{
  "best_global_step": 7359,
  "best_metric": 0.8462,
  "best_model_checkpoint": "models/NED/EMEA_human_only_tfidf_hybrid_long_v2_addheaders/Llama-3.1-8B-Instruct/checkpoint-7359",
  "epoch": 50.0,
  "eval_steps": 500,
  "global_step": 122650,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.1526817805758311,
      "epoch": 1.0,
      "grad_norm": 304.0,
      "learning_rate": 1.9989130434782608e-05,
      "loss": 0.7669,
      "mean_token_accuracy": 0.8752253057546777,
      "num_tokens": 15010779.0,
      "step": 2453
    },
    {
      "epoch": 1.0,
      "eval_entropy": 1.2358426589232225,
      "eval_loss": 0.6339517831802368,
      "eval_mean_token_accuracy": 0.8988095246828519,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 15010779.0,
      "eval_recall": 0.7308,
      "eval_runtime": 3.6399,
      "eval_samples_per_second": 7.143,
      "eval_steps_per_second": 3.571,
      "step": 2453
    },
    {
      "entropy": 1.3605892632720036,
      "epoch": 2.0,
      "grad_norm": 12.1875,
      "learning_rate": 2.9691098596284776e-05,
      "loss": 0.5437,
      "mean_token_accuracy": 0.9150349811612466,
      "num_tokens": 30021558.0,
      "step": 4906
    },
    {
      "epoch": 2.0,
      "eval_entropy": 1.1509519540346587,
      "eval_loss": 0.4853871166706085,
      "eval_mean_token_accuracy": 0.9201437464127173,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 30021558.0,
      "eval_recall": 0.7692,
      "eval_runtime": 3.627,
      "eval_samples_per_second": 7.168,
      "eval_steps_per_second": 3.584,
      "step": 4906
    },
    {
      "entropy": 1.1862413553719222,
      "epoch": 3.0,
      "grad_norm": 2.1875,
      "learning_rate": 2.9072539295620746e-05,
      "loss": 0.2619,
      "mean_token_accuracy": 0.9548876376794495,
      "num_tokens": 45032337.0,
      "step": 7359
    },
    {
      "epoch": 3.0,
      "eval_entropy": 1.019592651954064,
      "eval_loss": 0.5770813822746277,
      "eval_mean_token_accuracy": 0.9220362993387076,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 45032337.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6363,
      "eval_samples_per_second": 7.15,
      "eval_steps_per_second": 3.575,
      "step": 7359
    },
    {
      "entropy": 0.9634018300311497,
      "epoch": 4.0,
      "grad_norm": 0.1240234375,
      "learning_rate": 2.8453979994956713e-05,
      "loss": 0.1216,
      "mean_token_accuracy": 0.9782008502466845,
      "num_tokens": 60043116.0,
      "step": 9812
    },
    {
      "epoch": 4.0,
      "eval_entropy": 0.8699520321992728,
      "eval_loss": 0.5446107387542725,
      "eval_mean_token_accuracy": 0.940018314581651,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 60043116.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6143,
      "eval_samples_per_second": 7.194,
      "eval_steps_per_second": 3.597,
      "step": 9812
    },
    {
      "entropy": 0.7849812429144681,
      "epoch": 5.0,
      "grad_norm": 0.002227783203125,
      "learning_rate": 2.783542069429268e-05,
      "loss": 0.0517,
      "mean_token_accuracy": 0.9894482943411997,
      "num_tokens": 75053895.0,
      "step": 12265
    },
    {
      "epoch": 5.0,
      "eval_entropy": 0.6801113898937519,
      "eval_loss": 0.7289856672286987,
      "eval_mean_token_accuracy": 0.9444444454633273,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 75053895.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6486,
      "eval_samples_per_second": 7.126,
      "eval_steps_per_second": 3.563,
      "step": 12265
    },
    {
      "entropy": 0.6892432886826181,
      "epoch": 6.0,
      "grad_norm": 0.0004749298095703125,
      "learning_rate": 2.721686139362865e-05,
      "loss": 0.0209,
      "mean_token_accuracy": 0.9958273216359138,
      "num_tokens": 90064674.0,
      "step": 14718
    },
    {
      "epoch": 6.0,
      "eval_entropy": 0.577189931502709,
      "eval_loss": 0.7246649265289307,
      "eval_mean_token_accuracy": 0.9444444454633273,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 90064674.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6456,
      "eval_samples_per_second": 7.132,
      "eval_steps_per_second": 3.566,
      "step": 14718
    },
    {
      "entropy": 0.6557439389371696,
      "epoch": 7.0,
