{
  "best_global_step": 6696,
  "best_metric": 0.515357605277523,
  "best_model_checkpoint": "outputs/restaurant/vinai_phobert-base/checkpoint-6696",
  "epoch": 72.0,
  "eval_steps": 500,
  "global_step": 6696,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.10379958152771,
      "learning_rate": 2.9992756711044086e-05,
      "loss": 0.7015,
      "step": 93
    },
    {
      "epoch": 1.0,
      "eval_f1_aspect": 0.5933321447673382,
      "eval_f1_macro_aspect": 0.16885755258126195,
      "eval_f1_sentiment": 0.16885755258126195,
      "eval_loss": 0.48745372891426086,
      "eval_runtime": 8.9473,
      "eval_samples_per_second": 144.177,
      "eval_steps_per_second": 4.582,
      "step": 93
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6488051414489746,
      "learning_rate": 2.9970718237156808e-05,
      "loss": 0.5199,
      "step": 186
    },
    {
      "epoch": 2.0,
      "eval_f1_aspect": 0.624978526026456,
      "eval_f1_macro_aspect": 0.18175644028103044,
      "eval_f1_sentiment": 0.18175644028103044,
      "eval_loss": 0.45015349984169006,
      "eval_runtime": 9.1016,
      "eval_samples_per_second": 141.734,
      "eval_steps_per_second": 4.505,
      "step": 186
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.872970700263977,
      "learning_rate": 2.993390547180617e-05,
      "loss": 0.4759,
      "step": 279
    },
    {
      "epoch": 3.0,
      "eval_f1_aspect": 0.7459913082571556,
      "eval_f1_macro_aspect": 0.2446883510331042,
      "eval_f1_sentiment": 0.2446883510331042,
      "eval_loss": 0.40983885526657104,
      "eval_runtime": 9.2065,
      "eval_samples_per_second": 140.118,
      "eval_steps_per_second": 4.453,
      "step": 279
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.8576162457466125,
      "learning_rate": 2.9882354744747117e-05,
      "loss": 0.4218,
      "step": 372
    },
    {
      "epoch": 4.0,
      "eval_f1_aspect": 0.7686936281640966,
      "eval_f1_macro_aspect": 0.28575861718815654,
      "eval_f1_sentiment": 0.28575861718815654,
      "eval_loss": 0.38282328844070435,
      "eval_runtime": 9.2965,
      "eval_samples_per_second": 138.762,
      "eval_steps_per_second": 4.41,
      "step": 372
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.2181860208511353,
      "learning_rate": 2.981611693032346e-05,
      "loss": 0.3771,
      "step": 465
    },
    {
      "epoch": 5.0,
      "eval_f1_aspect": 0.7861789793262975,
      "eval_f1_macro_aspect": 0.32269120714687083,
      "eval_f1_sentiment": 0.32269120714687083,
      "eval_loss": 0.36723560094833374,
      "eval_runtime": 9.356,
      "eval_samples_per_second": 137.88,
      "eval_steps_per_second": 4.382,
      "step": 465
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.1722445487976074,
      "learning_rate": 2.9735257397261053e-05,
      "loss": 0.3386,
      "step": 558
    },
    {
      "epoch": 6.0,
      "eval_f1_aspect": 0.7791606367583213,
      "eval_f1_macro_aspect": 0.3425302041227573,
      "eval_f1_sentiment": 0.3425302041227573,
      "eval_loss": 0.3524690866470337,
      "eval_runtime": 9.3401,
      "eval_samples_per_second": 138.114,
      "eval_steps_per_second": 4.39,
      "step": 558
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.1638844013214111,
      "learning_rate": 2.9639855944156734e-05,
      "loss": 0.3011,
      "step": 651
    },
    {
      "epoch": 7.0,
      "eval_f1_aspect": 0.7984066012235026,
      "eval_f1_macro_aspect": 0.34551040596236665,
      "eval_f1_sentiment": 0.34551040596236665,
      "eval_loss": 0.346333771944046,
      "eval_runtime": 9.3542,
      "eval_samples_per_second": 137.906,
      "eval_steps_per_second": 4.383,
      "step": 651
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.6177209615707397,
      "learning_rate": 2.9530006720726755e-05,
      "loss": 0.2675,
      "step": 744
    },
    {
      "epoch": 8.0,
      "eval_f1_aspect": 0.8155061019382628,
      "eval_f1_macro_aspect": 0.3937242116265103,
      "eval_f1_sentiment": 0.3937242116265103,
      "eval_loss": 0.32574591040611267,
      "eval_runtime": 9.3295,
      "eval_samples_per_second": 138.271,
      "eval_steps_per_second": 4.395,
      "step": 744
    },
    {
      "epoch": 9.0,
      "grad_norm": 1.4512020349502563,
      "learning_rate": 2.9405818134892394e-05,
      "loss": 0.2355,
      "step": 837
    },
    {
      "epoch": 9.0,
      "eval_f1_aspect": 0.817910447761194,
      "eval_f1_macro_aspect": 0.4096234810909085,
      "eval_f1_sentiment": 0.4096234810909085,
      "eval_loss": 0.33384305238723755,
