{
  "best_metric": 0.49332907795906067,
  "best_model_checkpoint": "model_output/e2e_opentable_5_way__approximate__0-shot__seed-66__lstm/checkpoint-650",
  "epoch": 1.1149228130360207,
  "global_step": 650,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 4.9783362218370885e-05,
      "loss": 43.289,
      "step": 20
    },
    {
      "epoch": 0.07,
      "learning_rate": 4.956672443674177e-05,
      "loss": 39.7742,
      "step": 40
    },
    {
      "epoch": 0.09,
      "eval_distillation_accuracy_counterfactual": 0.3876346844535659,
      "eval_distillation_accuracy_factual": 0.7116469984607491,
      "eval_distillation_f1_counterfactual": 0.36789273743176754,
      "eval_distillation_f1_factual": 0.706355718089049,
      "eval_groundtruth_accuracy_counterfactual": 0.3170856849666496,
      "eval_groundtruth_f1_counterfactual": 0.30061734733599244,
      "eval_groundtruth_f1_factual": 0.5309938705776336,
      "eval_icace_cosine": 0.7443665266036987,
      "eval_icace_l2": 0.8025880455970764,
      "eval_icace_normdiff": 0.6179075837135315,
      "eval_loss": 21.722654342651367,
      "eval_runtime": 4.6356,
      "eval_samples_per_second": 840.887,
      "eval_steps_per_second": 6.687,
      "step": 50
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.935008665511265e-05,
      "loss": 36.1211,
      "step": 60
    },
    {
      "epoch": 0.14,
      "learning_rate": 4.913344887348354e-05,
      "loss": 33.0051,
      "step": 80
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.891681109185442e-05,
      "loss": 29.0866,
      "step": 100
    },
    {
      "epoch": 0.17,
      "eval_distillation_accuracy_counterfactual": 0.4179066187788609,
      "eval_distillation_accuracy_factual": 0.7213955874807594,
      "eval_distillation_f1_counterfactual": 0.4011779697256409,
      "eval_distillation_f1_factual": 0.7130234220148569,
      "eval_groundtruth_accuracy_counterfactual": 0.36198050282196,
      "eval_groundtruth_f1_counterfactual": 0.34119540585580543,
      "eval_groundtruth_f1_factual": 0.533506870286257,
      "eval_icace_cosine": 0.6045525074005127,
      "eval_icace_l2": 0.7391046285629272,
      "eval_icace_normdiff": 0.5667964220046997,
      "eval_loss": 15.981736183166504,
      "eval_runtime": 4.6279,
      "eval_samples_per_second": 842.278,
      "eval_steps_per_second": 6.698,
      "step": 100
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.8700173310225307e-05,
      "loss": 24.2452,
      "step": 120
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.848353552859619e-05,
      "loss": 20.9734,
      "step": 140
    },
    {
      "epoch": 0.26,
      "eval_distillation_accuracy_counterfactual": 0.4415084658799384,
      "eval_distillation_accuracy_factual": 0.6688045151359672,
      "eval_distillation_f1_counterfactual": 0.39743600677420876,
      "eval_distillation_f1_factual": 0.6386265115556637,
      "eval_groundtruth_accuracy_counterfactual": 0.42432016418676244,
      "eval_groundtruth_f1_counterfactual": 0.37094685868809796,
      "eval_groundtruth_f1_factual": 0.49392886445649786,
      "eval_icace_cosine": 0.5455384254455566,
      "eval_icace_l2": 0.6967277526855469,
      "eval_icace_normdiff": 0.46749597787857056,
      "eval_loss": 12.647016525268555,
      "eval_runtime": 4.5013,
      "eval_samples_per_second": 865.969,
      "eval_steps_per_second": 6.887,
      "step": 150
    },
    {
      "epoch": 0.27,
      "learning_rate": 4.826689774696707e-05,
      "loss": 19.1703,
      "step": 160
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.8050259965337955e-05,
      "loss": 17.6579,
