{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 49.01960784313726,
  "global_step": 12500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 4.9e-05,
      "loss": 2.0116,
      "step": 255
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4553,
      "eval_f1_macro": 0.1796,
      "eval_gen_len": 2.3716,
      "eval_loss": 0.7925581336021423,
      "eval_precision": 0.2144,
      "eval_recall": 0.2023,
      "eval_runtime": 18.41,
      "eval_samples_per_second": 264.584,
      "step": 255
    },
    {
      "epoch": 2.0,
      "learning_rate": 4.8e-05,
      "loss": 0.779,
      "step": 510
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4911,
      "eval_f1_macro": 0.2749,
      "eval_gen_len": 2.2535,
      "eval_loss": 0.752581775188446,
      "eval_precision": 0.4243,
      "eval_recall": 0.2719,
      "eval_runtime": 18.4504,
      "eval_samples_per_second": 264.005,
      "step": 510
    },
    {
      "epoch": 3.0,
      "learning_rate": 4.7e-05,
      "loss": 0.6777,
      "step": 765
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.503,
      "eval_f1_macro": 0.2851,
      "eval_gen_len": 2.3207,
      "eval_loss": 0.7490188479423523,
      "eval_precision": 0.3735,
      "eval_recall": 0.288,
      "eval_runtime": 18.569,
      "eval_samples_per_second": 262.318,
      "step": 765
    },
    {
      "epoch": 4.0,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.5968,
      "step": 1020
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5013,
      "eval_f1_macro": 0.2994,
      "eval_gen_len": 2.2948,
      "eval_loss": 0.7779901027679443,
      "eval_precision": 0.3833,
      "eval_recall": 0.2971,
      "eval_runtime": 18.5932,
      "eval_samples_per_second": 261.978,
      "step": 1020
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.5e-05,
      "loss": 0.5367,
      "step": 1275
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5044,
      "eval_f1_macro": 0.3052,
      "eval_gen_len": 2.3751,
      "eval_loss": 0.8049420118331909,
      "eval_precision": 0.3699,
      "eval_recall": 0.3016,
      "eval_runtime": 18.6267,
      "eval_samples_per_second": 261.506,
      "step": 1275
    },
    {
      "epoch": 6.0,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.475,
      "step": 1530
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5005,
      "eval_f1_macro": 0.3111,
      "eval_gen_len": 2.3361,
      "eval_loss": 0.8648290038108826,
      "eval_precision": 0.3545,
      "eval_recall": 0.3096,
      "eval_runtime": 18.6251,
      "eval_samples_per_second": 261.529,
      "step": 1530
    },
    {
      "epoch": 7.0,
      "learning_rate": 4.3e-05,
      "loss": 0.4194,
      "step": 1785
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4868,
      "eval_f1_macro": 0.2957,
      "eval_gen_len": 2.3318,
      "eval_loss": 0.9633024334907532,
      "eval_precision": 0.3316,
      "eval_recall": 0.2895,
      "eval_runtime": 18.3899,
      "eval_samples_per_second": 264.873,
      "step": 1785
    },
    {
      "epoch": 8.0,
      "learning_rate": 4.2e-05,
      "loss": 0.3619,
      "step": 2040
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.4925,
      "eval_f1_macro": 0.2992,
      "eval_gen_len": 2.3545,
      "eval_loss": 0.9422969818115234,
      "eval_precision": 0.3316,
      "eval_recall": 0.2964,
      "eval_runtime": 18.6364,
      "eval_samples_per_second": 261.37,
      "step": 2040
    },
    {
      "epoch": 9.0,
      "learning_rate": 4.1e-05,
      "loss": 0.3126,
      "step": 2295
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.4718,
      "eval_f1_macro": 0.2899,
      "eval_gen_len": 2.4246,
      "eval_loss": 1.086911678314209,
      "eval_precision": 0.3039,
      "eval_recall": 0.2956,
      "eval_runtime": 18.6675,
      "eval_samples_per_second": 260.935,
      "step": 2295
    },
    {
      "epoch": 10.0,
      "learning_rate": 4e-05,
      "loss": 0.2714,
      "step": 2550
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.4757,
      "eval_f1_macro": 0.2959,
      "eval_gen_len": 2.3593,
      "eval_loss": 1.1425822973251343,
      "eval_precision": 0.3162,
      "eval_recall": 0.2919,