      "eval_runtime": 9.3362,
      "eval_samples_per_second": 138.172,
      "eval_steps_per_second": 4.392,
      "step": 837
    },
    {
      "epoch": 10.0,
      "grad_norm": 2.0551767349243164,
      "learning_rate": 2.926741274579439e-05,
      "loss": 0.2072,
      "step": 930
    },
    {
      "epoch": 10.0,
      "eval_f1_aspect": 0.8134773955075348,
      "eval_f1_macro_aspect": 0.4091244468478088,
      "eval_f1_sentiment": 0.4091244468478088,
      "eval_loss": 0.3412579298019409,
      "eval_runtime": 9.3634,
      "eval_samples_per_second": 137.77,
      "eval_steps_per_second": 4.379,
      "step": 930
    },
    {
      "epoch": 11.0,
      "grad_norm": 2.016472578048706,
      "learning_rate": 2.9114927142841903e-05,
      "loss": 0.1839,
      "step": 1023
    },
    {
      "epoch": 11.0,
      "eval_f1_aspect": 0.8203693340794628,
      "eval_f1_macro_aspect": 0.43513403771495174,
      "eval_f1_sentiment": 0.43513403771495174,
      "eval_loss": 0.3348862826824188,
      "eval_runtime": 9.3321,
      "eval_samples_per_second": 138.232,
      "eval_steps_per_second": 4.393,
      "step": 1023
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.1789063215255737,
      "learning_rate": 2.894851181091524e-05,
      "loss": 0.1653,
      "step": 1116
    },
    {
      "epoch": 12.0,
      "eval_f1_aspect": 0.8179865387369325,
      "eval_f1_macro_aspect": 0.41853155653805585,
      "eval_f1_sentiment": 0.41853155653805585,
      "eval_loss": 0.339160680770874,
      "eval_runtime": 9.3338,
      "eval_samples_per_second": 138.207,
      "eval_steps_per_second": 4.393,
      "step": 1116
    },
    {
      "epoch": 13.0,
      "grad_norm": 1.1383475065231323,
      "learning_rate": 2.8768330981855445e-05,
      "loss": 0.1483,
      "step": 1209
    },
    {
      "epoch": 13.0,
      "eval_f1_aspect": 0.8207835509863851,
      "eval_f1_macro_aspect": 0.47172092276258015,
      "eval_f1_sentiment": 0.47172092276258015,
      "eval_loss": 0.35056257247924805,
      "eval_runtime": 9.3255,
      "eval_samples_per_second": 138.331,
      "eval_steps_per_second": 4.397,
      "step": 1209
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.7556346654891968,
      "learning_rate": 2.85745624723873e-05,
      "loss": 0.1336,
      "step": 1302
    },
    {
      "epoch": 14.0,
      "eval_f1_aspect": 0.8288747346072187,
      "eval_f1_macro_aspect": 0.45233775987096386,
      "eval_f1_sentiment": 0.45233775987096386,
      "eval_loss": 0.34405574202537537,
      "eval_runtime": 9.343,
      "eval_samples_per_second": 138.072,
      "eval_steps_per_second": 4.388,
      "step": 1302
    },
    {
      "epoch": 15.0,
      "grad_norm": 1.1931471824645996,
      "learning_rate": 2.8367397508635696e-05,
      "loss": 0.1215,
      "step": 1395
    },
    {
      "epoch": 15.0,
      "eval_f1_aspect": 0.8191606847045831,
      "eval_f1_macro_aspect": 0.46768270879244433,
      "eval_f1_sentiment": 0.46768270879244433,
      "eval_loss": 0.3684200644493103,
      "eval_runtime": 9.3313,
      "eval_samples_per_second": 138.245,
      "eval_steps_per_second": 4.394,
      "step": 1395
    },
    {
      "epoch": 16.0,
      "grad_norm": 1.4353535175323486,
      "learning_rate": 2.8147040537408527e-05,
      "loss": 0.1093,
      "step": 1488
    },
    {
      "epoch": 16.0,
      "eval_f1_aspect": 0.8141520943472956,
      "eval_f1_macro_aspect": 0.45828212910388655,
      "eval_f1_sentiment": 0.45828212910388655,
      "eval_loss": 0.3960510194301605,
      "eval_runtime": 9.3258,
      "eval_samples_per_second": 138.326,
      "eval_steps_per_second": 4.396,
      "step": 1488
    },
    {
      "epoch": 17.0,
      "grad_norm": 1.9182143211364746,
      "learning_rate": 2.791370902443238e-05,
      "loss": 0.0998,
      "step": 1581
    },
    {
      "epoch": 17.0,
      "eval_f1_aspect": 0.8258761601329824,
      "eval_f1_macro_aspect": 0.48051371313363417,
      "eval_f1_sentiment": 0.48051371313363417,
      "eval_loss": 0.3712829649448395,
      "eval_runtime": 9.3196,
      "eval_samples_per_second": 138.418,
      "eval_steps_per_second": 4.399,
      "step": 1581
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.9299781918525696,
      "learning_rate": 2.76676332397401e-05,
      "loss": 0.0914,
      "step": 1674
    },
    {
      "epoch": 18.0,
      "eval_f1_aspect": 0.819343316389614,
      "eval_f1_macro_aspect": 0.4569195105417062,
      "eval_f1_sentiment": 0.4569195105417062,
      "eval_loss": 0.38896748423576355,
      "eval_runtime": 9.3581,