      "step": 180
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.7833622183708845e-05,
      "loss": 17.1812,
      "step": 200
    },
    {
      "epoch": 0.34,
      "eval_distillation_accuracy_counterfactual": 0.46587993842996406,
      "eval_distillation_accuracy_factual": 0.7129297075423294,
      "eval_distillation_f1_counterfactual": 0.42585340396044125,
      "eval_distillation_f1_factual": 0.6801681890460928,
      "eval_groundtruth_accuracy_counterfactual": 0.44766546947152386,
      "eval_groundtruth_f1_counterfactual": 0.4011001985251402,
      "eval_groundtruth_f1_factual": 0.5200271047795778,
      "eval_icace_cosine": 0.5150377154350281,
      "eval_icace_l2": 0.6690810322761536,
      "eval_icace_normdiff": 0.4451918601989746,
      "eval_loss": 11.362848281860352,
      "eval_runtime": 4.5371,
      "eval_samples_per_second": 859.147,
      "eval_steps_per_second": 6.833,
      "step": 200
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.761698440207972e-05,
      "loss": 16.2793,
      "step": 220
    },
    {
      "epoch": 0.41,
      "learning_rate": 4.740034662045061e-05,
      "loss": 15.8969,
      "step": 240
    },
    {
      "epoch": 0.43,
      "eval_distillation_accuracy_counterfactual": 0.48255515649050795,
      "eval_distillation_accuracy_factual": 0.7508978963571062,
      "eval_distillation_f1_counterfactual": 0.4392245827288441,
      "eval_distillation_f1_factual": 0.7287650814209791,
      "eval_groundtruth_accuracy_counterfactual": 0.4574140584915341,
      "eval_groundtruth_f1_counterfactual": 0.4058681312407287,
      "eval_groundtruth_f1_factual": 0.5340988634733079,
      "eval_icace_cosine": 0.5094192028045654,
      "eval_icace_l2": 0.6640841960906982,
      "eval_icace_normdiff": 0.43985918164253235,
      "eval_loss": 11.062640190124512,
      "eval_runtime": 4.5851,
      "eval_samples_per_second": 850.14,
      "eval_steps_per_second": 6.761,
      "step": 250
    },
    {
      "epoch": 0.45,
      "learning_rate": 4.7183708838821494e-05,
      "loss": 15.567,
      "step": 260
    },
    {
      "epoch": 0.48,
      "learning_rate": 4.6967071057192376e-05,
      "loss": 15.5899,
      "step": 280
    },
    {
      "epoch": 0.51,
      "learning_rate": 4.675043327556326e-05,
      "loss": 15.4491,
      "step": 300
    },
    {
      "epoch": 0.51,
      "eval_distillation_accuracy_counterfactual": 0.49589533093894306,
      "eval_distillation_accuracy_factual": 0.7701385325808107,
      "eval_distillation_f1_counterfactual": 0.4490615827265524,
      "eval_distillation_f1_factual": 0.7488952781611599,
      "eval_groundtruth_accuracy_counterfactual": 0.47614161108260644,
      "eval_groundtruth_f1_counterfactual": 0.4213178039314829,
      "eval_groundtruth_f1_factual": 0.5320832145764172,
      "eval_icace_cosine": 0.5026495456695557,
      "eval_icace_l2": 0.6473169922828674,
      "eval_icace_normdiff": 0.4321483075618744,
      "eval_loss": 10.44105339050293,
      "eval_runtime": 4.6207,
      "eval_samples_per_second": 843.602,
      "eval_steps_per_second": 6.709,
      "step": 300
    },
    {
      "epoch": 0.55,
      "learning_rate": 4.653379549393415e-05,
      "loss": 15.2629,
      "step": 320
    },
    {
      "epoch": 0.58,
      "learning_rate": 4.6317157712305025e-05,
      "loss": 14.8008,
      "step": 340
    },
    {
      "epoch": 0.6,
      "eval_distillation_accuracy_counterfactual": 0.50846587993843,
      "eval_distillation_accuracy_factual": 0.7747562852744997,
      "eval_distillation_f1_counterfactual": 0.45682754564452965,