      "grad_norm": 0.000888824462890625,
      "learning_rate": 2.659830209296461e-05,
      "loss": 0.0078,
      "mean_token_accuracy": 0.9979321826393635,
      "num_tokens": 105075453.0,
      "step": 17171
    },
    {
      "epoch": 7.0,
      "eval_entropy": 0.5603500146132249,
      "eval_loss": 0.8045116662979126,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 105075453.0,
      "eval_recall": 0.8462,
      "eval_runtime": 5.557,
      "eval_samples_per_second": 4.679,
      "eval_steps_per_second": 2.339,
      "step": 17171
    },
    {
      "entropy": 0.6481096161976669,
      "epoch": 8.0,
      "grad_norm": 8.96453857421875e-05,
      "learning_rate": 2.597974279230058e-05,
      "loss": 0.0028,
      "mean_token_accuracy": 0.9993061645391568,
      "num_tokens": 120086232.0,
      "step": 19624
    },
    {
      "epoch": 8.0,
      "eval_entropy": 0.5650725089586698,
      "eval_loss": 0.8335245847702026,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 120086232.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6391,
      "eval_samples_per_second": 7.145,
      "eval_steps_per_second": 3.572,
      "step": 19624
    },
    {
      "entropy": 0.6384989822756452,
      "epoch": 9.0,
      "grad_norm": 0.00102996826171875,
      "learning_rate": 2.5361183491636548e-05,
      "loss": 0.0011,
      "mean_token_accuracy": 0.9997574686275129,
      "num_tokens": 135097011.0,
      "step": 22077
    },
    {
      "epoch": 9.0,
      "eval_entropy": 0.5437194108963013,
      "eval_loss": 0.8720409870147705,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 135097011.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6696,
      "eval_samples_per_second": 7.085,
      "eval_steps_per_second": 3.543,
      "step": 22077
    },
    {
      "entropy": 0.6327040182586792,
      "epoch": 10.0,
      "grad_norm": 0.00011968612670898438,
      "learning_rate": 2.4742624190972517e-05,
      "loss": 0.0002,
      "mean_token_accuracy": 0.9999592335818012,
      "num_tokens": 150107790.0,
      "step": 24530
    },
    {
      "epoch": 10.0,
      "eval_entropy": 0.5456434029799241,
      "eval_loss": 0.8786986470222473,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 150107790.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.7213,
      "eval_samples_per_second": 6.987,
      "eval_steps_per_second": 3.493,
      "step": 24530
    },
    {
      "entropy": 0.6342776355527636,
      "epoch": 11.0,
      "grad_norm": 2.9206275939941406e-05,
      "learning_rate": 2.412406489030848e-05,
      "loss": 0.0001,
      "mean_token_accuracy": 0.9999629396397,
      "num_tokens": 165118569.0,
      "step": 26983
    },
    {
      "epoch": 11.0,
      "eval_entropy": 0.5441241906239436,
      "eval_loss": 0.8776129484176636,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 165118569.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6285,
      "eval_samples_per_second": 7.165,
      "eval_steps_per_second": 3.583,
      "step": 26983
    },
    {
      "entropy": 0.6330991076222742,
      "epoch": 12.0,
      "grad_norm": 0.000823974609375,
      "learning_rate": 2.350550558964445e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 180129348.0,
      "step": 29436
    },
    {
      "epoch": 12.0,
      "eval_entropy": 0.544509245799138,
      "eval_loss": 0.88084477186203,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 180129348.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6661,
      "eval_samples_per_second": 7.092,
      "eval_steps_per_second": 3.546,
      "step": 29436
    },
    {
      "entropy": 0.6322705759061291,
      "epoch": 13.0,
      "grad_norm": 0.010498046875,
      "learning_rate": 2.2886946288980416e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 195140127.0,
      "step": 31889
    },
    {
      "epoch": 13.0,
      "eval_entropy": 0.5434356606923617,
      "eval_loss": 0.8842343091964722,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 195140127.0,
      "eval_recall": 0.8462,
      "eval_runtime": 4.1268,
      "eval_samples_per_second": 6.3,
      "eval_steps_per_second": 3.15,
      "step": 31889
    },
    {
      "entropy": 0.6316640121908612,
      "epoch": 14.0,
      "grad_norm": 0.0035552978515625,