      "eval_samples_per_second": 137.849,
      "eval_steps_per_second": 4.381,
      "step": 1674
    },
    {
      "epoch": 19.0,
      "grad_norm": 1.3669250011444092,
      "learning_rate": 2.740905603042209e-05,
      "loss": 0.0839,
      "step": 1767
    },
    {
      "epoch": 19.0,
      "eval_f1_aspect": 0.8201814682430575,
      "eval_f1_macro_aspect": 0.47456435634347893,
      "eval_f1_sentiment": 0.47456435634347893,
      "eval_loss": 0.37582284212112427,
      "eval_runtime": 9.37,
      "eval_samples_per_second": 137.673,
      "eval_steps_per_second": 4.376,
      "step": 1767
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.1103769540786743,
      "learning_rate": 2.713823258096555e-05,
      "loss": 0.0761,
      "step": 1860
    },
    {
      "epoch": 20.0,
      "eval_f1_aspect": 0.8232564487511942,
      "eval_f1_macro_aspect": 0.4851568517791377,
      "eval_f1_sentiment": 0.4851568517791377,
      "eval_loss": 0.39441487193107605,
      "eval_runtime": 9.367,
      "eval_samples_per_second": 137.718,
      "eval_steps_per_second": 4.377,
      "step": 1860
    },
    {
      "epoch": 21.0,
      "grad_norm": 1.348577618598938,
      "learning_rate": 2.685543016141817e-05,
      "loss": 0.0712,
      "step": 1953
    },
    {
      "epoch": 21.0,
      "eval_f1_aspect": 0.8264597795018375,
      "eval_f1_macro_aspect": 0.47069470951715775,
      "eval_f1_sentiment": 0.47069470951715775,
      "eval_loss": 0.40276068449020386,
      "eval_runtime": 9.3451,
      "eval_samples_per_second": 138.041,
      "eval_steps_per_second": 4.387,
      "step": 1953
    },
    {
      "epoch": 22.0,
      "grad_norm": 0.24034403264522552,
      "learning_rate": 2.65609278636249e-05,
      "loss": 0.065,
      "step": 2046
    },
    {
      "epoch": 22.0,
      "eval_f1_aspect": 0.8310792161192382,
      "eval_f1_macro_aspect": 0.47892394687508477,
      "eval_f1_sentiment": 0.47892394687508477,
      "eval_loss": 0.3948163688182831,
      "eval_runtime": 9.3847,
      "eval_samples_per_second": 137.458,
      "eval_steps_per_second": 4.369,
      "step": 2046
    },
    {
      "epoch": 23.0,
      "grad_norm": 0.6048269271850586,
      "learning_rate": 2.6255016325797975e-05,
      "loss": 0.0603,
      "step": 2139
    },
    {
      "epoch": 23.0,
      "eval_f1_aspect": 0.8304270218316627,
      "eval_f1_macro_aspect": 0.48533615854793977,
      "eval_f1_sentiment": 0.48533615854793977,
      "eval_loss": 0.39172783493995667,
      "eval_runtime": 9.3835,
      "eval_samples_per_second": 137.475,
      "eval_steps_per_second": 4.369,
      "step": 2139
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.7877061367034912,
      "learning_rate": 2.5937997445692067e-05,
      "loss": 0.0545,
      "step": 2232
    },
    {
      "epoch": 24.0,
      "eval_f1_aspect": 0.8241029113067028,
      "eval_f1_macro_aspect": 0.4876234699570111,
      "eval_f1_sentiment": 0.4876234699570111,
      "eval_loss": 0.4218457043170929,
      "eval_runtime": 9.3259,
      "eval_samples_per_second": 138.324,
      "eval_steps_per_second": 4.396,
      "step": 2232
    },
    {
      "epoch": 25.0,
      "grad_norm": 0.6037042140960693,
      "learning_rate": 2.56101840826677e-05,
      "loss": 0.05,
      "step": 2325
    },
    {
      "epoch": 25.0,
      "eval_f1_aspect": 0.8251955759374157,
      "eval_f1_macro_aspect": 0.49043349379612233,
      "eval_f1_sentiment": 0.49043349379612233,
      "eval_loss": 0.4337195158004761,
      "eval_runtime": 9.3447,
      "eval_samples_per_second": 138.046,
      "eval_steps_per_second": 4.388,
      "step": 2325
    },
    {
      "epoch": 26.0,
      "grad_norm": 0.35369500517845154,
      "learning_rate": 2.5271899748936786e-05,
      "loss": 0.0464,
      "step": 2418
    },
    {
      "epoch": 26.0,
      "eval_f1_aspect": 0.8145386766076421,
      "eval_f1_macro_aspect": 0.4989429955255985,
      "eval_f1_sentiment": 0.4989429955255985,
      "eval_loss": 0.4668835997581482,
      "eval_runtime": 9.3436,
      "eval_samples_per_second": 138.062,
      "eval_steps_per_second": 4.388,
      "step": 2418
    },
    {
      "epoch": 27.0,
      "grad_norm": 0.690770149230957,
      "learning_rate": 2.4923478290295158e-05,
      "loss": 0.0427,
      "step": 2511
    },
    {
      "epoch": 27.0,
      "eval_f1_aspect": 0.8269706124691019,
      "eval_f1_macro_aspect": 0.4950815490892515,
      "eval_f1_sentiment": 0.4950815490892515,
      "eval_loss": 0.4226386547088623,
      "eval_runtime": 9.3311,
      "eval_samples_per_second": 138.248,