      "eval_runtime": 18.4062,
      "eval_samples_per_second": 264.64,
      "step": 2550
    },
    {
      "epoch": 11.0,
      "learning_rate": 3.9000000000000006e-05,
      "loss": 0.2295,
      "step": 2805
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.4691,
      "eval_f1_macro": 0.289,
      "eval_gen_len": 2.4087,
      "eval_loss": 1.189605951309204,
      "eval_precision": 0.304,
      "eval_recall": 0.2901,
      "eval_runtime": 18.6533,
      "eval_samples_per_second": 261.133,
      "step": 2805
    },
    {
      "epoch": 12.0,
      "learning_rate": 3.8e-05,
      "loss": 0.1945,
      "step": 3060
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.4703,
      "eval_f1_macro": 0.3098,
      "eval_gen_len": 2.38,
      "eval_loss": 1.2930792570114136,
      "eval_precision": 0.3263,
      "eval_recall": 0.3061,
      "eval_runtime": 18.3552,
      "eval_samples_per_second": 265.375,
      "step": 3060
    },
    {
      "epoch": 13.0,
      "learning_rate": 3.7e-05,
      "loss": 0.1622,
      "step": 3315
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.4648,
      "eval_f1_macro": 0.3084,
      "eval_gen_len": 2.4237,
      "eval_loss": 1.3369712829589844,
      "eval_precision": 0.3194,
      "eval_recall": 0.3082,
      "eval_runtime": 18.3904,
      "eval_samples_per_second": 264.867,
      "step": 3315
    },
    {
      "epoch": 14.0,
      "learning_rate": 3.6e-05,
      "loss": 0.1339,
      "step": 3570
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.4757,
      "eval_f1_macro": 0.3098,
      "eval_gen_len": 2.3874,
      "eval_loss": 1.5157912969589233,
      "eval_precision": 0.3258,
      "eval_recall": 0.3087,
      "eval_runtime": 18.3743,
      "eval_samples_per_second": 265.099,
      "step": 3570
    },
    {
      "epoch": 15.0,
      "learning_rate": 3.5e-05,
      "loss": 0.1195,
      "step": 3825
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.4683,
      "eval_f1_macro": 0.3044,
      "eval_gen_len": 2.403,
      "eval_loss": 1.5008689165115356,
      "eval_precision": 0.3135,
      "eval_recall": 0.3059,
      "eval_runtime": 18.3877,
      "eval_samples_per_second": 264.905,
      "step": 3825
    },
    {
      "epoch": 16.0,
      "learning_rate": 3.4000000000000007e-05,
      "loss": 0.1019,
      "step": 4080
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.4738,
      "eval_f1_macro": 0.3054,
      "eval_gen_len": 2.3734,
      "eval_loss": 1.5503424406051636,
      "eval_precision": 0.3235,
      "eval_recall": 0.3035,
      "eval_runtime": 18.347,
      "eval_samples_per_second": 265.493,
      "step": 4080
    },
    {
      "epoch": 17.0,
      "learning_rate": 3.3e-05,
      "loss": 0.0853,
      "step": 4335
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.4759,
      "eval_f1_macro": 0.305,
      "eval_gen_len": 2.3954,
      "eval_loss": 1.7289695739746094,
      "eval_precision": 0.3219,
      "eval_recall": 0.3014,
      "eval_runtime": 18.3899,
      "eval_samples_per_second": 264.873,
      "step": 4335
    },
    {
      "epoch": 18.0,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 0.0773,
      "step": 4590
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.473,
      "eval_f1_macro": 0.3045,
      "eval_gen_len": 2.4233,
      "eval_loss": 1.7796562910079956,
      "eval_precision": 0.321,
      "eval_recall": 0.2996,
      "eval_runtime": 18.3356,
      "eval_samples_per_second": 265.658,
      "step": 4590
    },
    {
      "epoch": 19.0,
      "learning_rate": 3.1e-05,
      "loss": 0.0681,
      "step": 4845
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.4638,
      "eval_f1_macro": 0.2984,
      "eval_gen_len": 2.3843,
      "eval_loss": 1.753821611404419,
      "eval_precision": 0.3106,
      "eval_recall": 0.2996,
      "eval_runtime": 18.3825,
      "eval_samples_per_second": 264.98,
      "step": 4845
    },
    {
      "epoch": 20.0,
      "learning_rate": 3e-05,
      "loss": 0.0617,
      "step": 5100
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.4638,
      "eval_f1_macro": 0.3047,
      "eval_gen_len": 2.4092,
      "eval_loss": 1.868014931678772,
      "eval_precision": 0.3154,
