{
  "best_metric": 0.5754823088645935,
  "best_model_checkpoint": "data/Llama-31-8B_task-2_120-samples_config-1_auto/checkpoint-55",
  "epoch": 12.0,
  "eval_steps": 500,
  "global_step": 132,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09090909090909091,
      "grad_norm": 0.44860216975212097,
      "learning_rate": 1.818181818181818e-06,
      "loss": 0.881,
      "step": 1
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 0.4566896855831146,
      "learning_rate": 3.636363636363636e-06,
      "loss": 0.9713,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 0.39492738246917725,
      "learning_rate": 7.272727272727272e-06,
      "loss": 0.8458,
      "step": 4
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 0.46514591574668884,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 0.916,
      "step": 6
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 0.43416711688041687,
      "learning_rate": 1.4545454545454545e-05,
      "loss": 0.9211,
      "step": 8
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.4055511951446533,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 0.8648,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.8607686161994934,
      "eval_runtime": 25.0095,
      "eval_samples_per_second": 0.96,
      "eval_steps_per_second": 0.96,
      "step": 11
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 0.439016193151474,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 0.8753,
      "step": 12
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 0.3875233232975006,
      "learning_rate": 2.5454545454545454e-05,
      "loss": 0.8279,
      "step": 14
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 0.4131412208080292,
      "learning_rate": 2.909090909090909e-05,
      "loss": 0.827,
      "step": 16
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 0.324632853269577,
      "learning_rate": 3.272727272727273e-05,
      "loss": 0.8629,
      "step": 18
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.37663620710372925,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 0.7832,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3308725357055664,
      "learning_rate": 4e-05,
      "loss": 0.688,
      "step": 22
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.7372767329216003,
      "eval_runtime": 25.0199,
      "eval_samples_per_second": 0.959,
      "eval_steps_per_second": 0.959,
      "step": 22
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 0.336482971906662,
      "learning_rate": 4.3636363636363636e-05,
      "loss": 0.6865,
      "step": 24
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 0.3601330816745758,
      "learning_rate": 4.7272727272727275e-05,
      "loss": 0.7507,
      "step": 26
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.325101375579834,
      "learning_rate": 5.090909090909091e-05,
      "loss": 0.6832,
      "step": 28
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.3313121497631073,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 0.6488,
      "step": 30
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.35955771803855896,
      "learning_rate": 5.818181818181818e-05,
      "loss": 0.6118,
      "step": 32
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.6359238624572754,
      "eval_runtime": 25.0043,
      "eval_samples_per_second": 0.96,
      "eval_steps_per_second": 0.96,
      "step": 33
    },
    {
      "epoch": 3.090909090909091,
      "grad_norm": 0.36847934126853943,
      "learning_rate": 6.181818181818182e-05,
      "loss": 0.6509,
      "step": 34
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 0.32955634593963623,
      "learning_rate": 6.545454545454546e-05,
      "loss": 0.5562,
      "step": 36
    },
    {
      "epoch": 3.4545454545454546,
      "grad_norm": 0.3934621512889862,
      "learning_rate": 6.90909090909091e-05,
      "loss": 0.587,
      "step": 38
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.3738085925579071,
      "learning_rate": 7.272727272727273e-05,
      "loss": 0.5731,
      "step": 40
    },
    {
      "epoch": 3.8181818181818183,
      "grad_norm": 0.4378088414669037,
      "learning_rate": 7.636363636363637e-05,
      "loss": 0.5931,
      "step": 42
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.4045711159706116,
      "learning_rate": 8e-05,
      "loss": 0.5497,
      "step": 44
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.58708256483078,
      "eval_runtime": 25.0077,
      "eval_samples_per_second": 0.96,
      "eval_steps_per_second": 0.96,
      "step": 44
    },
    {
      "epoch": 4.181818181818182,
      "grad_norm": 0.4428233206272125,
      "learning_rate": 8.363636363636364e-05,
      "loss": 0.5404,
      "step": 46
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 0.3991241753101349,
      "learning_rate": 8.727272727272727e-05,
      "loss": 0.5197,
      "step": 48
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 0.4298638701438904,
      "learning_rate": 9.090909090909092e-05,
      "loss": 0.4817,
      "step": 50
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 0.5083165764808655,
      "learning_rate": 9.454545454545455e-05,
      "loss": 0.5148,
      "step": 52
    },
    {
      "epoch": 4.909090909090909,
      "grad_norm": 0.5078219175338745,
      "learning_rate": 9.818181818181818e-05,
      "loss": 0.4846,
      "step": 54
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.5754823088645935,
      "eval_runtime": 25.0095,
      "eval_samples_per_second": 0.96,
      "eval_steps_per_second": 0.96,
      "step": 55
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 0.39415740966796875,
      "learning_rate": 9.999899300364532e-05,
      "loss": 0.4644,
      "step": 56
    },
    {
      "epoch": 5.2727272727272725,
      "grad_norm": 0.4824540615081787,
      "learning_rate": 9.99909372761763e-05,
      "loss": 0.4429,
      "step": 58
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 0.4616365432739258,