      "learning_rate": 2.2268386988316383e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 210150906.0,
      "step": 34342
    },
    {
      "epoch": 14.0,
      "eval_entropy": 0.543243577847114,
      "eval_loss": 0.885927140712738,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 210150906.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.7188,
      "eval_samples_per_second": 6.991,
      "eval_steps_per_second": 3.496,
      "step": 34342
    },
    {
      "entropy": 0.6321596540070241,
      "epoch": 15.0,
      "grad_norm": 2.4199485778808594e-05,
      "learning_rate": 2.164982768765235e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 225161685.0,
      "step": 36795
    },
    {
      "epoch": 15.0,
      "eval_entropy": 0.5422769280580374,
      "eval_loss": 0.8823052644729614,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 225161685.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6723,
      "eval_samples_per_second": 7.08,
      "eval_steps_per_second": 3.54,
      "step": 36795
    },
    {
      "entropy": 0.6315903761194426,
      "epoch": 16.0,
      "grad_norm": 0.0291748046875,
      "learning_rate": 2.1031268386988316e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 240172464.0,
      "step": 39248
    },
    {
      "epoch": 16.0,
      "eval_entropy": 0.5426660546889672,
      "eval_loss": 0.8869765996932983,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 240172464.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6896,
      "eval_samples_per_second": 7.047,
      "eval_steps_per_second": 3.523,
      "step": 39248
    },
    {
      "entropy": 0.6317922561279472,
      "epoch": 17.0,
      "grad_norm": 0.0001850128173828125,
      "learning_rate": 2.0412709086324285e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 255183243.0,
      "step": 41701
    },
    {
      "epoch": 17.0,
      "eval_entropy": 0.542809899036701,
      "eval_loss": 0.8864607214927673,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 255183243.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6498,
      "eval_samples_per_second": 7.124,
      "eval_steps_per_second": 3.562,
      "step": 41701
    },
    {
      "entropy": 0.6319634849034763,
      "epoch": 18.0,
      "grad_norm": 2.1457672119140625e-05,
      "learning_rate": 1.979414978566025e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 270194022.0,
      "step": 44154
    },
    {
      "epoch": 18.0,
      "eval_entropy": 0.5426488243616544,
      "eval_loss": 0.8861849308013916,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 270194022.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6568,
      "eval_samples_per_second": 7.11,
      "eval_steps_per_second": 3.555,
      "step": 44154
    },
    {
      "entropy": 0.631338802688325,
      "epoch": 19.0,
      "grad_norm": 4.076957702636719e-05,
      "learning_rate": 1.9175590484996218e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 285204801.0,
      "step": 46607
    },
    {
      "epoch": 19.0,
      "eval_entropy": 0.5423762339812058,
      "eval_loss": 0.885791540145874,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 285204801.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.653,
      "eval_samples_per_second": 7.118,
      "eval_steps_per_second": 3.559,
      "step": 46607
    },
    {
      "entropy": 0.6311312203036976,
      "epoch": 20.0,
      "grad_norm": 0.0004634857177734375,
      "learning_rate": 1.8557031184332184e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 300215580.0,
      "step": 49060
    },
    {
      "epoch": 20.0,
      "eval_entropy": 0.5424229686076825,
      "eval_loss": 0.8889456987380981,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 300215580.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.651,
      "eval_samples_per_second": 7.121,
      "eval_steps_per_second": 3.561,
      "step": 49060
    },
    {
      "entropy": 0.631198678741249,
      "epoch": 21.0,
      "grad_norm": 0.00031280517578125,
      "learning_rate": 1.793847188366815e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 315226359.0,
      "step": 51513
    },
    {
      "epoch": 21.0,
      "eval_entropy": 0.5428222968028142,
