{
  "best_global_step": 400,
  "best_metric": 0.9414717555046082,
  "best_model_checkpoint": "./python-code-summarization-ministral-8B/checkpoint-400",
  "epoch": 1.2541176470588236,
  "eval_steps": 50,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.5161276400089263,
      "epoch": 0.03137254901960784,
      "grad_norm": 0.4193158745765686,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.7534,
      "mean_token_accuracy": 0.5859726279973984,
      "num_tokens": 40960.0,
      "step": 10
    },
    {
      "entropy": 1.5331396609544754,
      "epoch": 0.06274509803921569,
      "grad_norm": 0.34798431396484375,
      "learning_rate": 9.991971098713136e-06,
      "loss": 1.7374,
      "mean_token_accuracy": 0.5868768319487572,
      "num_tokens": 81920.0,
      "step": 20
    },
    {
      "entropy": 1.504272100329399,
      "epoch": 0.09411764705882353,
      "grad_norm": 0.29647937417030334,
      "learning_rate": 9.952707929045018e-06,
      "loss": 1.5992,
      "mean_token_accuracy": 0.6027859210968017,
      "num_tokens": 122880.0,
      "step": 30
    },
    {
      "entropy": 1.4829266160726546,
      "epoch": 0.12549019607843137,
      "grad_norm": 0.35300737619400024,
      "learning_rate": 9.88099273969759e-06,
      "loss": 1.4991,
      "mean_token_accuracy": 0.619208213686943,
      "num_tokens": 163840.0,
      "step": 40
    },
    {
      "entropy": 1.4123090922832489,
      "epoch": 0.1568627450980392,
      "grad_norm": 0.34853842854499817,
      "learning_rate": 9.777295435775165e-06,
      "loss": 1.4053,
      "mean_token_accuracy": 0.6398732364177704,
      "num_tokens": 204738.0,
      "step": 50
    },
    {
      "epoch": 0.1568627450980392,
      "eval_entropy": 1.3041793243090312,
      "eval_loss": 1.2978447675704956,
      "eval_mean_token_accuracy": 0.6662148380279541,
      "eval_num_tokens": 204738.0,
      "eval_runtime": 55.6346,
      "eval_samples_per_second": 4.044,
      "eval_steps_per_second": 4.044,
      "step": 50
    },
    {
      "entropy": 1.2701195865869521,
      "epoch": 0.18823529411764706,
      "grad_norm": 0.2973419427871704,
      "learning_rate": 9.64229548131959e-06,
      "loss": 1.2751,
      "mean_token_accuracy": 0.6745112419128418,
      "num_tokens": 245698.0,
      "step": 60
    },
    {
      "entropy": 1.0886118918657304,
      "epoch": 0.2196078431372549,
      "grad_norm": 0.2847767770290375,
      "learning_rate": 9.476877447204308e-06,
      "loss": 1.1225,
      "mean_token_accuracy": 0.7075268790125847,
      "num_tokens": 286658.0,
      "step": 70
    },
    {
      "entropy": 1.1023929134011268,
      "epoch": 0.25098039215686274,
      "grad_norm": 0.2556367814540863,
      "learning_rate": 9.282125215090694e-06,
      "loss": 1.1156,
      "mean_token_accuracy": 0.7074291288852692,
      "num_tokens": 327618.0,
      "step": 80
    },
    {
      "entropy": 1.0319992125034332,
      "epoch": 0.2823529411764706,
      "grad_norm": 0.2943298816680908,
      "learning_rate": 9.059314875424553e-06,
      "loss": 1.0371,
      "mean_token_accuracy": 0.7196480900049209,
      "num_tokens": 368578.0,
      "step": 90
    },
    {
      "entropy": 1.0712109550833702,
      "epoch": 0.3137254901960784,
      "grad_norm": 0.28212958574295044,
      "learning_rate": 8.809906366007832e-06,
      "loss": 1.0558,
      "mean_token_accuracy": 0.716348971426487,
      "num_tokens": 409538.0,
      "step": 100
    },
    {
      "epoch": 0.3137254901960784,
      "eval_entropy": 1.0366067356533475,
      "eval_loss": 1.0299878120422363,
      "eval_mean_token_accuracy": 0.7208515254656473,
      "eval_num_tokens": 409538.0,
      "eval_runtime": 55.7692,