      "eval_distillation_f1_factual": 0.7559415583306467,
      "eval_groundtruth_accuracy_counterfactual": 0.49461262185736277,
      "eval_groundtruth_f1_counterfactual": 0.4326301439786258,
      "eval_groundtruth_f1_factual": 0.5477637349101722,
      "eval_icace_cosine": 0.5030019283294678,
      "eval_icace_l2": 0.6434506177902222,
      "eval_icace_normdiff": 0.4238852560520172,
      "eval_loss": 10.315159797668457,
      "eval_runtime": 4.5363,
      "eval_samples_per_second": 859.299,
      "eval_steps_per_second": 6.834,
      "step": 350
    },
    {
      "epoch": 0.62,
      "learning_rate": 4.6100519930675915e-05,
      "loss": 14.6876,
      "step": 360
    },
    {
      "epoch": 0.65,
      "learning_rate": 4.58838821490468e-05,
      "loss": 14.8699,
      "step": 380
    },
    {
      "epoch": 0.69,
      "learning_rate": 4.566724436741768e-05,
      "loss": 14.3057,
      "step": 400
    },
    {
      "epoch": 0.69,
      "eval_distillation_accuracy_counterfactual": 0.5179579271421242,
      "eval_distillation_accuracy_factual": 0.7955361723961005,
      "eval_distillation_f1_counterfactual": 0.4694720290698628,
      "eval_distillation_f1_factual": 0.7787731275919787,
      "eval_groundtruth_accuracy_counterfactual": 0.5007696254489482,
      "eval_groundtruth_f1_counterfactual": 0.44241001562042664,
      "eval_groundtruth_f1_factual": 0.5479747801171813,
      "eval_icace_cosine": 0.4990290403366089,
      "eval_icace_l2": 0.6423947811126709,
      "eval_icace_normdiff": 0.4277788996696472,
      "eval_loss": 10.05081558227539,
      "eval_runtime": 4.5234,
      "eval_samples_per_second": 861.75,
      "eval_steps_per_second": 6.853,
      "step": 400
    },
    {
      "epoch": 0.72,
      "learning_rate": 4.5450606585788563e-05,
      "loss": 14.2025,
      "step": 420
    },
    {
      "epoch": 0.75,
      "learning_rate": 4.5233968804159446e-05,
      "loss": 14.1203,
      "step": 440
    },
    {
      "epoch": 0.77,
      "eval_distillation_accuracy_counterfactual": 0.5023088763468445,
      "eval_distillation_accuracy_factual": 0.8127244740892765,
      "eval_distillation_f1_counterfactual": 0.4656964220693382,
      "eval_distillation_f1_factual": 0.8006616317977919,
      "eval_groundtruth_accuracy_counterfactual": 0.4789635710620831,
      "eval_groundtruth_f1_counterfactual": 0.4346768246991469,
      "eval_groundtruth_f1_factual": 0.5734933636976969,
      "eval_icace_cosine": 0.5007656216621399,
      "eval_icace_l2": 0.6549262404441833,
      "eval_icace_normdiff": 0.4307872951030731,
      "eval_loss": 10.223078727722168,
      "eval_runtime": 4.7741,
      "eval_samples_per_second": 816.492,
      "eval_steps_per_second": 6.493,
      "step": 450
    },
    {
      "epoch": 0.79,
      "learning_rate": 4.501733102253033e-05,
      "loss": 14.0055,
      "step": 460
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.480069324090121e-05,
      "loss": 14.1732,
      "step": 480
    },
    {
      "epoch": 0.86,
      "learning_rate": 4.45840554592721e-05,
      "loss": 13.7579,
      "step": 500
    },
    {
      "epoch": 0.86,
      "eval_distillation_accuracy_counterfactual": 0.5017957927142124,
      "eval_distillation_accuracy_factual": 0.8217034376603386,
      "eval_distillation_f1_counterfactual": 0.45110748338345036,
      "eval_distillation_f1_factual": 0.8120033725947984,
      "eval_groundtruth_accuracy_counterfactual": 0.491277578245254,
      "eval_groundtruth_f1_counterfactual": 0.4306026913746316,
      "eval_groundtruth_f1_factual": 0.5751269063395339,
      "eval_icace_cosine": 0.5004109144210815,