      "eval_loss": 0.8843169808387756,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 315226359.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6619,
      "eval_samples_per_second": 7.1,
      "eval_steps_per_second": 3.55,
      "step": 51513
    },
    {
      "entropy": 0.6313406728478388,
      "epoch": 22.0,
      "grad_norm": 0.000759124755859375,
      "learning_rate": 1.731991258300412e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 330237138.0,
      "step": 53966
    },
    {
      "epoch": 22.0,
      "eval_entropy": 0.5427144765853882,
      "eval_loss": 0.8861469030380249,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 330237138.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6544,
      "eval_samples_per_second": 7.115,
      "eval_steps_per_second": 3.557,
      "step": 53966
    },
    {
      "entropy": 0.6313331465647263,
      "epoch": 23.0,
      "grad_norm": 0.00051116943359375,
      "learning_rate": 1.6701353282340083e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 345247917.0,
      "step": 56419
    },
    {
      "epoch": 23.0,
      "eval_entropy": 0.5423137545585632,
      "eval_loss": 0.8892049193382263,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 345247917.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6537,
      "eval_samples_per_second": 7.116,
      "eval_steps_per_second": 3.558,
      "step": 56419
    },
    {
      "entropy": 0.6310314053401527,
      "epoch": 24.0,
      "grad_norm": 3.600120544433594e-05,
      "learning_rate": 1.6082793981676053e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 360258696.0,
      "step": 58872
    },
    {
      "epoch": 24.0,
      "eval_entropy": 0.5423843631377587,
      "eval_loss": 0.8886714577674866,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 360258696.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6316,
      "eval_samples_per_second": 7.159,
      "eval_steps_per_second": 3.58,
      "step": 58872
    },
    {
      "entropy": 0.6315073234496484,
      "epoch": 25.0,
      "grad_norm": 7.82012939453125e-05,
      "learning_rate": 1.546423468101202e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 375269475.0,
      "step": 61325
    },
    {
      "epoch": 25.0,
      "eval_entropy": 0.5420686419193561,
      "eval_loss": 0.8865240812301636,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 375269475.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.613,
      "eval_samples_per_second": 7.196,
      "eval_steps_per_second": 3.598,
      "step": 61325
    },
    {
      "entropy": 0.632054461467718,
      "epoch": 26.0,
      "grad_norm": 0.00024318695068359375,
      "learning_rate": 1.4845675380347987e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 15010779.0,
      "step": 63778
    },
    {
      "epoch": 26.0,
      "eval_entropy": 0.5426568893285898,
      "eval_loss": 0.88667893409729,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 15010779.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.647,
      "eval_samples_per_second": 7.129,
      "eval_steps_per_second": 3.565,
      "step": 63778
    },
    {
      "entropy": 0.6314872418356777,
      "epoch": 27.0,
      "grad_norm": 0.00011396408081054688,
      "learning_rate": 1.4227116079683954e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 30021558.0,
      "step": 66231
    },
    {
      "epoch": 27.0,
      "eval_entropy": 0.5423887417866633,
      "eval_loss": 0.8907365798950195,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 30021558.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6242,
      "eval_samples_per_second": 7.174,
      "eval_steps_per_second": 3.587,
      "step": 66231
    },
    {
      "entropy": 0.6317801613055392,
      "epoch": 28.0,
      "grad_norm": 8.392333984375e-05,
      "learning_rate": 1.3608556779019922e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 45032337.0,
      "step": 68684
    },
    {
      "epoch": 28.0,
      "eval_entropy": 0.5428364735383254,
      "eval_loss": 0.885719358921051,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 45032337.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6828,
      "eval_samples_per_second": 7.06,
      "eval_steps_per_second": 3.53,
      "step": 68684
    },
    {
      "entropy": 0.6310389586555389,
      "epoch": 29.0,