      "eval_samples_per_second": 4.034,
      "eval_steps_per_second": 4.034,
      "step": 100
    },
    {
      "entropy": 1.0164774924516677,
      "epoch": 0.34509803921568627,
      "grad_norm": 0.28942403197288513,
      "learning_rate": 8.535533905932739e-06,
      "loss": 1.0079,
      "mean_token_accuracy": 0.7282013669610023,
      "num_tokens": 450498.0,
      "step": 110
    },
    {
      "entropy": 1.0560581550002097,
      "epoch": 0.3764705882352941,
      "grad_norm": 0.2502649426460266,
      "learning_rate": 8.237995287558801e-06,
      "loss": 1.037,
      "mean_token_accuracy": 0.7196969777345658,
      "num_tokens": 491458.0,
      "step": 120
    },
    {
      "entropy": 1.030363565683365,
      "epoch": 0.40784313725490196,
      "grad_norm": 0.27328935265541077,
      "learning_rate": 7.919240096695877e-06,
      "loss": 1.0113,
      "mean_token_accuracy": 0.727834801375866,
      "num_tokens": 532418.0,
      "step": 130
    },
    {
      "entropy": 1.022090318799019,
      "epoch": 0.4392156862745098,
      "grad_norm": 0.2800595462322235,
      "learning_rate": 7.58135693817893e-06,
      "loss": 0.9984,
      "mean_token_accuracy": 0.7261974602937699,
      "num_tokens": 573378.0,
      "step": 140
    },
    {
      "entropy": 1.0286607086658477,
      "epoch": 0.47058823529411764,
      "grad_norm": 0.29165375232696533,
      "learning_rate": 7.226559750537462e-06,
      "loss": 1.0059,
      "mean_token_accuracy": 0.7231427162885666,
      "num_tokens": 614338.0,
      "step": 150
    },
    {
      "epoch": 0.47058823529411764,
      "eval_entropy": 1.0056122451358371,
      "eval_loss": 0.983330488204956,
      "eval_mean_token_accuracy": 0.7300966668128968,
      "eval_num_tokens": 614338.0,
      "eval_runtime": 55.6969,
      "eval_samples_per_second": 4.04,
      "eval_steps_per_second": 4.04,
      "step": 150
    },
    {
      "entropy": 1.0160210341215135,
      "epoch": 0.5019607843137255,
      "grad_norm": 0.3049905598163605,
      "learning_rate": 6.857173299431084e-06,
      "loss": 0.9897,
      "mean_token_accuracy": 0.7299120262265205,
      "num_tokens": 655298.0,
      "step": 160
    },
    {
      "entropy": 1.0173547983169555,
      "epoch": 0.5333333333333333,
      "grad_norm": 0.29565489292144775,
      "learning_rate": 6.475617944903693e-06,
      "loss": 1.0005,
      "mean_token_accuracy": 0.7288611873984336,
      "num_tokens": 696258.0,
      "step": 170
    },
    {
      "entropy": 1.0144968718290328,
      "epoch": 0.5647058823529412,
      "grad_norm": 0.3220125436782837,
      "learning_rate": 6.084393782267039e-06,
      "loss": 0.9866,
      "mean_token_accuracy": 0.7293499529361724,
      "num_tokens": 737218.0,
      "step": 180
    },
    {
      "entropy": 0.9772199034690857,
      "epoch": 0.596078431372549,
      "grad_norm": 0.32867270708084106,
      "learning_rate": 5.686064260528577e-06,
      "loss": 0.9618,
      "mean_token_accuracy": 0.7332600191235542,
      "num_tokens": 778178.0,
      "step": 190
    },
    {
      "entropy": 0.9724952712655067,
      "epoch": 0.6274509803921569,
      "grad_norm": 0.3363828957080841,
      "learning_rate": 5.283239385701882e-06,
      "loss": 0.9558,
      "mean_token_accuracy": 0.7365591391921044,
      "num_tokens": 819138.0,
      "step": 200
    },
    {
      "epoch": 0.6274509803921569,
      "eval_entropy": 0.9774409135182699,
      "eval_loss": 0.9610546827316284,
      "eval_mean_token_accuracy": 0.7337851632965936,
      "eval_num_tokens": 819138.0,
      "eval_runtime": 55.7878,
      "eval_samples_per_second": 4.033,
      "eval_steps_per_second": 4.033,
