{
  "best_metric": 4.496387481689453,
  "best_model_checkpoint": "/home/p318482/babyLM_controlled/models_trained_last/en_wiki_mlm_30/checkpoint-40000",
  "epoch": 22.637238256932655,
  "eval_steps": 2000,
  "global_step": 40000,
  "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
    }
  ],
  "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": 1.2476180458635264e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}