{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.5231746031746032,
  "eval_steps": 300,
  "global_step": 600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.5461102724075317,
      "epoch": 0.0025396825396825397,
      "grad_norm": 3.896618366241455,
      "learning_rate": 0.0,
      "loss": 2.3711,
      "mean_token_accuracy": 0.6472751796245575,
      "num_tokens": 9621.0,
      "step": 1
    },
    {
      "entropy": 1.581420679233576,
      "epoch": 0.050793650793650794,
      "grad_norm": 2.02801251411438,
      "learning_rate": 9.5e-06,
      "loss": 1.9565,
      "mean_token_accuracy": 0.6899475496458379,
      "num_tokens": 194969.0,
      "step": 20
    },
    {
      "entropy": 1.8750008303672074,
      "epoch": 0.10158730158730159,
      "grad_norm": 0.83933025598526,
      "learning_rate": 1.95e-05,
      "loss": 1.4223,
      "mean_token_accuracy": 0.7304104179143905,
      "num_tokens": 390209.0,
      "step": 40
    },
    {
      "entropy": 1.868852799385786,
      "epoch": 0.1523809523809524,
      "grad_norm": 0.7984667420387268,
      "learning_rate": 1.9968176841806687e-05,
      "loss": 0.9479,
      "mean_token_accuracy": 0.7837442796677351,
      "num_tokens": 585927.0,
      "step": 60
    },
    {
      "entropy": 1.635760549083352,
      "epoch": 0.20317460317460317,
      "grad_norm": 0.7746185660362244,
      "learning_rate": 1.9866148103359362e-05,
      "loss": 0.6905,
      "mean_token_accuracy": 0.8289982877671719,
      "num_tokens": 781164.0,
      "step": 80
    },
    {
      "entropy": 1.4979693002998828,
      "epoch": 0.25396825396825395,
      "grad_norm": 0.7985681891441345,
      "learning_rate": 1.9694545073405348e-05,
      "loss": 0.5682,
      "mean_token_accuracy": 0.8550767470151186,
      "num_tokens": 976497.0,
      "step": 100
    },
    {
      "entropy": 1.3932939764112233,
      "epoch": 0.3047619047619048,
      "grad_norm": 0.8087366223335266,
      "learning_rate": 1.94545778654666e-05,
      "loss": 0.5126,
      "mean_token_accuracy": 0.8678769677877426,
      "num_tokens": 1171938.0,
      "step": 120
    },
    {
      "entropy": 1.3521239839494228,
      "epoch": 0.35555555555555557,
      "grad_norm": 1.0716849565505981,
      "learning_rate": 1.9147938684880213e-05,
      "loss": 0.4679,
      "mean_token_accuracy": 0.8757014229893685,
      "num_tokens": 1367065.0,
      "step": 140
    },
    {
      "entropy": 1.276645314320922,
      "epoch": 0.40634920634920635,
      "grad_norm": 1.04561448097229,
      "learning_rate": 1.8776789895672557e-05,
      "loss": 0.4319,
      "mean_token_accuracy": 0.8847867721691728,
      "num_tokens": 1562066.0,
      "step": 160
    },
    {
      "entropy": 1.237958626449108,
      "epoch": 0.45714285714285713,
      "grad_norm": 0.78230220079422,
      "learning_rate": 1.8343748771959346e-05,
      "loss": 0.4182,
      "mean_token_accuracy": 0.8867125200107694,
      "num_tokens": 1756828.0,
      "step": 180
    },
    {
      "entropy": 1.2167125567793846,
      "epoch": 0.5079365079365079,
      "grad_norm": 1.3130111694335938,
      "learning_rate": 1.785186904140207e-05,
      "loss": 0.384,
      "mean_token_accuracy": 0.8928723320364952,
      "num_tokens": 1952018.0,
      "step": 200
    },
    {
      "entropy": 1.1336687998846173,
