{
  "best_metric": 0.9703998565673828,
  "best_model_checkpoint": "data/Gemma-2-2B_task-1_120-samples_config-2_auto/checkpoint-33",
  "epoch": 12.909090909090908,
  "eval_steps": 500,
  "global_step": 71,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 1.7128347158432007,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.5946,
      "step": 1
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 1.9887312650680542,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.6593,
      "step": 2
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 1.6978522539138794,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.5728,
      "step": 4
    },
    {
      "epoch": 0.9090909090909091,
      "eval_loss": 1.6555770635604858,
      "eval_runtime": 2.1208,
      "eval_samples_per_second": 11.316,
      "eval_steps_per_second": 11.316,
      "step": 5
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 1.8235312700271606,
      "learning_rate": 2.4e-05,
      "loss": 1.7112,
      "step": 6
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 1.5804287195205688,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 1.4809,
      "step": 8
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 1.7760125398635864,
      "learning_rate": 4e-05,
      "loss": 1.5544,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.310595154762268,
      "eval_runtime": 2.148,
      "eval_samples_per_second": 11.173,
      "eval_steps_per_second": 11.173,
      "step": 11
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 1.2979093790054321,
      "learning_rate": 4.8e-05,
      "loss": 1.2278,
      "step": 12
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 1.1246834993362427,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 1.2293,
      "step": 14
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.9627875685691833,
      "learning_rate": 6.400000000000001e-05,
      "loss": 1.1135,
      "step": 16
    },
    {
      "epoch": 2.909090909090909,
      "eval_loss": 1.1137615442276,
      "eval_runtime": 2.1124,
      "eval_samples_per_second": 11.362,
      "eval_steps_per_second": 11.362,
      "step": 16
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 0.8638038039207458,
      "learning_rate": 7.2e-05,
      "loss": 0.9895,
      "step": 18
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.8232435584068298,
      "learning_rate": 8e-05,
      "loss": 1.029,
      "step": 20
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6949414014816284,
      "learning_rate": 8.800000000000001e-05,
      "loss": 0.9739,
      "step": 22
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.0346862077713013,
      "eval_runtime": 2.1116,
      "eval_samples_per_second": 11.366,
      "eval_steps_per_second": 11.366,
      "step": 22
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 0.7141179442405701,
      "learning_rate": 9.6e-05,
      "loss": 0.9627,
      "step": 24
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 0.7031455039978027,
      "learning_rate": 9.999512620046522e-05,
      "loss": 0.8888,
      "step": 26
    },
    {
      "epoch": 4.909090909090909,
      "eval_loss": 0.9918524622917175,
      "eval_runtime": 2.111,
      "eval_samples_per_second": 11.369,
      "eval_steps_per_second": 11.369,
      "step": 27
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 0.6778797507286072,
      "learning_rate": 9.995614150494293e-05,
      "loss": 0.8228,
      "step": 28
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 0.65517258644104,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.7832,
      "step": 30
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 0.6927575469017029,
      "learning_rate": 9.976136999909156e-05,
      "loss": 0.8107,
      "step": 32
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.9703998565673828,
      "eval_runtime": 2.1137,
      "eval_samples_per_second": 11.355,
      "eval_steps_per_second": 11.355,
      "step": 33
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 0.7274169325828552,
      "learning_rate": 9.96057350657239e-05,
      "loss": 0.7307,
      "step": 34
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 0.6062109470367432,
      "learning_rate": 9.941141907232765e-05,
      "loss": 0.7511,
      "step": 36
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 0.7088773250579834,
      "learning_rate": 9.917857354066931e-05,
      "loss": 0.6431,
      "step": 38
    },
    {
      "epoch": 6.909090909090909,
      "eval_loss": 0.9725914001464844,
      "eval_runtime": 2.0963,
      "eval_samples_per_second": 11.449,
      "eval_steps_per_second": 11.449,
      "step": 38
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 0.727410078048706,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.672,
      "step": 40
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 0.7320867776870728,
      "learning_rate": 9.859805002892732e-05,
      "loss": 0.5495,
      "step": 42
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.8117280602455139,
      "learning_rate": 9.825082472361557e-05,
      "loss": 0.5564,
      "step": 44
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.0095218420028687,
      "eval_runtime": 2.1109,
      "eval_samples_per_second": 11.369,
      "eval_steps_per_second": 11.369,
      "step": 44
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 0.7933902740478516,
      "learning_rate": 9.786597487660337e-05,
      "loss": 0.4522,
      "step": 46
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 0.950472891330719,
      "learning_rate": 9.744380058222483e-05,
      "loss": 0.4886,
      "step": 48
    },
    {
      "epoch": 8.909090909090908,
      "eval_loss": 1.1149295568466187,
      "eval_runtime": 2.1135,
      "eval_samples_per_second": 11.355,
      "eval_steps_per_second": 11.355,
      "step": 49
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 1.0439478158950806,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.4518,
      "step": 50
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 1.0246175527572632,
      "learning_rate": 9.648882429441257e-05,
      "loss": 0.3422,
      "step": 52
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 1.0976358652114868,
      "learning_rate": 9.595676696276172e-05,
      "loss": 0.3412,
      "step": 54
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.327033519744873,
      "eval_runtime": 2.1434,
      "eval_samples_per_second": 11.197,
      "eval_steps_per_second": 11.197,
      "step": 55
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 1.0725340843200684,
      "learning_rate": 9.538887392664544e-05,
      "loss": 0.2932,
      "step": 56
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 1.202645182609558,
      "learning_rate": 9.478558801197065e-05,
      "loss": 0.245,
      "step": 58
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 1.6887779235839844,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.2372,
      "step": 60
    },
    {
      "epoch": 10.909090909090908,
      "eval_loss": 1.5941375494003296,
      "eval_runtime": 2.1205,
      "eval_samples_per_second": 11.318,
      "eval_steps_per_second": 11.318,
      "step": 60
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 0.9777700901031494,
      "learning_rate": 9.347474647526095e-05,
      "loss": 0.1623,
      "step": 62
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 1.3054704666137695,
      "learning_rate": 9.276821300802534e-05,
      "loss": 0.1382,
      "step": 64
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.2604758739471436,
      "learning_rate": 9.202833017478422e-05,
      "loss": 0.1438,
      "step": 66
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.8735154867172241,
      "eval_runtime": 2.1146,
      "eval_samples_per_second": 11.349,
      "eval_steps_per_second": 11.349,
      "step": 66
    },
    {
      "epoch": 12.363636363636363,
      "grad_norm": 1.1091744899749756,
      "learning_rate": 9.125567491391476e-05,
      "loss": 0.0913,
      "step": 68
    },
    {
      "epoch": 12.727272727272727,
      "grad_norm": 1.0739688873291016,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0894,
      "step": 70
    },
    {
      "epoch": 12.909090909090908,
      "eval_loss": 2.111311674118042,
      "eval_runtime": 2.17,
      "eval_samples_per_second": 11.06,
      "eval_steps_per_second": 11.06,
      "step": 71
    },
    {
      "epoch": 12.909090909090908,
      "step": 71,
      "total_flos": 4963046722109440.0,
      "train_loss": 0.7389588935274474,
      "train_runtime": 498.1206,
      "train_samples_per_second": 8.833,
      "train_steps_per_second": 0.502
    }
  ],
  "logging_steps": 2,
  "max_steps": 250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 25,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 7,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4963046722109440.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}