{
  "best_metric": 1.1336115598678589,
  "best_model_checkpoint": "data/Llama-31-8B_task-3_120-samples_config-2_full/checkpoint-66",
  "epoch": 18.90909090909091,
  "eval_steps": 500,
  "global_step": 104,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 0.5719963312149048,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.7319,
      "step": 1
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 0.5006797313690186,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.672,
      "step": 2
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 0.44579824805259705,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.7269,
      "step": 4
    },
    {
      "epoch": 0.9090909090909091,
      "eval_loss": 1.6900886297225952,
      "eval_runtime": 20.9835,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 5
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 0.4690605700016022,
      "learning_rate": 2.4e-05,
      "loss": 1.7107,
      "step": 6
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 0.4857431948184967,
      "learning_rate": 3.2000000000000005e-05,
      "loss": 1.6594,
      "step": 8
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.36418014764785767,
      "learning_rate": 4e-05,
      "loss": 1.6699,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.5938736200332642,
      "eval_runtime": 20.9811,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 11
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 0.2819691300392151,
      "learning_rate": 4.8e-05,
      "loss": 1.5688,
      "step": 12
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.3870878517627716,
      "learning_rate": 5.6000000000000006e-05,
      "loss": 1.5426,
      "step": 14
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.2570940852165222,
      "learning_rate": 6.400000000000001e-05,
      "loss": 1.5273,
      "step": 16
    },
    {
      "epoch": 2.909090909090909,
      "eval_loss": 1.500514030456543,
      "eval_runtime": 20.9853,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 16
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 0.24655576050281525,
      "learning_rate": 7.2e-05,
      "loss": 1.4843,
      "step": 18
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.239831805229187,
      "learning_rate": 8e-05,
      "loss": 1.4156,
      "step": 20
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.20840594172477722,
      "learning_rate": 8.800000000000001e-05,
      "loss": 1.3953,
      "step": 22
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.3804607391357422,
      "eval_runtime": 20.9874,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 22
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 0.26833054423332214,
      "learning_rate": 9.6e-05,
      "loss": 1.3418,
      "step": 24
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 0.32587364315986633,
      "learning_rate": 9.999512620046522e-05,
      "loss": 1.2642,
      "step": 26
    },
    {
      "epoch": 4.909090909090909,
      "eval_loss": 1.2620800733566284,
      "eval_runtime": 20.985,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 27
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 0.4004269242286682,
      "learning_rate": 9.995614150494293e-05,
      "loss": 1.2394,
      "step": 28
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 0.24827851355075836,
      "learning_rate": 9.987820251299122e-05,
      "loss": 1.187,
      "step": 30
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 0.27338162064552307,
      "learning_rate": 9.976136999909156e-05,
      "loss": 1.1571,
      "step": 32
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.1986058950424194,
      "eval_runtime": 20.9798,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 33
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 0.17052970826625824,
      "learning_rate": 9.96057350657239e-05,
      "loss": 1.146,
      "step": 34
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 0.18164239823818207,
      "learning_rate": 9.941141907232765e-05,
      "loss": 1.1216,
      "step": 36
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 0.15134599804878235,
      "learning_rate": 9.917857354066931e-05,
      "loss": 1.1345,
      "step": 38
    },
    {
      "epoch": 6.909090909090909,
      "eval_loss": 1.1787580251693726,
      "eval_runtime": 20.9891,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 38
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 0.1562015265226364,
      "learning_rate": 9.890738003669029e-05,
      "loss": 1.1307,
      "step": 40
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 0.14140599966049194,
      "learning_rate": 9.859805002892732e-05,
      "loss": 1.0804,
      "step": 42
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.1359817236661911,
      "learning_rate": 9.825082472361557e-05,
      "loss": 1.1145,
      "step": 44
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.1605745553970337,
      "eval_runtime": 20.9782,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 44
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 0.1520766019821167,
      "learning_rate": 9.786597487660337e-05,
      "loss": 1.0777,
      "step": 46
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 0.16712282598018646,
      "learning_rate": 9.744380058222483e-05,
      "loss": 1.0654,
      "step": 48
    },
    {
      "epoch": 8.909090909090908,
      "eval_loss": 1.1504966020584106,
      "eval_runtime": 20.977,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 49
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 0.14631617069244385,
      "learning_rate": 9.698463103929542e-05,
      "loss": 1.0926,
      "step": 50
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 0.15264888107776642,
      "learning_rate": 9.648882429441257e-05,
      "loss": 1.0386,
      "step": 52
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 0.14912693202495575,
      "learning_rate": 9.595676696276172e-05,
      "loss": 1.0821,
      "step": 54
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.1416734457015991,
      "eval_runtime": 20.9871,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 55
