{
  "best_metric": 0.71900826446281,
  "best_model_checkpoint": "models/single_label/deberta-v3-large/exp5/checkpoint-1500",
  "epoch": 0.07913166189678593,
  "global_step": 1500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.1645569620253166e-08,
      "loss": 1.4101,
      "step": 100
    },
    {
      "epoch": 0.01,
      "learning_rate": 6.329113924050633e-08,
      "loss": 1.4037,
      "step": 200
    },
    {
      "epoch": 0.01,
      "eval_osu_accuracy": 0.5289256198347108,
      "eval_osu_loss": 1.3296300172805786,
      "eval_osu_matthews_correlation": 0.08430984949814109,
      "eval_osu_runtime": 4.6863,
      "eval_osu_samples_per_second": 51.64,
      "eval_osu_steps_per_second": 3.414,
      "step": 250
    },
    {
      "epoch": 0.01,
      "eval_deepset_1_accuracy": 0.42016806722689076,
      "eval_deepset_1_loss": 1.3601003885269165,
      "eval_deepset_1_matthews_correlation": 0.007340018855649166,
      "eval_deepset_1_runtime": 1.0598,
      "eval_deepset_1_samples_per_second": 112.29,
      "eval_deepset_1_steps_per_second": 7.549,
      "step": 250
    },
    {
      "epoch": 0.01,
      "eval_deepset_2_accuracy": 0.5128205128205128,
      "eval_deepset_2_loss": 1.341286540031433,
      "eval_deepset_2_matthews_correlation": 0.09702062677085851,
      "eval_deepset_2_runtime": 1.5595,
      "eval_deepset_2_samples_per_second": 100.032,
      "eval_deepset_2_steps_per_second": 6.412,
      "step": 250
    },
    {
      "epoch": 0.01,
      "eval_deepset_3_accuracy": 0.5260115606936416,
      "eval_deepset_3_loss": 1.3386189937591553,
      "eval_deepset_3_matthews_correlation": 0.13147932110355937,
      "eval_deepset_3_runtime": 3.0465,
      "eval_deepset_3_samples_per_second": 113.572,
      "eval_deepset_3_steps_per_second": 7.221,
      "step": 250
    },
    {
      "epoch": 0.01,
      "eval_deepset_4_accuracy": 0.46601941747572817,
      "eval_deepset_4_loss": 1.354163646697998,
      "eval_deepset_4_matthews_correlation": 0.04602315745107765,
      "eval_deepset_4_runtime": 1.7129,
      "eval_deepset_4_samples_per_second": 120.262,
      "eval_deepset_4_steps_per_second": 7.589,
      "step": 250
    },
    {
      "epoch": 0.01,
      "eval_anli_accuracy": 0.314,
      "eval_anli_loss": 1.3443490266799927,
      "eval_anli_matthews_correlation": -0.03669996966921473,
      "eval_anli_runtime": 5.1544,
      "eval_anli_samples_per_second": 194.01,
      "eval_anli_steps_per_second": 12.223,
      "step": 250
    },
    {
      "epoch": 0.02,
      "learning_rate": 9.462025316455696e-08,
      "loss": 1.381,
      "step": 300
    },
    {
      "epoch": 0.02,
      "learning_rate": 1.2626582278481014e-07,
      "loss": 1.3016,
      "step": 400
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.5791139240506326e-07,
      "loss": 1.1103,
      "step": 500
    },
    {
      "epoch": 0.03,
      "eval_osu_accuracy": 0.3677685950413223,
      "eval_osu_loss": 1.2251540422439575,
      "eval_osu_matthews_correlation": 0.009389673632694094,
      "eval_osu_runtime": 4.6743,
      "eval_osu_samples_per_second": 51.773,
      "eval_osu_steps_per_second": 3.423,
      "step": 500
    },
    {
      "epoch": 0.03,
      "eval_deepset_1_accuracy": 0.2605042016806723,
      "eval_deepset_1_loss": 1.498146414756775,
      "eval_deepset_1_matthews_correlation": -0.12070085589088507,
      "eval_deepset_1_runtime": 1.0684,
      "eval_deepset_1_samples_per_second": 111.378,
      "eval_deepset_1_steps_per_second": 7.488,
      "step": 500
    },
    {
      "epoch": 0.03,
      "eval_deepset_2_accuracy": 0.3076923076923077,