      "eval_steps_per_second": 4.394,
      "step": 2511
    },
    {
      "epoch": 28.0,
      "grad_norm": 0.25553351640701294,
      "learning_rate": 2.4565263556657065e-05,
      "loss": 0.0393,
      "step": 2604
    },
    {
      "epoch": 28.0,
      "eval_f1_aspect": 0.8211371237458194,
      "eval_f1_macro_aspect": 0.49076482790770953,
      "eval_f1_sentiment": 0.49076482790770953,
      "eval_loss": 0.450793981552124,
      "eval_runtime": 9.3603,
      "eval_samples_per_second": 137.816,
      "eval_steps_per_second": 4.38,
      "step": 2604
    },
    {
      "epoch": 29.0,
      "grad_norm": 0.8166686296463013,
      "learning_rate": 2.419760906271681e-05,
      "loss": 0.0361,
      "step": 2697
    },
    {
      "epoch": 29.0,
      "eval_f1_aspect": 0.829605530703538,
      "eval_f1_macro_aspect": 0.497223760698809,
      "eval_f1_sentiment": 0.497223760698809,
      "eval_loss": 0.4382466971874237,
      "eval_runtime": 9.3474,
      "eval_samples_per_second": 138.006,
      "eval_steps_per_second": 4.386,
      "step": 2697
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.502255916595459,
      "learning_rate": 2.382087763907246e-05,
      "loss": 0.0337,
      "step": 2790
    },
    {
      "epoch": 30.0,
      "eval_f1_aspect": 0.8287307849272207,
      "eval_f1_macro_aspect": 0.4848470830656446,
      "eval_f1_sentiment": 0.4848470830656446,
      "eval_loss": 0.44884979724884033,
      "eval_runtime": 9.3411,
      "eval_samples_per_second": 138.099,
      "eval_steps_per_second": 4.389,
      "step": 2790
    },
    {
      "epoch": 31.0,
      "grad_norm": 0.9951657652854919,
      "learning_rate": 2.3435441074155805e-05,
      "loss": 0.0318,
      "step": 2883
    },
    {
      "epoch": 31.0,
      "eval_f1_aspect": 0.8262215641405304,
      "eval_f1_macro_aspect": 0.4980498748854937,
      "eval_f1_sentiment": 0.4980498748854937,
      "eval_loss": 0.4665895998477936,
      "eval_runtime": 9.3449,
      "eval_samples_per_second": 138.043,
      "eval_steps_per_second": 4.387,
      "step": 2883
    },
    {
      "epoch": 32.0,
      "grad_norm": 0.36832740902900696,
      "learning_rate": 2.30416797473221e-05,
      "loss": 0.0286,
      "step": 2976
    },
    {
      "epoch": 32.0,
      "eval_f1_aspect": 0.8308532423208191,
      "eval_f1_macro_aspect": 0.49750043202051564,
      "eval_f1_sentiment": 0.49750043202051564,
      "eval_loss": 0.4392968416213989,
      "eval_runtime": 9.3083,
      "eval_samples_per_second": 138.586,
      "eval_steps_per_second": 4.405,
      "step": 2976
    },
    {
      "epoch": 33.0,
      "grad_norm": 0.32439595460891724,
      "learning_rate": 2.2639982253461552e-05,
      "loss": 0.0268,
      "step": 3069
    },
    {
      "epoch": 33.0,
      "eval_f1_aspect": 0.8321267586395301,
      "eval_f1_macro_aspect": 0.4950515397265748,
      "eval_f1_sentiment": 0.4950515397265748,
      "eval_loss": 0.4425518214702606,
      "eval_runtime": 9.3231,
      "eval_samples_per_second": 138.365,
      "eval_steps_per_second": 4.398,
      "step": 3069
    },
    {
      "epoch": 34.0,
      "grad_norm": 0.35072898864746094,
      "learning_rate": 2.223074501950307e-05,
      "loss": 0.0248,
      "step": 3162
    },
    {
      "epoch": 34.0,
      "eval_f1_aspect": 0.83188484519283,
      "eval_f1_macro_aspect": 0.502266735609415,
      "eval_f1_sentiment": 0.502266735609415,
      "eval_loss": 0.4622325003147125,
      "eval_runtime": 9.3309,
      "eval_samples_per_second": 138.251,
      "eval_steps_per_second": 4.394,
      "step": 3162
    },
    {
      "epoch": 35.0,
      "grad_norm": 0.7930346131324768,
      "learning_rate": 2.1814371913188768e-05,
      "loss": 0.0235,
      "step": 3255
    },
    {
      "epoch": 35.0,
      "eval_f1_aspect": 0.8264797087771336,
      "eval_f1_macro_aspect": 0.4988629225098729,
      "eval_f1_sentiment": 0.4988629225098729,
      "eval_loss": 0.47125911712646484,
      "eval_runtime": 9.3454,
      "eval_samples_per_second": 138.035,
      "eval_steps_per_second": 4.387,
      "step": 3255
    },
    {
      "epoch": 36.0,
      "grad_norm": 0.4293688237667084,
      "learning_rate": 2.139127384450524e-05,
      "loss": 0.0226,
      "step": 3348
    },
    {
      "epoch": 36.0,
      "eval_f1_aspect": 0.8337644656228727,
      "eval_f1_macro_aspect": 0.48743686932731856,
      "eval_f1_sentiment": 0.48743686932731856,
      "eval_loss": 0.45808976888656616,
      "eval_runtime": 9.3644,
      "eval_samples_per_second": 137.756,
      "eval_steps_per_second": 4.378,