      "eval_recall": 0.3036,
      "eval_runtime": 18.3921,
      "eval_samples_per_second": 264.841,
      "step": 5100
    },
    {
      "epoch": 21.0,
      "learning_rate": 2.9e-05,
      "loss": 0.0537,
      "step": 5355
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.4642,
      "eval_f1_macro": 0.3035,
      "eval_gen_len": 2.3738,
      "eval_loss": 1.9632675647735596,
      "eval_precision": 0.3201,
      "eval_recall": 0.2996,
      "eval_runtime": 18.4289,
      "eval_samples_per_second": 264.314,
      "step": 5355
    },
    {
      "epoch": 22.0,
      "learning_rate": 2.8000000000000003e-05,
      "loss": 0.0473,
      "step": 5610
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.4726,
      "eval_f1_macro": 0.303,
      "eval_gen_len": 2.3862,
      "eval_loss": 1.8952040672302246,
      "eval_precision": 0.3214,
      "eval_recall": 0.2998,
      "eval_runtime": 18.3834,
      "eval_samples_per_second": 264.967,
      "step": 5610
    },
    {
      "epoch": 23.0,
      "learning_rate": 2.7000000000000002e-05,
      "loss": 0.0465,
      "step": 5865
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.466,
      "eval_f1_macro": 0.3035,
      "eval_gen_len": 2.4024,
      "eval_loss": 1.9048091173171997,
      "eval_precision": 0.3173,
      "eval_recall": 0.3012,
      "eval_runtime": 18.3842,
      "eval_samples_per_second": 264.956,
      "step": 5865
    },
    {
      "epoch": 24.0,
      "learning_rate": 2.6000000000000002e-05,
      "loss": 0.0406,
      "step": 6120
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.4634,
      "eval_f1_macro": 0.3068,
      "eval_gen_len": 2.426,
      "eval_loss": 2.021580696105957,
      "eval_precision": 0.3153,
      "eval_recall": 0.3044,
      "eval_runtime": 18.3498,
      "eval_samples_per_second": 265.452,
      "step": 6120
    },
    {
      "epoch": 25.0,
      "learning_rate": 2.5e-05,
      "loss": 0.0358,
      "step": 6375
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.4742,
      "eval_f1_macro": 0.3003,
      "eval_gen_len": 2.3597,
      "eval_loss": 2.116412401199341,
      "eval_precision": 0.3236,
      "eval_recall": 0.2931,
      "eval_runtime": 18.3974,
      "eval_samples_per_second": 264.766,
      "step": 6375
    },
    {
      "epoch": 26.0,
      "learning_rate": 2.4e-05,
      "loss": 0.0353,
      "step": 6630
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.4668,
      "eval_f1_macro": 0.3004,
      "eval_gen_len": 2.4484,
      "eval_loss": 2.0235698223114014,
      "eval_precision": 0.3084,
      "eval_recall": 0.2995,
      "eval_runtime": 18.3702,
      "eval_samples_per_second": 265.158,
      "step": 6630
    },
    {
      "epoch": 27.0,
      "learning_rate": 2.3000000000000003e-05,
      "loss": 0.0314,
      "step": 6885
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.464,
      "eval_f1_macro": 0.3013,
      "eval_gen_len": 2.4204,
      "eval_loss": 2.124769926071167,
      "eval_precision": 0.3066,
      "eval_recall": 0.3019,
      "eval_runtime": 18.4132,
      "eval_samples_per_second": 264.538,
      "step": 6885
    },
    {
      "epoch": 28.0,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 0.0296,
      "step": 7140
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.4722,
      "eval_f1_macro": 0.2997,
      "eval_gen_len": 2.3732,
      "eval_loss": 2.124000072479248,
      "eval_precision": 0.3261,
      "eval_recall": 0.294,
      "eval_runtime": 18.3663,
      "eval_samples_per_second": 265.213,
      "step": 7140
    },
    {
      "epoch": 29.0,
      "learning_rate": 2.1e-05,
      "loss": 0.0274,
      "step": 7395
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.467,
      "eval_f1_macro": 0.3011,
      "eval_gen_len": 2.3636,
      "eval_loss": 2.1549148559570312,
      "eval_precision": 0.3197,
      "eval_recall": 0.2963,
      "eval_runtime": 18.4004,
      "eval_samples_per_second": 264.722,
      "step": 7395
    },
    {
      "epoch": 30.0,
      "learning_rate": 2e-05,
      "loss": 0.0248,
      "step": 7650
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.4697,
      "eval_f1_macro": 0.2982,
      "eval_gen_len": 2.3798,