      "grad_norm": 0.000774383544921875,
      "learning_rate": 1.2989997478355888e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 60043116.0,
      "step": 71137
    },
    {
      "epoch": 29.0,
      "eval_entropy": 0.5424722524789664,
      "eval_loss": 0.8864960074424744,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 60043116.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6359,
      "eval_samples_per_second": 7.151,
      "eval_steps_per_second": 3.576,
      "step": 71137
    },
    {
      "entropy": 0.6310345640461444,
      "epoch": 30.0,
      "grad_norm": 3.5762786865234375e-05,
      "learning_rate": 1.2371438177691856e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 75053895.0,
      "step": 73590
    },
    {
      "epoch": 30.0,
      "eval_entropy": 0.5427528161268967,
      "eval_loss": 0.8871183395385742,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 75053895.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6648,
      "eval_samples_per_second": 7.095,
      "eval_steps_per_second": 3.547,
      "step": 73590
    },
    {
      "entropy": 0.6307261824680745,
      "epoch": 31.0,
      "grad_norm": 0.00015163421630859375,
      "learning_rate": 1.1752878877027823e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 90064674.0,
      "step": 76043
    },
    {
      "epoch": 31.0,
      "eval_entropy": 0.5423439878683823,
      "eval_loss": 0.890313982963562,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 90064674.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6589,
      "eval_samples_per_second": 7.106,
      "eval_steps_per_second": 3.553,
      "step": 76043
    },
    {
      "entropy": 0.6317850742056279,
      "epoch": 32.0,
      "grad_norm": 0.0005035400390625,
      "learning_rate": 1.113431957636379e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 105075453.0,
      "step": 78496
    },
    {
      "epoch": 32.0,
      "eval_entropy": 0.5422184283916767,
      "eval_loss": 0.8882402181625366,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 105075453.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6075,
      "eval_samples_per_second": 7.207,
      "eval_steps_per_second": 3.604,
      "step": 78496
    },
    {
      "entropy": 0.6315069926961121,
      "epoch": 33.0,
      "grad_norm": 0.0079345703125,
      "learning_rate": 1.0515760275699757e-05,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 120086232.0,
      "step": 80949
    },
    {
      "epoch": 33.0,
      "eval_entropy": 0.5428683024186355,
      "eval_loss": 0.8859032988548279,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 120086232.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6537,
      "eval_samples_per_second": 7.116,
      "eval_steps_per_second": 3.558,
      "step": 80949
    },
    {
      "entropy": 0.6313212784246381,
      "epoch": 34.0,
      "grad_norm": 0.000885009765625,
      "learning_rate": 9.897200975035723e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 135097011.0,
      "step": 83402
    },
    {
      "epoch": 34.0,
      "eval_entropy": 0.5425068598527175,
      "eval_loss": 0.887780487537384,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 135097011.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6448,
      "eval_samples_per_second": 7.133,
      "eval_steps_per_second": 3.567,
      "step": 83402
    },
    {
      "entropy": 0.6308202771352254,
      "epoch": 35.0,
      "grad_norm": 0.00032806396484375,
      "learning_rate": 9.27864167437169e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 150107790.0,
      "step": 85855
    },
    {
      "epoch": 35.0,
      "eval_entropy": 0.54246619114509,
      "eval_loss": 0.8900800347328186,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 150107790.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6253,
      "eval_samples_per_second": 7.172,
      "eval_steps_per_second": 3.586,
      "step": 85855
    },
    {
      "entropy": 0.6310893858737767,
      "epoch": 36.0,
      "grad_norm": 0.00543212890625,
      "learning_rate": 8.660082373707658e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 165118569.0,
      "step": 88308
    },
    {
      "epoch": 36.0,
      "eval_entropy": 0.542354785479032,