      "eval_icace_l2": 0.6552226543426514,
      "eval_icace_normdiff": 0.43052297830581665,
      "eval_loss": 10.176661491394043,
      "eval_runtime": 8.7551,
      "eval_samples_per_second": 445.225,
      "eval_steps_per_second": 3.541,
      "step": 500
    },
    {
      "epoch": 0.89,
      "learning_rate": 4.436741767764298e-05,
      "loss": 13.774,
      "step": 520
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.415077989601387e-05,
      "loss": 13.8,
      "step": 540
    },
    {
      "epoch": 0.94,
      "eval_distillation_accuracy_counterfactual": 0.5059004617752694,
      "eval_distillation_accuracy_factual": 0.8234992303745511,
      "eval_distillation_f1_counterfactual": 0.4512025853019249,
      "eval_distillation_f1_factual": 0.8135629436467913,
      "eval_groundtruth_accuracy_counterfactual": 0.49461262185736277,
      "eval_groundtruth_f1_counterfactual": 0.43348852199044147,
      "eval_groundtruth_f1_factual": 0.5789128560578634,
      "eval_icace_cosine": 0.498872309923172,
      "eval_icace_l2": 0.6550716757774353,
      "eval_icace_normdiff": 0.42937180399894714,
      "eval_loss": 10.230228424072266,
      "eval_runtime": 7.836,
      "eval_samples_per_second": 497.446,
      "eval_steps_per_second": 3.956,
      "step": 550
    },
    {
      "epoch": 0.96,
      "learning_rate": 4.393414211438475e-05,
      "loss": 13.5074,
      "step": 560
    },
    {
      "epoch": 0.99,
      "learning_rate": 4.371750433275563e-05,
      "loss": 13.6928,
      "step": 580
    },
    {
      "epoch": 1.03,
      "learning_rate": 4.3500866551126516e-05,
      "loss": 13.7141,
      "step": 600
    },
    {
      "epoch": 1.03,
      "eval_distillation_accuracy_counterfactual": 0.5179579271421242,
      "eval_distillation_accuracy_factual": 0.8306824012314007,
      "eval_distillation_f1_counterfactual": 0.45833699531570044,
      "eval_distillation_f1_factual": 0.8222217849303248,
      "eval_groundtruth_accuracy_counterfactual": 0.5033350436121088,
      "eval_groundtruth_f1_counterfactual": 0.433570321967066,
      "eval_groundtruth_f1_factual": 0.5763197755988125,
      "eval_icace_cosine": 0.49790823459625244,
      "eval_icace_l2": 0.6532407402992249,
      "eval_icace_normdiff": 0.4311390519142151,
      "eval_loss": 10.064295768737793,
      "eval_runtime": 4.6211,
      "eval_samples_per_second": 843.531,
      "eval_steps_per_second": 6.708,
      "step": 600
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.3284228769497406e-05,
      "loss": 13.3673,
      "step": 620
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.306759098786828e-05,
      "loss": 13.4799,
      "step": 640
    },
    {
      "epoch": 1.11,
      "eval_distillation_accuracy_counterfactual": 0.5187275525910724,
      "eval_distillation_accuracy_factual": 0.8283735248845562,
      "eval_distillation_f1_counterfactual": 0.46382143672624687,
      "eval_distillation_f1_factual": 0.8207720593293605,
      "eval_groundtruth_accuracy_counterfactual": 0.5064135454079015,
      "eval_groundtruth_f1_counterfactual": 0.4409147466060839,
      "eval_groundtruth_f1_factual": 0.5772188705192105,
      "eval_icace_cosine": 0.49332907795906067,
      "eval_icace_l2": 0.6394243240356445,
      "eval_icace_normdiff": 0.4169619083404541,
      "eval_loss": 9.886384963989258,
      "eval_runtime": 4.5468,
      "eval_samples_per_second": 857.313,
      "eval_steps_per_second": 6.818,
      "step": 650
    }
  ],
  "max_steps": 4616,
  "num_train_epochs": 8,
  "total_flos": 98183903155200.0,
  "trial_name": null,
  "trial_params": null
}