      "eval_deepset_2_loss": 1.3761013746261597,
      "eval_deepset_2_matthews_correlation": -0.08397491442025497,
      "eval_deepset_2_runtime": 1.5607,
      "eval_deepset_2_samples_per_second": 99.953,
      "eval_deepset_2_steps_per_second": 6.407,
      "step": 500
    },
    {
      "epoch": 0.03,
      "eval_deepset_3_accuracy": 0.31213872832369943,
      "eval_deepset_3_loss": 1.4205697774887085,
      "eval_deepset_3_matthews_correlation": -0.06528708661392457,
      "eval_deepset_3_runtime": 3.0447,
      "eval_deepset_3_samples_per_second": 113.639,
      "eval_deepset_3_steps_per_second": 7.226,
      "step": 500
    },
    {
      "epoch": 0.03,
      "eval_deepset_4_accuracy": 0.25728155339805825,
      "eval_deepset_4_loss": 1.5763508081436157,
      "eval_deepset_4_matthews_correlation": -0.11908261162663966,
      "eval_deepset_4_runtime": 1.7183,
      "eval_deepset_4_samples_per_second": 119.883,
      "eval_deepset_4_steps_per_second": 7.565,
      "step": 500
    },
    {
      "epoch": 0.03,
      "eval_anli_accuracy": 0.33,
      "eval_anli_loss": 1.1667168140411377,
      "eval_anli_matthews_correlation": 0.0012307081400291311,
      "eval_anli_runtime": 5.1504,
      "eval_anli_samples_per_second": 194.161,
      "eval_anli_steps_per_second": 12.232,
      "step": 500
    },
    {
      "epoch": 0.03,
      "learning_rate": 1.8955696202531644e-07,
      "loss": 0.8892,
      "step": 600
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.2120253164556962e-07,
      "loss": 0.7838,
      "step": 700
    },
    {
      "epoch": 0.04,
      "eval_osu_accuracy": 0.6942148760330579,
      "eval_osu_loss": 0.9712508916854858,
      "eval_osu_matthews_correlation": 0.4897642238477718,
      "eval_osu_runtime": 4.6823,
      "eval_osu_samples_per_second": 51.684,
      "eval_osu_steps_per_second": 3.417,
      "step": 750
    },
    {
      "epoch": 0.04,
      "eval_deepset_1_accuracy": 0.6134453781512605,
      "eval_deepset_1_loss": 1.6580817699432373,
      "eval_deepset_1_matthews_correlation": 0.41985911107813456,
      "eval_deepset_1_runtime": 1.0652,
      "eval_deepset_1_samples_per_second": 111.721,
      "eval_deepset_1_steps_per_second": 7.511,
      "step": 750
    },
    {
      "epoch": 0.04,
      "eval_deepset_2_accuracy": 0.7051282051282052,
      "eval_deepset_2_loss": 1.3619518280029297,
      "eval_deepset_2_matthews_correlation": 0.5328568142577198,
      "eval_deepset_2_runtime": 1.5605,
      "eval_deepset_2_samples_per_second": 99.966,
      "eval_deepset_2_steps_per_second": 6.408,
      "step": 750
    },
    {
      "epoch": 0.04,
      "eval_deepset_3_accuracy": 0.5578034682080925,
      "eval_deepset_3_loss": 1.4653511047363281,
      "eval_deepset_3_matthews_correlation": 0.2932617859503659,
      "eval_deepset_3_runtime": 3.0504,
      "eval_deepset_3_samples_per_second": 113.429,
      "eval_deepset_3_steps_per_second": 7.212,
      "step": 750
    },
    {
      "epoch": 0.04,
      "eval_deepset_4_accuracy": 0.529126213592233,
      "eval_deepset_4_loss": 1.7414307594299316,
      "eval_deepset_4_matthews_correlation": 0.3054166536717615,
      "eval_deepset_4_runtime": 1.7235,
      "eval_deepset_4_samples_per_second": 119.522,
      "eval_deepset_4_steps_per_second": 7.543,
      "step": 750
    },
    {
      "epoch": 0.04,
      "eval_anli_accuracy": 0.351,
      "eval_anli_loss": 1.1338268518447876,
      "eval_anli_matthews_correlation": 0.030305510765310642,
      "eval_anli_runtime": 5.1432,
      "eval_anli_samples_per_second": 194.432,
      "eval_anli_steps_per_second": 12.249,
      "step": 750
    },
    {