      "step": 3348
    },
    {
      "epoch": 37.0,
      "grad_norm": 0.26249465346336365,
      "learning_rate": 2.0961868360165025e-05,
      "loss": 0.0206,
      "step": 3441
    },
    {
      "epoch": 37.0,
      "eval_f1_aspect": 0.8222400427693131,
      "eval_f1_macro_aspect": 0.5054661843425796,
      "eval_f1_sentiment": 0.5054661843425796,
      "eval_loss": 0.5085671544075012,
      "eval_runtime": 9.3502,
      "eval_samples_per_second": 137.966,
      "eval_steps_per_second": 4.385,
      "step": 3441
    },
    {
      "epoch": 38.0,
      "grad_norm": 0.4812083840370178,
      "learning_rate": 2.0526579231538418e-05,
      "loss": 0.0202,
      "step": 3534
    },
    {
      "epoch": 38.0,
      "eval_f1_aspect": 0.8257322455530293,
      "eval_f1_macro_aspect": 0.49273853416492536,
      "eval_f1_sentiment": 0.49273853416492536,
      "eval_loss": 0.5079413056373596,
      "eval_runtime": 9.3389,
      "eval_samples_per_second": 138.132,
      "eval_steps_per_second": 4.39,
      "step": 3534
    },
    {
      "epoch": 39.0,
      "grad_norm": 0.41249096393585205,
      "learning_rate": 2.008583603644228e-05,
      "loss": 0.0186,
      "step": 3627
    },
    {
      "epoch": 39.0,
      "eval_f1_aspect": 0.827416712179137,
      "eval_f1_macro_aspect": 0.505304770419867,
      "eval_f1_sentiment": 0.505304770419867,
      "eval_loss": 0.4699024260044098,
      "eval_runtime": 9.3379,
      "eval_samples_per_second": 138.147,
      "eval_steps_per_second": 4.391,
      "step": 3627
    },
    {
      "epoch": 40.0,
      "grad_norm": 0.5531007051467896,
      "learning_rate": 1.9640073735198596e-05,
      "loss": 0.0177,
      "step": 3720
    },
    {
      "epoch": 40.0,
      "eval_f1_aspect": 0.836542400655469,
      "eval_f1_macro_aspect": 0.501600642979722,
      "eval_f1_sentiment": 0.501600642979722,
      "eval_loss": 0.470309853553772,
      "eval_runtime": 9.348,
      "eval_samples_per_second": 137.997,
      "eval_steps_per_second": 4.386,
      "step": 3720
    },
    {
      "epoch": 41.0,
      "grad_norm": 0.5626578330993652,
      "learning_rate": 1.9189732241381162e-05,
      "loss": 0.0168,
      "step": 3813
    },
    {
      "epoch": 41.0,
      "eval_f1_aspect": 0.8326975476839237,
      "eval_f1_macro_aspect": 0.5021347533739974,
      "eval_f1_sentiment": 0.5021347533739974,
      "eval_loss": 0.47961315512657166,
      "eval_runtime": 9.3145,
      "eval_samples_per_second": 138.494,
      "eval_steps_per_second": 4.402,
      "step": 3813
    },
    {
      "epoch": 42.0,
      "grad_norm": 0.2621760666370392,
      "learning_rate": 1.873525598767399e-05,
      "loss": 0.0153,
      "step": 3906
    },
    {
      "epoch": 42.0,
      "eval_f1_aspect": 0.8300459335314779,
      "eval_f1_macro_aspect": 0.5035559634999267,
      "eval_f1_sentiment": 0.5035559634999267,
      "eval_loss": 0.4952414333820343,
      "eval_runtime": 9.3623,
      "eval_samples_per_second": 137.787,
      "eval_steps_per_second": 4.379,
      "step": 3906
    },
    {
      "epoch": 43.0,
      "grad_norm": 0.5963602662086487,
      "learning_rate": 1.8277093487269912e-05,
      "loss": 0.0145,
      "step": 3999
    },
    {
      "epoch": 43.0,
      "eval_f1_aspect": 0.8300027374760471,
      "eval_f1_macro_aspect": 0.5022869266053374,
      "eval_f1_sentiment": 0.5022869266053374,
      "eval_loss": 0.4856065511703491,
      "eval_runtime": 9.3475,
      "eval_samples_per_second": 138.005,
      "eval_steps_per_second": 4.386,
      "step": 3999
    },
    {
      "epoch": 44.0,
      "grad_norm": 0.10686498880386353,
      "learning_rate": 1.781569689124218e-05,
      "loss": 0.0138,
      "step": 4092
    },
    {
      "epoch": 44.0,
      "eval_f1_aspect": 0.8269858541893362,
      "eval_f1_macro_aspect": 0.5039532789685118,
      "eval_f1_sentiment": 0.5039532789685118,
      "eval_loss": 0.4970964193344116,
      "eval_runtime": 9.3393,
      "eval_samples_per_second": 138.126,
      "eval_steps_per_second": 4.39,
      "step": 4092
    },
    {
      "epoch": 45.0,
      "grad_norm": 0.33090656995773315,
      "learning_rate": 1.7351521542326e-05,
      "loss": 0.0128,
      "step": 4185
    },
    {
      "epoch": 45.0,
      "eval_f1_aspect": 0.8324442620708521,
      "eval_f1_macro_aspect": 0.5030698686911683,
      "eval_f1_sentiment": 0.5030698686911683,
      "eval_loss": 0.4739266335964203,
      "eval_runtime": 9.3292,
      "eval_samples_per_second": 138.276,
      "eval_steps_per_second": 4.395,
      "step": 4185
    },
    {
      "epoch": 46.0,
