{
  "best_global_step": 200,
  "best_metric": 0.9610546827316284,
  "best_model_checkpoint": "./python-code-summarization-ministral-8B/checkpoint-200",
  "epoch": 0.6274509803921569,
  "eval_steps": 50,
  "global_step": 200,
  "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
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.763541949957734e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}