{
  "best_metric": 3.254521369934082,
  "best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained_last/en_wiki_mlm_30/checkpoint-84000",
  "epoch": 47.53820033955857,
  "eval_steps": 2000,
  "global_step": 84000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.1318619128466327,
      "eval_loss": 7.872045516967773,
      "eval_runtime": 3.9374,
      "eval_samples_per_second": 1252.08,
      "eval_steps_per_second": 78.477,
      "step": 2000
    },
    {
      "epoch": 2.2637238256932655,
      "grad_norm": 1.0014455318450928,
      "learning_rate": 1e-05,
      "loss": 7.9236,
      "step": 4000
    },
    {
      "epoch": 2.2637238256932655,
      "eval_loss": 7.107393264770508,
      "eval_runtime": 3.9376,
      "eval_samples_per_second": 1252.036,
      "eval_steps_per_second": 78.474,
      "step": 4000
    },
    {
      "epoch": 3.395585738539898,
      "eval_loss": 7.030406951904297,
      "eval_runtime": 3.9404,
      "eval_samples_per_second": 1251.146,
      "eval_steps_per_second": 78.419,
      "step": 6000
    },
    {
      "epoch": 4.527447651386531,
      "grad_norm": 1.1162114143371582,
      "learning_rate": 2e-05,
      "loss": 7.0257,
      "step": 8000
    },
    {
      "epoch": 4.527447651386531,
      "eval_loss": 6.953151226043701,
      "eval_runtime": 4.0906,
      "eval_samples_per_second": 1205.208,
      "eval_steps_per_second": 75.539,
      "step": 8000
    },
    {
      "epoch": 5.659309564233164,
      "eval_loss": 6.87604284286499,
      "eval_runtime": 3.912,
      "eval_samples_per_second": 1260.235,
      "eval_steps_per_second": 78.988,
      "step": 10000
    },
    {
      "epoch": 6.791171477079796,
      "grad_norm": 1.346594214439392,
      "learning_rate": 3e-05,
      "loss": 6.8811,
      "step": 12000
    },
    {
      "epoch": 6.791171477079796,
      "eval_loss": 6.811014175415039,
      "eval_runtime": 3.951,
      "eval_samples_per_second": 1247.773,
      "eval_steps_per_second": 78.207,
      "step": 12000
    },
    {
      "epoch": 7.923033389926429,
      "eval_loss": 6.731590747833252,
      "eval_runtime": 3.8635,
      "eval_samples_per_second": 1276.05,
      "eval_steps_per_second": 79.98,
      "step": 14000
    },
    {
      "epoch": 9.054895302773062,
      "grad_norm": 1.458003044128418,
      "learning_rate": 4e-05,
      "loss": 6.7587,
      "step": 16000
    },
    {
      "epoch": 9.054895302773062,
      "eval_loss": 6.689206600189209,
      "eval_runtime": 3.8397,
      "eval_samples_per_second": 1283.96,
      "eval_steps_per_second": 80.475,
      "step": 16000
    },
    {
      "epoch": 10.186757215619695,
      "eval_loss": 6.650099754333496,
      "eval_runtime": 3.9002,
      "eval_samples_per_second": 1264.041,
      "eval_steps_per_second": 79.227,
      "step": 18000
    },
    {
      "epoch": 11.318619128466327,
      "grad_norm": 1.5871217250823975,
      "learning_rate": 5e-05,
      "loss": 6.6566,
      "step": 20000
    },
    {
      "epoch": 11.318619128466327,
      "eval_loss": 6.595088958740234,
      "eval_runtime": 3.9235,
      "eval_samples_per_second": 1256.533,
      "eval_steps_per_second": 78.756,
      "step": 20000
    },
    {
      "epoch": 12.45048104131296,
      "eval_loss": 6.525483131408691,
      "eval_runtime": 3.8651,
      "eval_samples_per_second": 1275.512,
      "eval_steps_per_second": 79.946,
      "step": 22000
    },
    {
      "epoch": 13.582342954159593,
      "grad_norm": 2.502711296081543,
      "learning_rate": 6e-05,
      "loss": 6.546,
      "step": 24000
    },
    {
      "epoch": 13.582342954159593,
      "eval_loss": 6.440639972686768,
      "eval_runtime": 3.8611,
      "eval_samples_per_second": 1276.85,
      "eval_steps_per_second": 80.03,
      "step": 24000
    },
    {
      "epoch": 14.714204867006226,
      "eval_loss": 6.316535472869873,
      "eval_runtime": 3.8871,
      "eval_samples_per_second": 1268.291,
      "eval_steps_per_second": 79.493,
      "step": 26000
    },
    {
      "epoch": 15.846066779852858,
      "grad_norm": 4.2022857666015625,
      "learning_rate": 7e-05,
      "loss": 6.3494,
      "step": 28000
    },
    {
      "epoch": 15.846066779852858,
      "eval_loss": 6.149945259094238,
      "eval_runtime": 3.9074,
      "eval_samples_per_second": 1261.705,
      "eval_steps_per_second": 79.081,