      "epoch": 0.04,
      "learning_rate": 2.5284810126582275e-07,
      "loss": 0.6807,
      "step": 800
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.844936708860759e-07,
      "loss": 0.502,
      "step": 900
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.991503304270561e-07,
      "loss": 0.4006,
      "step": 1000
    },
    {
      "epoch": 0.05,
      "eval_osu_accuracy": 0.7024793388429752,
      "eval_osu_loss": 0.7255080938339233,
      "eval_osu_matthews_correlation": 0.4761839486992106,
      "eval_osu_runtime": 4.6909,
      "eval_osu_samples_per_second": 51.589,
      "eval_osu_steps_per_second": 3.411,
      "step": 1000
    },
    {
      "epoch": 0.05,
      "eval_deepset_1_accuracy": 0.6134453781512605,
      "eval_deepset_1_loss": 1.6959413290023804,
      "eval_deepset_1_matthews_correlation": 0.42105052198193255,
      "eval_deepset_1_runtime": 1.0672,
      "eval_deepset_1_samples_per_second": 111.505,
      "eval_deepset_1_steps_per_second": 7.496,
      "step": 1000
    },
    {
      "epoch": 0.05,
      "eval_deepset_2_accuracy": 0.7243589743589743,
      "eval_deepset_2_loss": 1.2153555154800415,
      "eval_deepset_2_matthews_correlation": 0.5466918371684817,
      "eval_deepset_2_runtime": 1.556,
      "eval_deepset_2_samples_per_second": 100.254,
      "eval_deepset_2_steps_per_second": 6.427,
      "step": 1000
    },
    {
      "epoch": 0.05,
      "eval_deepset_3_accuracy": 0.6473988439306358,
      "eval_deepset_3_loss": 1.4247812032699585,
      "eval_deepset_3_matthews_correlation": 0.4017768477834457,
      "eval_deepset_3_runtime": 3.045,
      "eval_deepset_3_samples_per_second": 113.628,
      "eval_deepset_3_steps_per_second": 7.225,
      "step": 1000
    },
    {
      "epoch": 0.05,
      "eval_deepset_4_accuracy": 0.6504854368932039,
      "eval_deepset_4_loss": 1.7334672212600708,
      "eval_deepset_4_matthews_correlation": 0.4508615553005039,
      "eval_deepset_4_runtime": 1.7091,
      "eval_deepset_4_samples_per_second": 120.534,
      "eval_deepset_4_steps_per_second": 7.606,
      "step": 1000
    },
    {
      "epoch": 0.05,
      "eval_anli_accuracy": 0.477,
      "eval_anli_loss": 1.3187198638916016,
      "eval_anli_matthews_correlation": 0.218035345395415,
      "eval_anli_runtime": 5.1567,
      "eval_anli_samples_per_second": 193.922,
      "eval_anli_steps_per_second": 12.217,
      "step": 1000
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.9748431165657797e-07,
      "loss": 0.3456,
      "step": 1100
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.9581829288609986e-07,
      "loss": 0.3098,
      "step": 1200
    },
    {
      "epoch": 0.07,
      "eval_osu_accuracy": 0.6900826446280992,
      "eval_osu_loss": 0.794772207736969,
      "eval_osu_matthews_correlation": 0.4632054751499522,
      "eval_osu_runtime": 4.6759,
      "eval_osu_samples_per_second": 51.755,
      "eval_osu_steps_per_second": 3.422,
      "step": 1250
    },
    {
      "epoch": 0.07,
      "eval_deepset_1_accuracy": 0.5882352941176471,
      "eval_deepset_1_loss": 1.9199906587600708,
      "eval_deepset_1_matthews_correlation": 0.3787987593817238,
      "eval_deepset_1_runtime": 1.0581,
      "eval_deepset_1_samples_per_second": 112.466,
      "eval_deepset_1_steps_per_second": 7.561,
      "step": 1250
    },
    {
      "epoch": 0.07,
      "eval_deepset_2_accuracy": 0.6923076923076923,
      "eval_deepset_2_loss": 1.3611226081848145,
      "eval_deepset_2_matthews_correlation": 0.5007307553834116,
      "eval_deepset_2_runtime": 1.5603,
      "eval_deepset_2_samples_per_second": 99.978,
