loss": 0.8872,
      "step": 1680
    },
    {
      "epoch": 13.0,
      "grad_norm": 3.812871217727661,
      "learning_rate": 2.666666666666667e-06,
      "loss": 0.9032,
      "step": 1690
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.7969924812030075,
      "eval_loss": 0.8698711395263672,
      "eval_runtime": 0.7423,
      "eval_samples_per_second": 179.165,
      "eval_steps_per_second": 22.901,
      "step": 1690
    },
    {
      "epoch": 13.076923076923077,
      "grad_norm": 2.8741540908813477,
      "learning_rate": 2.564102564102564e-06,
      "loss": 0.8842,
      "step": 1700
    },
    {
      "epoch": 13.153846153846153,
      "grad_norm": 2.3278818130493164,
      "learning_rate": 2.461538461538462e-06,
      "loss": 0.9131,
      "step": 1710
    },
    {
      "epoch": 13.23076923076923,
      "grad_norm": 2.8419501781463623,
      "learning_rate": 2.358974358974359e-06,
      "loss": 0.8965,
      "step": 1720
    },
    {
      "epoch": 13.307692307692308,
      "grad_norm": 1.8506221771240234,
      "learning_rate": 2.2564102564102566e-06,
      "loss": 0.8967,
      "step": 1730
    },
    {
      "epoch": 13.384615384615385,
      "grad_norm": 2.6166839599609375,
      "learning_rate": 2.153846153846154e-06,
      "loss": 0.8785,
      "step": 1740
    },
    {
      "epoch": 13.461538461538462,
      "grad_norm": 4.287515640258789,
      "learning_rate": 2.0512820512820513e-06,
      "loss": 0.914,
      "step": 1750
    },
    {
      "epoch": 13.538461538461538,
      "grad_norm": 2.516889810562134,
      "learning_rate": 1.948717948717949e-06,
      "loss": 0.9286,
      "step": 1760
    },
    {
      "epoch": 13.615384615384615,
      "grad_norm": 1.8332946300506592,
      "learning_rate": 1.8461538461538465e-06,
      "loss": 0.8995,
      "step": 1770
    },
    {
      "epoch": 13.692307692307692,
      "grad_norm": 2.2418551445007324,
      "learning_rate": 1.7435897435897436e-06,
      "loss": 0.8818,
      "step": 1780
    },
    {
      "epoch": 13.76923076923077,
      "grad_norm": 1.794832706451416,
      "learning_rate": 1.6410256410256412e-06,
      "loss": 0.9044,
      "step": 1790
    },
    {
      "epoch": 13.846153846153847,
      "grad_norm": 3.0142152309417725,
      "learning_rate": 1.5384615384615387e-06,
      "loss": 0.8826,
      "step": 1800
    },
    {
      "epoch": 13.923076923076923,
      "grad_norm": 2.5891315937042236,
      "learning_rate": 1.4358974358974359e-06,
      "loss": 0.8387,
      "step": 1810
    },
    {
      "epoch": 14.0,
      "grad_norm": 5.37412691116333,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 0.9001,
      "step": 1820
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.7819548872180451,
      "eval_loss": 0.8673797845840454,
      "eval_runtime": 0.7642,
      "eval_samples_per_second": 174.027,
      "eval_steps_per_second": 22.244,
      "step": 1820
    },
    {
      "epoch": 14.076923076923077,
      "grad_norm": 1.8213236331939697,
      "learning_rate": 1.230769230769231e-06,
      "loss": 0.9047,
      "step": 1830
    },
    {
      "epoch": 14.153846153846153,
      "grad_norm": 1.8006333112716675,
      "learning_rate": 1.1282051282051283e-06,
      "loss": 0.8768,
      "step": 1840
    },
    {
      "epoch": 14.23076923076923,
      "grad_norm": 2.691574811935425,
      "learning_rate": 1.0256410256410257e-06,
      "loss": 0.8757,
      "step": 1850
    },
    {
      "epoch": 14.307692307692308,
      "grad_norm": 5.015848636627197,
      "learning_rate": 9.230769230769232e-07,
      "loss": 0.8734,
      "step": 1860
    },
    {
      "epoch": 14.384615384615385,
      "grad_norm": 2.5233821868896484,
      "learning_rate": 8.205128205128206e-07,
      "loss": 0.8787,
      "step": 1870
    },
    {
      "epoch": 14.461538461538462,
      "grad_norm": 2.1718924045562744,
      "learning_rate": 7.179487179487179e-07,
      "loss": 0.8767,
      "step": 1880
    },
    {
      "epoch": 14.538461538461538,
      "grad_norm": 3.0364015102386475,
      "learning_rate": 6.153846153846155e-07,
      "loss": 0.873,
      "step": 1890
    },
    {
      "epoch": 14.615384615384615,
      "grad_norm": 2.5152034759521484,
      "learning_rate": 5.128205128205128e-07,
      "loss": 0.9096,
      "step": 1900
    },
    {
      "epoch": 14.692307692307692,
      "grad_norm": 1.819096565246582,
      "learning_rate": 4.102564102564103e-07,
      "loss": 0.892,
      "step": 1910
    },
    {
      "epoch": 14.76923076923077,
      "grad_norm": 3.512732982635498,
      "learning_rate": 3.0769230769230774e-07,
      "loss": 0.8937,
      "step": 1920
    },
    {
      "epoch": 14.846153846153847,
      "grad_norm": 2.917677879333496,
      "learning_rate": 2.0512820512820514e-07,
      "loss": 0.9231,
      "step": 1930
    },
    {
      "epoch": 14.923076923076923,
      "grad_norm": 2.0683395862579346,
      "learning_rate": 1.0256410256410257e-07,
      "loss": 0.8613,
      "step": 1940
    },
    {
      "epoch": 15.0,
      "grad_norm": 4.704519271850586,
      "learning_rate": 0.0,
      "loss": 0.8842,
      "step": 1950
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.7894736842105263,
      "eval_loss": 0.8550169467926025,
      "eval_runtime": 0.7718,
      "eval_samples_per_second": 172.329,
      "eval_steps_per_second": 22.027,
      "step": 1950
    },
    {
      "epoch": 15.0,
      "step": 1950,
      "total_flos": 1.5658365504595968e+17,
      "train_loss": 0.6299933981284117,
      "train_runtime": 92.772,
      "train_samples_per_second": 167.184,
      "train_steps_per_second": 21.019
    }
  ],
  "logging_steps": 10,
  "max_steps": 1950,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.5658365504595968e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}