      "eval_loss": 2.2189269065856934,
      "eval_precision": 0.3152,
      "eval_recall": 0.2951,
      "eval_runtime": 18.3622,
      "eval_samples_per_second": 265.273,
      "step": 7650
    },
    {
      "epoch": 31.0,
      "learning_rate": 1.9e-05,
      "loss": 0.0219,
      "step": 7905
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.4736,
      "eval_f1_macro": 0.3048,
      "eval_gen_len": 2.3942,
      "eval_loss": 2.3595752716064453,
      "eval_precision": 0.3157,
      "eval_recall": 0.3049,
      "eval_runtime": 18.3521,
      "eval_samples_per_second": 265.419,
      "step": 7905
    },
    {
      "epoch": 32.0,
      "learning_rate": 1.8e-05,
      "loss": 0.0205,
      "step": 8160
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.4705,
      "eval_f1_macro": 0.3013,
      "eval_gen_len": 2.3909,
      "eval_loss": 2.4317517280578613,
      "eval_precision": 0.3151,
      "eval_recall": 0.3001,
      "eval_runtime": 18.3589,
      "eval_samples_per_second": 265.321,
      "step": 8160
    },
    {
      "epoch": 33.0,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 0.0189,
      "step": 8415
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.4767,
      "eval_f1_macro": 0.3084,
      "eval_gen_len": 2.3751,
      "eval_loss": 2.4803547859191895,
      "eval_precision": 0.3242,
      "eval_recall": 0.3045,
      "eval_runtime": 18.4166,
      "eval_samples_per_second": 264.489,
      "step": 8415
    },
    {
      "epoch": 34.0,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.0211,
      "step": 8670
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.4699,
      "eval_f1_macro": 0.304,
      "eval_gen_len": 2.4102,
      "eval_loss": 2.371544361114502,
      "eval_precision": 0.3223,
      "eval_recall": 0.2999,
      "eval_runtime": 18.4037,
      "eval_samples_per_second": 264.675,
      "step": 8670
    },
    {
      "epoch": 35.0,
      "learning_rate": 1.5e-05,
      "loss": 0.0156,
      "step": 8925
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.4753,
      "eval_f1_macro": 0.3014,
      "eval_gen_len": 2.395,
      "eval_loss": 2.55307674407959,
      "eval_precision": 0.3136,
      "eval_recall": 0.2973,
      "eval_runtime": 18.4323,
      "eval_samples_per_second": 264.264,
      "step": 8925
    },
    {
      "epoch": 36.0,
      "learning_rate": 1.4000000000000001e-05,
      "loss": 0.0179,
      "step": 9180
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.4829,
      "eval_f1_macro": 0.3055,
      "eval_gen_len": 2.3621,
      "eval_loss": 2.49765682220459,
      "eval_precision": 0.3308,
      "eval_recall": 0.2991,
      "eval_runtime": 18.4231,
      "eval_samples_per_second": 264.396,
      "step": 9180
    },
    {
      "epoch": 37.0,
      "learning_rate": 1.3000000000000001e-05,
      "loss": 0.0155,
      "step": 9435
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.4808,
      "eval_f1_macro": 0.3086,
      "eval_gen_len": 2.3775,
      "eval_loss": 2.4960439205169678,
      "eval_precision": 0.3245,
      "eval_recall": 0.3039,
      "eval_runtime": 18.4083,
      "eval_samples_per_second": 264.609,
      "step": 9435
    },
    {
      "epoch": 38.0,
      "learning_rate": 1.2e-05,
      "loss": 0.0154,
      "step": 9690
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.4652,
      "eval_f1_macro": 0.3047,
      "eval_gen_len": 2.3983,
      "eval_loss": 2.59112811088562,
      "eval_precision": 0.3196,
      "eval_recall": 0.2998,
      "eval_runtime": 18.406,
      "eval_samples_per_second": 264.642,
      "step": 9690
    },
    {
      "epoch": 39.0,
      "learning_rate": 1.1000000000000001e-05,
      "loss": 0.0144,
      "step": 9945
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.473,
      "eval_f1_macro": 0.3097,
      "eval_gen_len": 2.3634,
      "eval_loss": 2.646393299102783,
      "eval_precision": 0.3283,
      "eval_recall": 0.3046,
      "eval_runtime": 18.3793,
      "eval_samples_per_second": 265.027,
      "step": 9945
    },
    {
      "epoch": 40.0,
      "learning_rate": 1e-05,
      "loss": 0.0135,
      "step": 10200
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.4695,
      "eval_f1_macro": 0.3035,