      "epoch": 0.5587301587301587,
      "grad_norm": 1.0740300416946411,
      "learning_rate": 1.7304619350872992e-05,
      "loss": 0.355,
      "mean_token_accuracy": 0.8990522997453809,
      "num_tokens": 2146747.0,
      "step": 220
    },
    {
      "entropy": 1.0902384620159864,
      "epoch": 0.6095238095238096,
      "grad_norm": 1.2321722507476807,
      "learning_rate": 1.6705858806184933e-05,
      "loss": 0.3421,
      "mean_token_accuracy": 0.8982272742316126,
      "num_tokens": 2342526.0,
      "step": 240
    },
    {
      "entropy": 1.072350986301899,
      "epoch": 0.6603174603174603,
      "grad_norm": 1.2107694149017334,
      "learning_rate": 1.605980975837524e-05,
      "loss": 0.3293,
      "mean_token_accuracy": 0.9014129877090454,
      "num_tokens": 2537429.0,
      "step": 260
    },
    {
      "entropy": 1.0297158515080809,
      "epoch": 0.7111111111111111,
      "grad_norm": 1.3480980396270752,
      "learning_rate": 1.5371028028450152e-05,
      "loss": 0.3115,
      "mean_token_accuracy": 0.9040320562198758,
      "num_tokens": 2732125.0,
      "step": 280
    },
    {
      "entropy": 1.0070750426501036,
      "epoch": 0.7619047619047619,
      "grad_norm": 1.2340924739837646,
      "learning_rate": 1.4644370780559265e-05,
      "loss": 0.297,
      "mean_token_accuracy": 0.906495463848114,
      "num_tokens": 2926977.0,
      "step": 300
    },
    {
      "epoch": 0.7619047619047619,
      "eval_entropy": 0.9912602304560798,
      "eval_loss": 0.28733107447624207,
      "eval_mean_token_accuracy": 0.9105286079645157,
      "eval_num_tokens": 2926977.0,
      "eval_runtime": 254.0073,
      "eval_samples_per_second": 2.756,
      "eval_steps_per_second": 2.756,
      "step": 300
    },
    {
      "entropy": 0.9841975728049874,
      "epoch": 0.8126984126984127,
      "grad_norm": 1.2695492506027222,
      "learning_rate": 1.3884962270152693e-05,
      "loss": 0.2871,
      "mean_token_accuracy": 0.9104899806901813,
      "num_tokens": 3122184.0,
      "step": 320
    },
    {
      "entropy": 0.9748819842934608,
      "epoch": 0.8634920634920635,
      "grad_norm": 1.4477434158325195,
      "learning_rate": 1.3098157708658657e-05,
      "loss": 0.2584,
      "mean_token_accuracy": 0.9179832600057125,
      "num_tokens": 3317637.0,
      "step": 340
    },
    {
      "entropy": 0.9418716534972191,
      "epoch": 0.9142857142857143,
      "grad_norm": 1.3718384504318237,
      "learning_rate": 1.2289505499501341e-05,
      "loss": 0.2509,
      "mean_token_accuracy": 0.9180495567619801,
      "num_tokens": 3513113.0,
      "step": 360
    },
    {
      "entropy": 0.946486484631896,
      "epoch": 0.9650793650793651,
      "grad_norm": 1.3464258909225464,
      "learning_rate": 1.1464708111763723e-05,
      "loss": 0.2477,
      "mean_token_accuracy": 0.9207174494862557,
      "num_tokens": 3707967.0,
      "step": 380
    },
    {
      "entropy": 0.9170130017814757,
      "epoch": 1.0152380952380953,
      "grad_norm": 1.5472830533981323,
      "learning_rate": 1.0629581867407241e-05,
      "loss": 0.2329,
      "mean_token_accuracy": 0.9222704160817062,
      "num_tokens": 3900703.0,
      "step": 400
    },
    {
      "entropy": 0.927550189383328,
      "epoch": 1.066031746031746,
      "grad_norm": 1.8993698358535767,