      "eval_loss": 0.882867157459259,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 165118569.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6309,
      "eval_samples_per_second": 7.161,
      "eval_steps_per_second": 3.58,
      "step": 88308
    },
    {
      "entropy": 0.6313383878492308,
      "epoch": 37.0,
      "grad_norm": 0.0014495849609375,
      "learning_rate": 8.041523073043624e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 180129348.0,
      "step": 90761
    },
    {
      "epoch": 37.0,
      "eval_entropy": 0.5429406670423654,
      "eval_loss": 0.8894430994987488,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 180129348.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6047,
      "eval_samples_per_second": 7.213,
      "eval_steps_per_second": 3.606,
      "step": 90761
    },
    {
      "entropy": 0.6315074832012738,
      "epoch": 38.0,
      "grad_norm": 1.8477439880371094e-05,
      "learning_rate": 7.422963772379592e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 195140127.0,
      "step": 93214
    },
    {
      "epoch": 38.0,
      "eval_entropy": 0.5428708929281968,
      "eval_loss": 0.8853751420974731,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 195140127.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6095,
      "eval_samples_per_second": 7.203,
      "eval_steps_per_second": 3.602,
      "step": 93214
    },
    {
      "entropy": 0.6316086658156264,
      "epoch": 39.0,
      "grad_norm": 0.0019378662109375,
      "learning_rate": 6.804404471715559e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 210150906.0,
      "step": 95667
    },
    {
      "epoch": 39.0,
      "eval_entropy": 0.5423155472828791,
      "eval_loss": 0.8865050673484802,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 210150906.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6105,
      "eval_samples_per_second": 7.201,
      "eval_steps_per_second": 3.601,
      "step": 95667
    },
    {
      "entropy": 0.6319762418161253,
      "epoch": 40.0,
      "grad_norm": 0.0076904296875,
      "learning_rate": 6.185845171051526e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 225161685.0,
      "step": 98120
    },
    {
      "epoch": 40.0,
      "eval_entropy": 0.5423448315033546,
      "eval_loss": 0.887237012386322,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 225161685.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6062,
      "eval_samples_per_second": 7.21,
      "eval_steps_per_second": 3.605,
      "step": 98120
    },
    {
      "entropy": 0.6316094772090632,
      "epoch": 41.0,
      "grad_norm": 0.00040435791015625,
      "learning_rate": 5.567285870387493e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 240172464.0,
      "step": 100573
    },
    {
      "epoch": 41.0,
      "eval_entropy": 0.5424330555475675,
      "eval_loss": 0.8862788081169128,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 240172464.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6042,
      "eval_samples_per_second": 7.214,
      "eval_steps_per_second": 3.607,
      "step": 100573
    },
    {
      "entropy": 0.6310035889118581,
      "epoch": 42.0,
      "grad_norm": 0.0020294189453125,
      "learning_rate": 4.94872656972346e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 255183243.0,
      "step": 103026
    },
    {
      "epoch": 42.0,
      "eval_entropy": 0.5431472292313209,
      "eval_loss": 0.890018105506897,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 255183243.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6041,
      "eval_samples_per_second": 7.214,
      "eval_steps_per_second": 3.607,
      "step": 103026
    },
    {
      "entropy": 0.6312229550229838,
      "epoch": 43.0,
      "grad_norm": 0.0012969970703125,
      "learning_rate": 4.330167269059427e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 270194022.0,
      "step": 105479
    },
    {
      "epoch": 43.0,
      "eval_entropy": 0.5424636235603919,
      "eval_loss": 0.8868480324745178,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 270194022.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.606,
      "eval_samples_per_second": 7.21,
      "eval_steps_per_second": 3.605,
      "step": 105479
    },
    {
      "entropy": 0.631434175660063,