      "eval_gen_len": 2.3812,
      "eval_loss": 2.711395740509033,
      "eval_precision": 0.3185,
      "eval_recall": 0.2989,
      "eval_runtime": 18.3802,
      "eval_samples_per_second": 265.013,
      "step": 10200
    },
    {
      "epoch": 41.0,
      "learning_rate": 9e-06,
      "loss": 0.0132,
      "step": 10455
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.4707,
      "eval_f1_macro": 0.307,
      "eval_gen_len": 2.4055,
      "eval_loss": 2.706991672515869,
      "eval_precision": 0.3218,
      "eval_recall": 0.3029,
      "eval_runtime": 18.4175,
      "eval_samples_per_second": 264.476,
      "step": 10455
    },
    {
      "epoch": 42.0,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0113,
      "step": 10710
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.4705,
      "eval_f1_macro": 0.3041,
      "eval_gen_len": 2.3833,
      "eval_loss": 2.7490220069885254,
      "eval_precision": 0.3226,
      "eval_recall": 0.3009,
      "eval_runtime": 18.3891,
      "eval_samples_per_second": 264.885,
      "step": 10710
    },
    {
      "epoch": 43.0,
      "learning_rate": 7.000000000000001e-06,
      "loss": 0.0104,
      "step": 10965
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.4707,
      "eval_f1_macro": 0.3094,
      "eval_gen_len": 2.3868,
      "eval_loss": 2.8594679832458496,
      "eval_precision": 0.3257,
      "eval_recall": 0.304,
      "eval_runtime": 18.4068,
      "eval_samples_per_second": 264.631,
      "step": 10965
    },
    {
      "epoch": 44.0,
      "learning_rate": 6e-06,
      "loss": 0.0125,
      "step": 11220
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.4705,
      "eval_f1_macro": 0.3068,
      "eval_gen_len": 2.3921,
      "eval_loss": 2.777812957763672,
      "eval_precision": 0.3207,
      "eval_recall": 0.3034,
      "eval_runtime": 18.3628,
      "eval_samples_per_second": 265.264,
      "step": 11220
    },
    {
      "epoch": 45.0,
      "learning_rate": 5e-06,
      "loss": 0.0094,
      "step": 11475
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.4753,
      "eval_f1_macro": 0.3102,
      "eval_gen_len": 2.4036,
      "eval_loss": 2.7739901542663574,
      "eval_precision": 0.3301,
      "eval_recall": 0.3028,
      "eval_runtime": 18.4413,
      "eval_samples_per_second": 264.135,
      "step": 11475
    },
    {
      "epoch": 46.0,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0094,
      "step": 11730
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.4755,
      "eval_f1_macro": 0.3122,
      "eval_gen_len": 2.3979,
      "eval_loss": 2.759046792984009,
      "eval_precision": 0.3284,
      "eval_recall": 0.3064,
      "eval_runtime": 18.4399,
      "eval_samples_per_second": 264.155,
      "step": 11730
    },
    {
      "epoch": 47.0,
      "learning_rate": 3e-06,
      "loss": 0.0085,
      "step": 11985
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.4759,
      "eval_f1_macro": 0.3092,
      "eval_gen_len": 2.3903,
      "eval_loss": 2.8234634399414062,
      "eval_precision": 0.3248,
      "eval_recall": 0.304,
      "eval_runtime": 18.4435,
      "eval_samples_per_second": 264.104,
      "step": 11985
    },
    {
      "epoch": 48.0,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0087,
      "step": 12240
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.474,
      "eval_f1_macro": 0.3097,
      "eval_gen_len": 2.3977,
      "eval_loss": 2.834430694580078,
      "eval_precision": 0.3247,
      "eval_recall": 0.3045,
      "eval_runtime": 18.4163,
      "eval_samples_per_second": 264.494,
      "step": 12240
    },
    {
      "epoch": 49.0,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.009,
      "step": 12495
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.4732,
      "eval_f1_macro": 0.3081,
      "eval_gen_len": 2.3868,
      "eval_loss": 2.841893196105957,
      "eval_precision": 0.3256,
      "eval_recall": 0.3018,
      "eval_runtime": 18.4481,
      "eval_samples_per_second": 264.037,
      "step": 12495
    }
  ],
  "max_steps": 12750,
  "num_train_epochs": 50,
  "total_flos": 1.14107158131029e+17,
  "trial_name": null,
  "trial_params": null
}