      "grad_norm": 0.1663566678762436,
      "learning_rate": 1.688502552555018e-05,
      "loss": 0.0123,
      "step": 4278
    },
    {
      "epoch": 46.0,
      "eval_f1_aspect": 0.834061135371179,
      "eval_f1_macro_aspect": 0.509172021050766,
      "eval_f1_sentiment": 0.509172021050766,
      "eval_loss": 0.48807352781295776,
      "eval_runtime": 9.3687,
      "eval_samples_per_second": 137.693,
      "eval_steps_per_second": 4.376,
      "step": 4278
    },
    {
      "epoch": 47.0,
      "grad_norm": 0.3777198791503906,
      "learning_rate": 1.6416669216162534e-05,
      "loss": 0.0111,
      "step": 4371
    },
    {
      "epoch": 47.0,
      "eval_f1_aspect": 0.8331075589271201,
      "eval_f1_macro_aspect": 0.5121130129547204,
      "eval_f1_sentiment": 0.5121130129547204,
      "eval_loss": 0.4972289204597473,
      "eval_runtime": 9.3277,
      "eval_samples_per_second": 138.297,
      "eval_steps_per_second": 4.395,
      "step": 4371
    },
    {
      "epoch": 48.0,
      "grad_norm": 0.12145251035690308,
      "learning_rate": 1.594691482529508e-05,
      "loss": 0.0108,
      "step": 4464
    },
    {
      "epoch": 48.0,
      "eval_f1_aspect": 0.8309993201903467,
      "eval_f1_macro_aspect": 0.5086307029029409,
      "eval_f1_sentiment": 0.5086307029029409,
      "eval_loss": 0.5049077868461609,
      "eval_runtime": 9.3524,
      "eval_samples_per_second": 137.932,
      "eval_steps_per_second": 4.384,
      "step": 4464
    },
    {
      "epoch": 49.0,
      "grad_norm": 0.07569893449544907,
      "learning_rate": 1.5476225943817477e-05,
      "loss": 0.0103,
      "step": 4557
    },
    {
      "epoch": 49.0,
      "eval_f1_aspect": 0.8257667068434444,
      "eval_f1_macro_aspect": 0.5098420899290514,
      "eval_f1_sentiment": 0.5098420899290514,
      "eval_loss": 0.527094304561615,
      "eval_runtime": 9.3619,
      "eval_samples_per_second": 137.793,
      "eval_steps_per_second": 4.379,
      "step": 4557
    },
    {
      "epoch": 50.0,
      "grad_norm": 0.32851970195770264,
      "learning_rate": 1.5005067084828775e-05,
      "loss": 0.0097,
      "step": 4650
    },
    {
      "epoch": 50.0,
      "eval_f1_aspect": 0.8259358288770053,
      "eval_f1_macro_aspect": 0.511524325131937,
      "eval_f1_sentiment": 0.511524325131937,
      "eval_loss": 0.54547119140625,
      "eval_runtime": 9.3572,
      "eval_samples_per_second": 137.862,
      "eval_steps_per_second": 4.382,
      "step": 4650
    },
    {
      "epoch": 51.0,
      "grad_norm": 0.23753219842910767,
      "learning_rate": 1.4533903225239106e-05,
      "loss": 0.0094,
      "step": 4743
    },
    {
      "epoch": 51.0,
      "eval_f1_aspect": 0.8308108108108108,
      "eval_f1_macro_aspect": 0.5060447117801372,
      "eval_f1_sentiment": 0.5060447117801372,
      "eval_loss": 0.5215189456939697,
      "eval_runtime": 9.3414,
      "eval_samples_per_second": 138.094,
      "eval_steps_per_second": 4.389,
      "step": 4743
    },
    {
      "epoch": 52.0,
      "grad_norm": 0.06389125436544418,
      "learning_rate": 1.406319934689358e-05,
      "loss": 0.0088,
      "step": 4836
    },
    {
      "epoch": 52.0,
      "eval_f1_aspect": 0.8305039073026138,
      "eval_f1_macro_aspect": 0.5091081741888608,
      "eval_f1_sentiment": 0.5091081741888608,
      "eval_loss": 0.5314034819602966,
      "eval_runtime": 9.3322,
      "eval_samples_per_second": 138.231,
      "eval_steps_per_second": 4.393,
      "step": 4836
    },
    {
      "epoch": 53.0,
      "grad_norm": 0.10959524661302567,
      "learning_rate": 1.359341997769136e-05,
      "loss": 0.0085,
      "step": 4929
    },
    {
      "epoch": 53.0,
      "eval_f1_aspect": 0.8315402424077353,
      "eval_f1_macro_aspect": 0.5054988827192769,
      "eval_f1_sentiment": 0.5054988827192769,
      "eval_loss": 0.522325873374939,
      "eval_runtime": 9.3288,
      "eval_samples_per_second": 138.281,
      "eval_steps_per_second": 4.395,
      "step": 4929
    },
    {
      "epoch": 54.0,
      "grad_norm": 0.10368623584508896,
      "learning_rate": 1.3125028733152692e-05,
      "loss": 0.0083,
      "step": 5022
    },
    {
      "epoch": 54.0,
      "eval_f1_aspect": 0.8303258820360894,
      "eval_f1_macro_aspect": 0.5126767891177076,
      "eval_f1_sentiment": 0.5126767891177076,
      "eval_loss": 0.5269874930381775,
      "eval_runtime": 9.3141,
      "eval_samples_per_second": 138.499,
      "eval_steps_per_second": 4.402,
      "step": 5022
    },
    {
      "epoch": 55.0,
      "grad_norm": 0.1779782474040985,