      "step": 200
    },
    {
      "entropy": 0.98491101115942,
      "epoch": 0.6588235294117647,
      "grad_norm": 0.3083605170249939,
      "learning_rate": 4.8785586190579355e-06,
      "loss": 0.9618,
      "mean_token_accuracy": 0.7304252177476883,
      "num_tokens": 860098.0,
      "step": 210
    },
    {
      "entropy": 1.0157848447561264,
      "epoch": 0.6901960784313725,
      "grad_norm": 0.30706891417503357,
      "learning_rate": 4.474673582374381e-06,
      "loss": 1.0052,
      "mean_token_accuracy": 0.7238514170050621,
      "num_tokens": 901058.0,
      "step": 220
    },
    {
      "entropy": 0.9612855270504952,
      "epoch": 0.7215686274509804,
      "grad_norm": 0.3250674605369568,
      "learning_rate": 4.074230683504541e-06,
      "loss": 0.9545,
      "mean_token_accuracy": 0.7387096747756005,
      "num_tokens": 942018.0,
      "step": 230
    },
    {
      "entropy": 0.9811745539307595,
      "epoch": 0.7529411764705882,
      "grad_norm": 0.3410372734069824,
      "learning_rate": 3.679853776110035e-06,
      "loss": 0.9529,
      "mean_token_accuracy": 0.7300586521625518,
      "num_tokens": 982978.0,
      "step": 240
    },
    {
      "entropy": 0.9820658221840859,
      "epoch": 0.7843137254901961,
      "grad_norm": 0.3213532567024231,
      "learning_rate": 3.2941269671770193e-06,
      "loss": 0.9632,
      "mean_token_accuracy": 0.7317448645830155,
      "num_tokens": 1023938.0,
      "step": 250
    },
    {
      "epoch": 0.7843137254901961,
      "eval_entropy": 0.9663468877474467,
      "eval_loss": 0.9494678974151611,
      "eval_mean_token_accuracy": 0.7356924083497789,
      "eval_num_tokens": 1023938.0,
      "eval_runtime": 55.6726,
      "eval_samples_per_second": 4.041,
      "eval_steps_per_second": 4.041,
      "step": 250
    },
    {
      "entropy": 0.9845806032419204,
      "epoch": 0.8156862745098039,
      "grad_norm": 0.3576700687408447,
      "learning_rate": 2.9195776849677037e-06,
      "loss": 0.9711,
      "mean_token_accuracy": 0.7322091892361641,
      "num_tokens": 1064898.0,
      "step": 260
    },
    {
      "entropy": 0.9538720920681953,
      "epoch": 0.8470588235294118,
      "grad_norm": 0.34291872382164,
      "learning_rate": 2.5586601183523485e-06,
      "loss": 0.934,
      "mean_token_accuracy": 0.7407869040966034,
      "num_tokens": 1105858.0,
      "step": 270
    },
    {
      "entropy": 0.931637741625309,
      "epoch": 0.8784313725490196,
      "grad_norm": 0.3526791036128998,
      "learning_rate": 2.213739136033533e-06,
      "loss": 0.9127,
      "mean_token_accuracy": 0.745845553278923,
      "num_tokens": 1146818.0,
      "step": 280
    },
    {
      "entropy": 0.9876263022422791,
      "epoch": 0.9098039215686274,
      "grad_norm": 0.36476704478263855,
      "learning_rate": 1.8870747910300064e-06,
      "loss": 0.9755,
      "mean_token_accuracy": 0.7316226810216904,
      "num_tokens": 1187778.0,
      "step": 290
    },
    {
      "entropy": 0.9182186275720596,
      "epoch": 0.9411764705882353,
      "grad_norm": 0.3551705777645111,
      "learning_rate": 1.5808075119526323e-06,
      "loss": 0.8992,
      "mean_token_accuracy": 0.7443059593439102,
      "num_tokens": 1228738.0,
      "step": 300
    },
    {
      "epoch": 0.9411764705882353,
      "eval_entropy": 0.960735419591268,
      "eval_loss": 0.9439367651939392,
      "eval_mean_token_accuracy": 0.7366395137045119,
      "eval_num_tokens": 1228738.0,
      "eval_runtime": 55.8238,
      "eval_samples_per_second": 4.031,
      "eval_steps_per_second": 4.031,
      "step": 300
    },
    {
      "entropy": 0.9542466774582863,
      "epoch": 0.9725490196078431,