      "eval_deepset_2_steps_per_second": 6.409,
      "step": 1250
    },
    {
      "epoch": 0.07,
      "eval_deepset_3_accuracy": 0.630057803468208,
      "eval_deepset_3_loss": 1.5691180229187012,
      "eval_deepset_3_matthews_correlation": 0.38258183358832687,
      "eval_deepset_3_runtime": 3.052,
      "eval_deepset_3_samples_per_second": 113.37,
      "eval_deepset_3_steps_per_second": 7.208,
      "step": 1250
    },
    {
      "epoch": 0.07,
      "eval_deepset_4_accuracy": 0.5922330097087378,
      "eval_deepset_4_loss": 1.946365237236023,
      "eval_deepset_4_matthews_correlation": 0.36618144302865463,
      "eval_deepset_4_runtime": 1.7085,
      "eval_deepset_4_samples_per_second": 120.574,
      "eval_deepset_4_steps_per_second": 7.609,
      "step": 1250
    },
    {
      "epoch": 0.07,
      "eval_anli_accuracy": 0.577,
      "eval_anli_loss": 1.1495600938796997,
      "eval_anli_matthews_correlation": 0.36758926218287485,
      "eval_anli_runtime": 5.1325,
      "eval_anli_samples_per_second": 194.837,
      "eval_anli_steps_per_second": 12.275,
      "step": 1250
    },
    {
      "epoch": 0.07,
      "learning_rate": 2.941522741156217e-07,
      "loss": 0.2936,
      "step": 1300
    },
    {
      "epoch": 0.07,
      "learning_rate": 2.9248625534514355e-07,
      "loss": 0.275,
      "step": 1400
    },
    {
      "epoch": 0.08,
      "learning_rate": 2.9083689676237014e-07,
      "loss": 0.2711,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "eval_osu_accuracy": 0.71900826446281,
      "eval_osu_loss": 0.8360257744789124,
      "eval_osu_matthews_correlation": 0.5055263088346037,
      "eval_osu_runtime": 4.6825,
      "eval_osu_samples_per_second": 51.682,
      "eval_osu_steps_per_second": 3.417,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "eval_deepset_1_accuracy": 0.6302521008403361,
      "eval_deepset_1_loss": 2.0135045051574707,
      "eval_deepset_1_matthews_correlation": 0.43404618683091417,
      "eval_deepset_1_runtime": 1.0652,
      "eval_deepset_1_samples_per_second": 111.718,
      "eval_deepset_1_steps_per_second": 7.51,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "eval_deepset_2_accuracy": 0.7115384615384616,
      "eval_deepset_2_loss": 1.3997838497161865,
      "eval_deepset_2_matthews_correlation": 0.5105532533477752,
      "eval_deepset_2_runtime": 1.5611,
      "eval_deepset_2_samples_per_second": 99.928,
      "eval_deepset_2_steps_per_second": 6.406,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "eval_deepset_3_accuracy": 0.6676300578034682,
      "eval_deepset_3_loss": 1.5432208776474,
      "eval_deepset_3_matthews_correlation": 0.42876096695199906,
      "eval_deepset_3_runtime": 3.0533,
      "eval_deepset_3_samples_per_second": 113.322,
      "eval_deepset_3_steps_per_second": 7.205,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "eval_deepset_4_accuracy": 0.6359223300970874,
      "eval_deepset_4_loss": 1.9390015602111816,
      "eval_deepset_4_matthews_correlation": 0.4108775577859797,
      "eval_deepset_4_runtime": 1.7173,
      "eval_deepset_4_samples_per_second": 119.953,
      "eval_deepset_4_steps_per_second": 7.57,
      "step": 1500
    },
    {
      "epoch": 0.08,
      "eval_anli_accuracy": 0.609,
      "eval_anli_loss": 1.042230486869812,
      "eval_anli_matthews_correlation": 0.42100205280404024,
      "eval_anli_runtime": 5.1501,
      "eval_anli_samples_per_second": 194.17,
      "eval_anli_steps_per_second": 12.233,
      "step": 1500
    }
  ],
  "max_steps": 18955,
  "num_train_epochs": 1,
  "total_flos": 2.042867740823347e+16,
  "trial_name": null,
  "trial_params": null
}