      "learning_rate": 1.2658487858886377e-05,
      "loss": 0.0081,
      "step": 5115
    },
    {
      "epoch": 55.0,
      "eval_f1_aspect": 0.8290253096392031,
      "eval_f1_macro_aspect": 0.5115460278242395,
      "eval_f1_sentiment": 0.5115460278242395,
      "eval_loss": 0.5419005751609802,
      "eval_runtime": 9.3319,
      "eval_samples_per_second": 138.235,
      "eval_steps_per_second": 4.394,
      "step": 5115
    },
    {
      "epoch": 56.0,
      "grad_norm": 0.1706460565328598,
      "learning_rate": 1.2194257774409154e-05,
      "loss": 0.0077,
      "step": 5208
    },
    {
      "epoch": 56.0,
      "eval_f1_aspect": 0.8326342849411447,
      "eval_f1_macro_aspect": 0.5092695492835801,
      "eval_f1_sentiment": 0.5092695492835801,
      "eval_loss": 0.527721107006073,
      "eval_runtime": 9.3661,
      "eval_samples_per_second": 137.731,
      "eval_steps_per_second": 4.378,
      "step": 5208
    },
    {
      "epoch": 57.0,
      "grad_norm": 0.12452230602502823,
      "learning_rate": 1.1732796618767227e-05,
      "loss": 0.0073,
      "step": 5301
    },
    {
      "epoch": 57.0,
      "eval_f1_aspect": 0.8280682887484877,
      "eval_f1_macro_aspect": 0.5039250475125738,
      "eval_f1_sentiment": 0.5039250475125738,
      "eval_loss": 0.5472165942192078,
      "eval_runtime": 9.3353,
      "eval_samples_per_second": 138.185,
      "eval_steps_per_second": 4.392,
      "step": 5301
    },
    {
      "epoch": 58.0,
      "grad_norm": 0.17888185381889343,
      "learning_rate": 1.1274559798408335e-05,
      "loss": 0.007,
      "step": 5394
    },
    {
      "epoch": 58.0,
      "eval_f1_aspect": 0.8300618445818768,
      "eval_f1_macro_aspect": 0.5115141204508211,
      "eval_f1_sentiment": 0.5115141204508211,
      "eval_loss": 0.5492764115333557,
      "eval_runtime": 9.3839,
      "eval_samples_per_second": 137.47,
      "eval_steps_per_second": 4.369,
      "step": 5394
    },
    {
      "epoch": 59.0,
      "grad_norm": 0.14206252992153168,
      "learning_rate": 1.0819999537750566e-05,
      "loss": 0.0066,
      "step": 5487
    },
    {
      "epoch": 59.0,
      "eval_f1_aspect": 0.8283018867924529,
      "eval_f1_macro_aspect": 0.5054447791896768,
      "eval_f1_sentiment": 0.5054447791896768,
      "eval_loss": 0.5587204694747925,
      "eval_runtime": 9.358,
      "eval_samples_per_second": 137.85,
      "eval_steps_per_second": 4.381,
      "step": 5487
    },
    {
      "epoch": 60.0,
      "grad_norm": 0.061074741184711456,
      "learning_rate": 1.0369564432891463e-05,
      "loss": 0.0064,
      "step": 5580
    },
    {
      "epoch": 60.0,
      "eval_f1_aspect": 0.8348274925292041,
      "eval_f1_macro_aspect": 0.5075519832469101,
      "eval_f1_sentiment": 0.5075519832469101,
      "eval_loss": 0.5378357172012329,
      "eval_runtime": 9.3288,
      "eval_samples_per_second": 138.281,
      "eval_steps_per_second": 4.395,
      "step": 5580
    },
    {
      "epoch": 61.0,
      "grad_norm": 0.10277830809354782,
      "learning_rate": 9.923699008897836e-06,
      "loss": 0.0061,
      "step": 5673
    },
    {
      "epoch": 61.0,
      "eval_f1_aspect": 0.8329488397340209,
      "eval_f1_macro_aspect": 0.508721066148471,
      "eval_f1_sentiment": 0.508721066148471,
      "eval_loss": 0.5482782125473022,
      "eval_runtime": 9.3421,
      "eval_samples_per_second": 138.085,
      "eval_steps_per_second": 4.389,
      "step": 5673
    },
    {
      "epoch": 62.0,
      "grad_norm": 0.19184204936027527,
      "learning_rate": 9.48284328111318e-06,
      "loss": 0.006,
      "step": 5766
    },
    {
      "epoch": 62.0,
      "eval_f1_aspect": 0.8309517390132111,
      "eval_f1_macro_aspect": 0.5086792171655632,
      "eval_f1_sentiment": 0.5086792171655632,
      "eval_loss": 0.548111081123352,
      "eval_runtime": 9.4091,
      "eval_samples_per_second": 137.101,
      "eval_steps_per_second": 4.357,
      "step": 5766
    },
    {
      "epoch": 63.0,
      "grad_norm": 0.08960551023483276,
      "learning_rate": 9.047432320915662e-06,
      "loss": 0.0058,
      "step": 5859
    },
    {
      "epoch": 63.0,
      "eval_f1_aspect": 0.8318009734991888,
      "eval_f1_macro_aspect": 0.5046846413594825,
      "eval_f1_sentiment": 0.5046846413594825,
      "eval_loss": 0.5570544600486755,
      "eval_runtime": 9.341,
      "eval_samples_per_second": 138.1,
      "eval_steps_per_second": 4.389,
      "step": 5859
    },
    {
      "epoch": 64.0,
      "grad_norm": 0.21702516078948975,
      "learning_rate": 8.617895826355196e-06,
      "loss": 0.0057,