      "learning_rate": 9.790015925621588e-06,
      "loss": 0.2196,
      "mean_token_accuracy": 0.9251768393442035,
      "num_tokens": 4096441.0,
      "step": 420
    },
    {
      "entropy": 0.9159683456644416,
      "epoch": 1.116825396825397,
      "grad_norm": 1.489464521408081,
      "learning_rate": 8.951930753539521e-06,
      "loss": 0.2162,
      "mean_token_accuracy": 0.9265725754201413,
      "num_tokens": 4290838.0,
      "step": 440
    },
    {
      "entropy": 0.8982374468818307,
      "epoch": 1.1676190476190476,
      "grad_norm": 1.4005104303359985,
      "learning_rate": 8.121236376173745e-06,
      "loss": 0.205,
      "mean_token_accuracy": 0.9285883469507098,
      "num_tokens": 4485804.0,
      "step": 460
    },
    {
      "entropy": 0.8816795371472835,
      "epoch": 1.2184126984126984,
      "grad_norm": 1.45193612575531,
      "learning_rate": 7.303790699989714e-06,
      "loss": 0.2085,
      "mean_token_accuracy": 0.928891065903008,
      "num_tokens": 4681086.0,
      "step": 480
    },
    {
      "entropy": 0.8763484323397279,
      "epoch": 1.2692063492063492,
      "grad_norm": 1.3778793811798096,
      "learning_rate": 6.505358204009018e-06,
      "loss": 0.1982,
      "mean_token_accuracy": 0.9306870764121413,
      "num_tokens": 4876429.0,
      "step": 500
    },
    {
      "entropy": 0.8736646875739098,
      "epoch": 1.32,
      "grad_norm": 1.3831270933151245,
      "learning_rate": 5.731569289746193e-06,
      "loss": 0.1897,
      "mean_token_accuracy": 0.9305133303627372,
      "num_tokens": 5071394.0,
      "step": 520
    },
    {
      "entropy": 0.868809700757265,
      "epoch": 1.370793650793651,
      "grad_norm": 1.8496570587158203,
      "learning_rate": 4.98788057663585e-06,
      "loss": 0.1931,
      "mean_token_accuracy": 0.9316842250525952,
      "num_tokens": 5265948.0,
      "step": 540
    },
    {
      "entropy": 0.8576494121924043,
      "epoch": 1.4215873015873015,
      "grad_norm": 1.5876587629318237,
      "learning_rate": 4.279536422939606e-06,
      "loss": 0.1873,
      "mean_token_accuracy": 0.9321241827681661,
      "num_tokens": 5461394.0,
      "step": 560
    },
    {
      "entropy": 0.854162979312241,
      "epoch": 1.4723809523809523,
      "grad_norm": 1.3804458379745483,
      "learning_rate": 3.6115319434803897e-06,
      "loss": 0.1901,
      "mean_token_accuracy": 0.9312955033034086,
      "num_tokens": 5656093.0,
      "step": 580
    },
    {
      "entropy": 0.8487729975953698,
      "epoch": 1.5231746031746032,
      "grad_norm": 2.6718533039093018,
      "learning_rate": 2.9885777849964016e-06,
      "loss": 0.1862,
      "mean_token_accuracy": 0.9315127771347761,
      "num_tokens": 5851549.0,
      "step": 600
    },
    {
      "epoch": 1.5231746031746032,
      "eval_entropy": 0.844009120634624,
      "eval_loss": 0.19104033708572388,
      "eval_mean_token_accuracy": 0.9312942716905049,
      "eval_num_tokens": 5851549.0,
      "eval_runtime": 253.7619,
      "eval_samples_per_second": 2.758,
      "eval_steps_per_second": 2.758,
      "step": 600
    }
  ],
  "logging_steps": 20,
  "max_steps": 788,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 300,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.847129416270643e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}