      "step": 28000
    },
    {
      "epoch": 16.97792869269949,
      "eval_loss": 5.94103479385376,
      "eval_runtime": 3.8837,
      "eval_samples_per_second": 1269.393,
      "eval_steps_per_second": 79.562,
      "step": 30000
    },
    {
      "epoch": 18.109790605546124,
      "grad_norm": 4.1835408210754395,
      "learning_rate": 8e-05,
      "loss": 6.0156,
      "step": 32000
    },
    {
      "epoch": 18.109790605546124,
      "eval_loss": 5.63771390914917,
      "eval_runtime": 3.9903,
      "eval_samples_per_second": 1235.504,
      "eval_steps_per_second": 77.438,
      "step": 32000
    },
    {
      "epoch": 19.241652518392755,
      "eval_loss": 5.117396354675293,
      "eval_runtime": 3.9867,
      "eval_samples_per_second": 1236.617,
      "eval_steps_per_second": 77.508,
      "step": 34000
    },
    {
      "epoch": 20.37351443123939,
      "grad_norm": 3.6266136169433594,
      "learning_rate": 9e-05,
      "loss": 5.2999,
      "step": 36000
    },
    {
      "epoch": 20.37351443123939,
      "eval_loss": 4.855127811431885,
      "eval_runtime": 3.8488,
      "eval_samples_per_second": 1280.914,
      "eval_steps_per_second": 80.284,
      "step": 36000
    },
    {
      "epoch": 21.50537634408602,
      "eval_loss": 4.6650190353393555,
      "eval_runtime": 3.8768,
      "eval_samples_per_second": 1271.655,
      "eval_steps_per_second": 79.704,
      "step": 38000
    },
    {
      "epoch": 22.637238256932655,
      "grad_norm": 3.870995044708252,
      "learning_rate": 0.0001,
      "loss": 4.7633,
      "step": 40000
    },
    {
      "epoch": 22.637238256932655,
      "eval_loss": 4.496387481689453,
      "eval_runtime": 3.9179,
      "eval_samples_per_second": 1258.325,
      "eval_steps_per_second": 78.869,
      "step": 40000
    },
    {
      "epoch": 23.769100169779286,
      "eval_loss": 4.324868679046631,
      "eval_runtime": 3.916,
      "eval_samples_per_second": 1258.94,
      "eval_steps_per_second": 78.907,
      "step": 42000
    },
    {
      "epoch": 24.90096208262592,
      "grad_norm": 4.243465423583984,
      "learning_rate": 9.333333333333334e-05,
      "loss": 4.4471,
      "step": 44000
    },
    {
      "epoch": 24.90096208262592,
      "eval_loss": 4.211695194244385,
      "eval_runtime": 3.9298,
      "eval_samples_per_second": 1254.526,
      "eval_steps_per_second": 78.631,
      "step": 44000
    },
    {
      "epoch": 26.03282399547255,
      "eval_loss": 4.076672077178955,
      "eval_runtime": 3.9702,
      "eval_samples_per_second": 1241.746,
      "eval_steps_per_second": 77.829,
      "step": 46000
    },
    {
      "epoch": 27.164685908319186,
      "grad_norm": 4.304368495941162,
      "learning_rate": 8.666666666666667e-05,
      "loss": 4.1884,
      "step": 48000
    },
    {
      "epoch": 27.164685908319186,
      "eval_loss": 3.9929893016815186,
      "eval_runtime": 3.9338,
      "eval_samples_per_second": 1253.246,
      "eval_steps_per_second": 78.55,
      "step": 48000
    },
    {
      "epoch": 28.296547821165817,
      "eval_loss": 3.9029903411865234,
      "eval_runtime": 3.9273,
      "eval_samples_per_second": 1255.305,
      "eval_steps_per_second": 78.679,
      "step": 50000
    },
    {
      "epoch": 29.42840973401245,
      "grad_norm": 4.121150970458984,
      "learning_rate": 8e-05,
      "loss": 3.9939,
      "step": 52000
    },
    {
      "epoch": 29.42840973401245,
      "eval_loss": 3.8125526905059814,
      "eval_runtime": 3.9176,
      "eval_samples_per_second": 1258.429,
      "eval_steps_per_second": 78.875,
      "step": 52000
    },
    {
      "epoch": 30.560271646859082,
      "eval_loss": 3.7701168060302734,
      "eval_runtime": 3.9559,
      "eval_samples_per_second": 1246.255,
      "eval_steps_per_second": 78.112,
      "step": 54000
    },
    {
      "epoch": 31.692133559705717,
      "grad_norm": 4.4801177978515625,
      "learning_rate": 7.333333333333333e-05,
      "loss": 3.8479,
      "step": 56000
    },
    {
      "epoch": 31.692133559705717,
      "eval_loss": 3.6775288581848145,
      "eval_runtime": 3.8905,
      "eval_samples_per_second": 1267.203,
      "eval_steps_per_second": 79.425,
      "step": 56000
    },
    {
      "epoch": 32.82399547255235,
      "eval_loss": 3.643172025680542,
      "eval_runtime": 3.9657,