      "epoch": 44.0,
      "grad_norm": 7.390975952148438e-05,
      "learning_rate": 3.711607968395394e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 285204801.0,
      "step": 107932
    },
    {
      "epoch": 44.0,
      "eval_entropy": 0.5421680899766775,
      "eval_loss": 0.8860384821891785,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 285204801.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6344,
      "eval_samples_per_second": 7.154,
      "eval_steps_per_second": 3.577,
      "step": 107932
    },
    {
      "entropy": 0.6307510763127319,
      "epoch": 45.0,
      "grad_norm": 0.00927734375,
      "learning_rate": 3.0930486677313608e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 300215580.0,
      "step": 110385
    },
    {
      "epoch": 45.0,
      "eval_entropy": 0.54229736328125,
      "eval_loss": 0.8853968977928162,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 300215580.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.61,
      "eval_samples_per_second": 7.202,
      "eval_steps_per_second": 3.601,
      "step": 110385
    },
    {
      "entropy": 0.6315490893937595,
      "epoch": 46.0,
      "grad_norm": 0.0001239776611328125,
      "learning_rate": 2.474489367067328e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 315226359.0,
      "step": 112838
    },
    {
      "epoch": 46.0,
      "eval_entropy": 0.5422170620698196,
      "eval_loss": 0.8882192373275757,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 315226359.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.7084,
      "eval_samples_per_second": 7.011,
      "eval_steps_per_second": 3.506,
      "step": 112838
    },
    {
      "entropy": 0.6317317981380761,
      "epoch": 47.0,
      "grad_norm": 3.3855438232421875e-05,
      "learning_rate": 1.855930066403295e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 330237138.0,
      "step": 115291
    },
    {
      "epoch": 47.0,
      "eval_entropy": 0.5427549022894639,
      "eval_loss": 0.8879793882369995,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 330237138.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.6923,
      "eval_samples_per_second": 7.042,
      "eval_steps_per_second": 3.521,
      "step": 115291
    },
    {
      "entropy": 0.6314135375092869,
      "epoch": 48.0,
      "grad_norm": 0.0025634765625,
      "learning_rate": 1.2373707657392621e-06,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 345247917.0,
      "step": 117744
    },
    {
      "epoch": 48.0,
      "eval_entropy": 0.5423269546948947,
      "eval_loss": 0.887828528881073,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 345247917.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.661,
      "eval_samples_per_second": 7.102,
      "eval_steps_per_second": 3.551,
      "step": 117744
    },
    {
      "entropy": 0.6317788491650499,
      "epoch": 49.0,
      "grad_norm": 0.0015106201171875,
      "learning_rate": 6.18811465075229e-07,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 360258696.0,
      "step": 120197
    },
    {
      "epoch": 49.0,
      "eval_entropy": 0.5421000031324533,
      "eval_loss": 0.886226236820221,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 360258696.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.8724,
      "eval_samples_per_second": 6.714,
      "eval_steps_per_second": 3.357,
      "step": 120197
    },
    {
      "entropy": 0.6307675256881722,
      "epoch": 50.0,
      "grad_norm": 0.0003414154052734375,
      "learning_rate": 2.5216441119609984e-10,
      "loss": 0.0,
      "mean_token_accuracy": 1.0,
      "num_tokens": 375269475.0,
      "step": 122650
    },
    {
      "epoch": 50.0,
      "eval_entropy": 0.5427401478473957,
      "eval_loss": 0.888108491897583,
      "eval_mean_token_accuracy": 0.9358974374257601,
      "eval_num_gold": 26,
      "eval_num_guess": 26,
      "eval_num_tokens": 375269475.0,
      "eval_recall": 0.8462,
      "eval_runtime": 3.7116,
      "eval_samples_per_second": 7.005,
      "eval_steps_per_second": 3.503,
      "step": 122650
    }
  ],
  "logging_steps": 0,
  "max_steps": 122650,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.3796448253168845e+19,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}