      "step": 5952
    },
    {
      "epoch": 64.0,
      "eval_f1_aspect": 0.8305540613232921,
      "eval_f1_macro_aspect": 0.5038690490749029,
      "eval_f1_sentiment": 0.5038690490749029,
      "eval_loss": 0.558007001876831,
      "eval_runtime": 9.3124,
      "eval_samples_per_second": 138.525,
      "eval_steps_per_second": 4.403,
      "step": 5952
    },
    {
      "epoch": 65.0,
      "grad_norm": 0.1779937744140625,
      "learning_rate": 8.19465769809334e-06,
      "loss": 0.0056,
      "step": 6045
    },
    {
      "epoch": 65.0,
      "eval_f1_aspect": 0.8275676403284425,
      "eval_f1_macro_aspect": 0.5100772980163969,
      "eval_f1_sentiment": 0.5100772980163969,
      "eval_loss": 0.5643439292907715,
      "eval_runtime": 9.3701,
      "eval_samples_per_second": 137.672,
      "eval_steps_per_second": 4.376,
      "step": 6045
    },
    {
      "epoch": 66.0,
      "grad_norm": 0.16287781298160553,
      "learning_rate": 7.778135621064538e-06,
      "loss": 0.0056,
      "step": 6138
    },
    {
      "epoch": 66.0,
      "eval_f1_aspect": 0.8326331481230519,
      "eval_f1_macro_aspect": 0.5051451274344735,
      "eval_f1_sentiment": 0.5051451274344735,
      "eval_loss": 0.5547571182250977,
      "eval_runtime": 9.3211,
      "eval_samples_per_second": 138.396,
      "eval_steps_per_second": 4.399,
      "step": 6138
    },
    {
      "epoch": 67.0,
      "grad_norm": 0.10883475840091705,
      "learning_rate": 7.3687406522714935e-06,
      "loss": 0.0054,
      "step": 6231
    },
    {
      "epoch": 67.0,
      "eval_f1_aspect": 0.8321128288581503,
      "eval_f1_macro_aspect": 0.5017256104433099,
      "eval_f1_sentiment": 0.5017256104433099,
      "eval_loss": 0.5490269660949707,
      "eval_runtime": 9.3495,
      "eval_samples_per_second": 137.976,
      "eval_steps_per_second": 4.385,
      "step": 6231
    },
    {
      "epoch": 68.0,
      "grad_norm": 0.03851054236292839,
      "learning_rate": 6.966876815121545e-06,
      "loss": 0.005,
      "step": 6324
    },
    {
      "epoch": 68.0,
      "eval_f1_aspect": 0.8308853254278399,
      "eval_f1_macro_aspect": 0.5116775487268761,
      "eval_f1_sentiment": 0.5116775487268761,
      "eval_loss": 0.5618360638618469,
      "eval_runtime": 9.3528,
      "eval_samples_per_second": 137.926,
      "eval_steps_per_second": 4.384,
      "step": 6324
    },
    {
      "epoch": 69.0,
      "grad_norm": 0.07619833946228027,
      "learning_rate": 6.5729407007043604e-06,
      "loss": 0.0049,
      "step": 6417
    },
    {
      "epoch": 69.0,
      "eval_f1_aspect": 0.8336261656980674,
      "eval_f1_macro_aspect": 0.5103642555675019,
      "eval_f1_sentiment": 0.5103642555675019,
      "eval_loss": 0.5541679263114929,
      "eval_runtime": 9.3571,
      "eval_samples_per_second": 137.863,
      "eval_steps_per_second": 4.382,
      "step": 6417
    },
    {
      "epoch": 70.0,
      "grad_norm": 0.06904967874288559,
      "learning_rate": 6.187321076404377e-06,
      "loss": 0.0048,
      "step": 6510
    },
    {
      "epoch": 70.0,
      "eval_f1_aspect": 0.8326342849411447,
      "eval_f1_macro_aspect": 0.5086953588309843,
      "eval_f1_sentiment": 0.5086953588309843,
      "eval_loss": 0.5588096976280212,
      "eval_runtime": 9.3394,
      "eval_samples_per_second": 138.125,
      "eval_steps_per_second": 4.39,
      "step": 6510
    },
    {
      "epoch": 71.0,
      "grad_norm": 0.17583686113357544,
      "learning_rate": 5.8103985022343725e-06,
      "loss": 0.0046,
      "step": 6603
    },
    {
      "epoch": 71.0,
      "eval_f1_aspect": 0.8329952670723462,
      "eval_f1_macro_aspect": 0.5078880365837567,
      "eval_f1_sentiment": 0.5078880365837567,
      "eval_loss": 0.5640448331832886,
      "eval_runtime": 9.3374,
      "eval_samples_per_second": 138.154,
      "eval_steps_per_second": 4.391,
      "step": 6603
    },
    {
      "epoch": 72.0,
      "grad_norm": 0.03971230611205101,
      "learning_rate": 5.4425449552686176e-06,
      "loss": 0.0047,
      "step": 6696
    },
    {
      "epoch": 72.0,
      "eval_f1_aspect": 0.831053901850362,
      "eval_f1_macro_aspect": 0.515357605277523,
      "eval_f1_sentiment": 0.515357605277523,
      "eval_loss": 0.5797184705734253,
      "eval_runtime": 9.337,
      "eval_samples_per_second": 138.16,
      "eval_steps_per_second": 4.391,
      "step": 6696
    }
  ],
  "logging_steps": 93,
  "max_steps": 9300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.2928415218023104e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}