      "eval_samples_per_second": 1243.163,
      "eval_steps_per_second": 77.918,
      "step": 58000
    },
    {
      "epoch": 33.95585738539898,
      "grad_norm": 4.373381614685059,
      "learning_rate": 6.666666666666667e-05,
      "loss": 3.7265,
      "step": 60000
    },
    {
      "epoch": 33.95585738539898,
      "eval_loss": 3.5951223373413086,
      "eval_runtime": 3.9071,
      "eval_samples_per_second": 1261.815,
      "eval_steps_per_second": 79.087,
      "step": 60000
    },
    {
      "epoch": 35.08771929824562,
      "eval_loss": 3.5469629764556885,
      "eval_runtime": 3.9205,
      "eval_samples_per_second": 1257.484,
      "eval_steps_per_second": 78.816,
      "step": 62000
    },
    {
      "epoch": 36.21958121109225,
      "grad_norm": 3.978059768676758,
      "learning_rate": 6e-05,
      "loss": 3.6305,
      "step": 64000
    },
    {
      "epoch": 36.21958121109225,
      "eval_loss": 3.5206313133239746,
      "eval_runtime": 3.9167,
      "eval_samples_per_second": 1258.71,
      "eval_steps_per_second": 78.893,
      "step": 64000
    },
    {
      "epoch": 37.35144312393888,
      "eval_loss": 3.494868278503418,
      "eval_runtime": 3.9115,
      "eval_samples_per_second": 1260.397,
      "eval_steps_per_second": 78.998,
      "step": 66000
    },
    {
      "epoch": 38.48330503678551,
      "grad_norm": 4.399171352386475,
      "learning_rate": 5.333333333333333e-05,
      "loss": 3.5483,
      "step": 68000
    },
    {
      "epoch": 38.48330503678551,
      "eval_loss": 3.476797103881836,
      "eval_runtime": 3.9327,
      "eval_samples_per_second": 1253.586,
      "eval_steps_per_second": 78.572,
      "step": 68000
    },
    {
      "epoch": 39.61516694963215,
      "eval_loss": 3.422701358795166,
      "eval_runtime": 3.878,
      "eval_samples_per_second": 1271.27,
      "eval_steps_per_second": 79.68,
      "step": 70000
    },
    {
      "epoch": 40.74702886247878,
      "grad_norm": 4.03302001953125,
      "learning_rate": 4.666666666666667e-05,
      "loss": 3.4798,
      "step": 72000
    },
    {
      "epoch": 40.74702886247878,
      "eval_loss": 3.373471975326538,
      "eval_runtime": 3.9535,
      "eval_samples_per_second": 1247.004,
      "eval_steps_per_second": 78.159,
      "step": 72000
    },
    {
      "epoch": 41.87889077532541,
      "eval_loss": 3.3893837928771973,
      "eval_runtime": 3.8713,
      "eval_samples_per_second": 1273.464,
      "eval_steps_per_second": 79.818,
      "step": 74000
    },
    {
      "epoch": 43.01075268817204,
      "grad_norm": 4.291697978973389,
      "learning_rate": 4e-05,
      "loss": 3.4256,
      "step": 76000
    },
    {
      "epoch": 43.01075268817204,
      "eval_loss": 3.354308843612671,
      "eval_runtime": 3.9005,
      "eval_samples_per_second": 1263.926,
      "eval_steps_per_second": 79.22,
      "step": 76000
    },
    {
      "epoch": 44.14261460101868,
      "eval_loss": 3.3211026191711426,
      "eval_runtime": 3.9503,
      "eval_samples_per_second": 1248.001,
      "eval_steps_per_second": 78.222,
      "step": 78000
    },
    {
      "epoch": 45.27447651386531,
      "grad_norm": 4.428945541381836,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 3.3707,
      "step": 80000
    },
    {
      "epoch": 45.27447651386531,
      "eval_loss": 3.3156309127807617,
      "eval_runtime": 3.9036,
      "eval_samples_per_second": 1262.941,
      "eval_steps_per_second": 79.158,
      "step": 80000
    },
    {
      "epoch": 46.40633842671194,
      "eval_loss": 3.289860963821411,
      "eval_runtime": 3.8723,
      "eval_samples_per_second": 1273.15,
      "eval_steps_per_second": 79.798,
      "step": 82000
    },
    {
      "epoch": 47.53820033955857,
      "grad_norm": 4.017796516418457,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 3.3325,
      "step": 84000
    },
    {
      "epoch": 47.53820033955857,
      "eval_loss": 3.254521369934082,
      "eval_runtime": 3.8949,
      "eval_samples_per_second": 1265.772,
      "eval_steps_per_second": 79.335,
      "step": 84000
    }
  ],
  "logging_steps": 4000,
  "max_steps": 100000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 57,
  "save_steps": 4000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.6199932172632064e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}