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 0.16033270955085754,
      "learning_rate": 9.538887392664544e-05,
      "loss": 1.0224,
      "step": 56
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 0.18686003983020782,
      "learning_rate": 9.478558801197065e-05,
      "loss": 1.0213,
      "step": 58
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 0.16541936993598938,
      "learning_rate": 9.414737964294636e-05,
      "loss": 1.0346,
      "step": 60
    },
    {
      "epoch": 10.909090909090908,
      "eval_loss": 1.1381394863128662,
      "eval_runtime": 20.9809,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 60
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 0.1957688182592392,
      "learning_rate": 9.347474647526095e-05,
      "loss": 1.0144,
      "step": 62
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 0.18131205439567566,
      "learning_rate": 9.276821300802534e-05,
      "loss": 1.0238,
      "step": 64
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.22676895558834076,
      "learning_rate": 9.202833017478422e-05,
      "loss": 1.0027,
      "step": 66
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.1336115598678589,
      "eval_runtime": 20.9843,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 66
    },
    {
      "epoch": 12.363636363636363,
      "grad_norm": 0.20228485763072968,
      "learning_rate": 9.125567491391476e-05,
      "loss": 0.9864,
      "step": 68
    },
    {
      "epoch": 12.727272727272727,
      "grad_norm": 0.22149056196212769,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.9693,
      "step": 70
    },
    {
      "epoch": 12.909090909090908,
      "eval_loss": 1.1349502801895142,
      "eval_runtime": 20.9835,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 71
    },
    {
      "epoch": 13.090909090909092,
      "grad_norm": 0.24571332335472107,
      "learning_rate": 8.961448216775954e-05,
      "loss": 0.9601,
      "step": 72
    },
    {
      "epoch": 13.454545454545455,
      "grad_norm": 0.19281752407550812,
      "learning_rate": 8.874722443520899e-05,
      "loss": 1.0015,
      "step": 74
    },
    {
      "epoch": 13.818181818181818,
      "grad_norm": 0.23641857504844666,
      "learning_rate": 8.784975278258783e-05,
      "loss": 0.9594,
      "step": 76
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.1413918733596802,
      "eval_runtime": 20.9878,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 77
    },
    {
      "epoch": 14.181818181818182,
      "grad_norm": 0.2770022451877594,
      "learning_rate": 8.692276703129421e-05,
      "loss": 0.9033,
      "step": 78
    },
    {
      "epoch": 14.545454545454545,
      "grad_norm": 0.2724754214286804,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.9652,
      "step": 80
    },
    {
      "epoch": 14.909090909090908,
      "grad_norm": 0.34605658054351807,
      "learning_rate": 8.498316702566828e-05,
      "loss": 0.8512,
      "step": 82
    },
    {
      "epoch": 14.909090909090908,
      "eval_loss": 1.1548329591751099,
      "eval_runtime": 20.9813,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 82
    },
    {
      "epoch": 15.272727272727273,
      "grad_norm": 0.38165947794914246,
      "learning_rate": 8.397206521307584e-05,
      "loss": 0.9087,
      "step": 84
    },
    {
      "epoch": 15.636363636363637,
      "grad_norm": 0.4297967255115509,
      "learning_rate": 8.293447300593402e-05,
      "loss": 0.8638,
      "step": 86
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.3778313398361206,
      "learning_rate": 8.18711994874345e-05,
      "loss": 0.863,
      "step": 88
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.160701870918274,
      "eval_runtime": 20.9855,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 88
    },
    {
      "epoch": 16.363636363636363,
      "grad_norm": 0.3163960576057434,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.91,
      "step": 90
    },
    {
      "epoch": 16.727272727272727,
      "grad_norm": 0.6867598295211792,
      "learning_rate": 7.967094433018508e-05,
      "loss": 0.8028,
      "step": 92
    },
    {
      "epoch": 16.90909090909091,
      "eval_loss": 1.1746482849121094,
      "eval_runtime": 20.975,
      "eval_samples_per_second": 1.144,
      "eval_steps_per_second": 1.144,
      "step": 93
    },
    {
      "epoch": 17.09090909090909,
      "grad_norm": 0.5571116209030151,
      "learning_rate": 7.85356783842216e-05,
      "loss": 0.7761,
      "step": 94
    },
    {
      "epoch": 17.454545454545453,
      "grad_norm": 0.7103570699691772,
      "learning_rate": 7.737816117462752e-05,
      "loss": 0.7825,
      "step": 96
    },
    {
      "epoch": 17.818181818181817,
      "grad_norm": 0.5428041219711304,
      "learning_rate": 7.619929529850397e-05,
      "loss": 0.731,
      "step": 98
    },
    {
      "epoch": 18.0,
      "eval_loss": 1.2033344507217407,
      "eval_runtime": 21.0145,
      "eval_samples_per_second": 1.142,
      "eval_steps_per_second": 1.142,
      "step": 99
    },
    {
      "epoch": 18.181818181818183,
      "grad_norm": 0.564351499080658,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.7785,
      "step": 100
    },
    {
      "epoch": 18.545454545454547,
      "grad_norm": 1.0754562616348267,
      "learning_rate": 7.378121045351378e-05,
      "loss": 0.7117,
      "step": 102
    },
    {
      "epoch": 18.90909090909091,
      "grad_norm": 0.6552812457084656,
      "learning_rate": 7.254387703447154e-05,
      "loss": 0.7089,
      "step": 104
    },
    {
      "epoch": 18.90909090909091,
      "eval_loss": 1.2295643091201782,
      "eval_runtime": 21.0231,
      "eval_samples_per_second": 1.142,
      "eval_steps_per_second": 1.142,
      "step": 104
    },
    {
      "epoch": 18.90909090909091,
      "step": 104,
      "total_flos": 1.3050541697400832e+17,
      "train_loss": 1.1235424764454365,
      "train_runtime": 4811.9738,
      "train_samples_per_second": 0.914,
      "train_steps_per_second": 0.052
    }
  ],
  "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": 1.3050541697400832e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}