      "grad_norm": 0.3290831744670868,
      "learning_rate": 1.2969440781048015e-06,
      "loss": 0.9373,
      "mean_token_accuracy": 0.7406158342957496,
      "num_tokens": 1269698.0,
      "step": 310
    },
    {
      "entropy": 0.992382885553898,
      "epoch": 1.0031372549019608,
      "grad_norm": 0.35219520330429077,
      "learning_rate": 1.0373444703037644e-06,
      "loss": 0.9847,
      "mean_token_accuracy": 0.7291275018300766,
      "num_tokens": 1309634.0,
      "step": 320
    },
    {
      "entropy": 0.9549886897206307,
      "epoch": 1.0345098039215685,
      "grad_norm": 0.33679652214050293,
      "learning_rate": 8.037096835812885e-07,
      "loss": 0.9334,
      "mean_token_accuracy": 0.7392717510461807,
      "num_tokens": 1350594.0,
      "step": 330
    },
    {
      "entropy": 0.9770316660404206,
      "epoch": 1.0658823529411765,
      "grad_norm": 0.3329678475856781,
      "learning_rate": 5.97570581619446e-07,
      "loss": 0.9528,
      "mean_token_accuracy": 0.7311094805598259,
      "num_tokens": 1391554.0,
      "step": 340
    },
    {
      "entropy": 0.9595620423555374,
      "epoch": 1.0972549019607842,
      "grad_norm": 0.3248395323753357,
      "learning_rate": 4.2027786595149555e-07,
      "loss": 0.9332,
      "mean_token_accuracy": 0.7348729267716407,
      "num_tokens": 1432514.0,
      "step": 350
    },
    {
      "epoch": 1.0972549019607842,
      "eval_entropy": 0.9590845820638868,
      "eval_loss": 0.9417916536331177,
      "eval_mean_token_accuracy": 0.737252090771993,
      "eval_num_tokens": 1432514.0,
      "eval_runtime": 55.6945,
      "eval_samples_per_second": 4.04,
      "eval_steps_per_second": 4.04,
      "step": 350
    },
    {
      "entropy": 0.9660995692014694,
      "epoch": 1.1286274509803922,
      "grad_norm": 0.3455366790294647,
      "learning_rate": 2.7299322565344956e-07,
      "loss": 0.9624,
      "mean_token_accuracy": 0.7324291303753853,
      "num_tokens": 1473474.0,
      "step": 360
    },
    {
      "entropy": 0.9640696182847023,
      "epoch": 1.16,
      "grad_norm": 0.36158645153045654,
      "learning_rate": 1.5668172551691174e-07,
      "loss": 0.9559,
      "mean_token_accuracy": 0.7369257047772407,
      "num_tokens": 1514434.0,
      "step": 370
    },
    {
      "entropy": 0.9180433616042137,
      "epoch": 1.1913725490196079,
      "grad_norm": 0.33794501423835754,
      "learning_rate": 7.210548257874927e-08,
      "loss": 0.8909,
      "mean_token_accuracy": 0.7486803516745567,
      "num_tokens": 1555394.0,
      "step": 380
    },
    {
      "entropy": 0.9517706975340843,
      "epoch": 1.2227450980392156,
      "grad_norm": 0.365311861038208,
      "learning_rate": 1.981867244139124e-08,
      "loss": 0.9339,
      "mean_token_accuracy": 0.7413001000881195,
      "num_tokens": 1596354.0,
      "step": 390
    },
    {
      "entropy": 0.9845273345708847,
      "epoch": 1.2541176470588236,
      "grad_norm": 0.3574397563934326,
      "learning_rate": 1.6389810421846286e-10,
      "loss": 0.9673,
      "mean_token_accuracy": 0.7334555193781853,
      "num_tokens": 1637314.0,
      "step": 400
    },
    {
      "epoch": 1.2541176470588236,
      "eval_entropy": 0.9607294366094801,
      "eval_loss": 0.9414717555046082,
      "eval_mean_token_accuracy": 0.7371565114127265,
      "eval_num_tokens": 1637314.0,
      "eval_runtime": 55.7023,
      "eval_samples_per_second": 4.039,
      "eval_steps_per_second": 4.039,
      "step": 400
    }
  ],
  "logging_steps": 10,
  "max_steps": 400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.522663976342323e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}