      "learning_rate": 9.997482711915927e-05,
      "loss": 0.391,
      "step": 60
    },
    {
      "epoch": 5.636363636363637,
      "grad_norm": 0.5370848774909973,
      "learning_rate": 9.99506651282272e-05,
      "loss": 0.4113,
      "step": 62
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 0.5910454988479614,
      "learning_rate": 9.991845519630678e-05,
      "loss": 0.4175,
      "step": 64
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.5805835127830505,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.3946,
      "step": 66
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.5995001792907715,
      "eval_runtime": 25.0088,
      "eval_samples_per_second": 0.96,
      "eval_steps_per_second": 0.96,
      "step": 66
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 0.5539088249206543,
      "learning_rate": 9.982991356370404e-05,
      "loss": 0.3231,
      "step": 68
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 0.5546351075172424,
      "learning_rate": 9.977359612865423e-05,
      "loss": 0.3238,
      "step": 70
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 0.638353168964386,
      "learning_rate": 9.970925928158274e-05,
      "loss": 0.3096,
      "step": 72
    },
    {
      "epoch": 6.7272727272727275,
      "grad_norm": 0.7306607961654663,
      "learning_rate": 9.963691338830044e-05,
      "loss": 0.2636,
      "step": 74
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 0.675478994846344,
      "learning_rate": 9.955657010501806e-05,
      "loss": 0.2736,
      "step": 76
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.6777470111846924,
      "eval_runtime": 25.0055,
      "eval_samples_per_second": 0.96,
      "eval_steps_per_second": 0.96,
      "step": 77
    },
    {
      "epoch": 7.090909090909091,
      "grad_norm": 0.682803213596344,
      "learning_rate": 9.946824237646824e-05,
      "loss": 0.2445,
      "step": 78
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 0.7058468461036682,
      "learning_rate": 9.937194443381972e-05,
      "loss": 0.179,
      "step": 80
    },
    {
      "epoch": 7.454545454545454,
      "grad_norm": 1.1515408754348755,
      "learning_rate": 9.926769179238466e-05,
      "loss": 0.139,
      "step": 82
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 0.8244071006774902,
      "learning_rate": 9.915550124911866e-05,
      "loss": 0.1774,
      "step": 84
    },
    {
      "epoch": 7.818181818181818,
      "grad_norm": 0.6870058178901672,
      "learning_rate": 9.903539087991462e-05,
      "loss": 0.1817,
      "step": 86
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.7544240355491638,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.1769,
      "step": 88
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.7587476372718811,
      "eval_runtime": 25.0072,
      "eval_samples_per_second": 0.96,
      "eval_steps_per_second": 0.96,
      "step": 88
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 0.6142439842224121,
      "learning_rate": 9.877148934427037e-05,
      "loss": 0.1093,
      "step": 90
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 0.7897180914878845,
      "learning_rate": 9.862774069706346e-05,
      "loss": 0.0872,
      "step": 92
    },
    {
      "epoch": 8.545454545454545,
      "grad_norm": 0.9514899253845215,
      "learning_rate": 9.847615725553456e-05,
      "loss": 0.0842,
      "step": 94
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 0.8123044967651367,
      "learning_rate": 9.831676344247342e-05,
      "loss": 0.0912,
      "step": 96
    },
    {
      "epoch": 8.909090909090908,
      "grad_norm": 0.7924637794494629,
      "learning_rate": 9.814958493905963e-05,
      "loss": 0.1044,
      "step": 98
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.881658136844635,
      "eval_runtime": 25.0074,
      "eval_samples_per_second": 0.96,
      "eval_steps_per_second": 0.96,
      "step": 99
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 0.6067517399787903,
      "learning_rate": 9.797464868072488e-05,
      "loss": 0.084,
      "step": 100
    },
    {
      "epoch": 9.272727272727273,
      "grad_norm": 0.5800299644470215,
      "learning_rate": 9.779198285281325e-05,
      "loss": 0.0522,
      "step": 102
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 0.8385024666786194,
      "learning_rate": 9.760161688604008e-05,
      "loss": 0.0515,
      "step": 104
    },
    {
      "epoch": 9.636363636363637,
      "grad_norm": 0.7786041498184204,
      "learning_rate": 9.740358145174998e-05,
      "loss": 0.0549,
      "step": 106
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 0.883116602897644,
      "learning_rate": 9.719790845697533e-05,
      "loss": 0.0522,
      "step": 108
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.6970568299293518,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0471,
      "step": 110
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.9811952710151672,
      "eval_runtime": 25.0048,
      "eval_samples_per_second": 0.96,
      "eval_steps_per_second": 0.96,
      "step": 110
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 0.6507841944694519,
      "learning_rate": 9.676378356149734e-05,
      "loss": 0.0334,
      "step": 112
    },
    {
      "epoch": 10.363636363636363,
      "grad_norm": 0.4860031008720398,
      "learning_rate": 9.653540160603956e-05,
      "loss": 0.0325,
      "step": 114
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 0.724877655506134,
      "learning_rate": 9.629952196931901e-05,
      "loss": 0.0326,
      "step": 116
    },
    {
      "epoch": 10.727272727272727,
      "grad_norm": 0.8310495615005493,
      "learning_rate": 9.60561826557425e-05,
      "loss": 0.0483,
      "step": 118
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 0.7164592146873474,
      "learning_rate": 9.580542287160348e-05,
      "loss": 0.0359,
      "step": 120
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.001274585723877,
      "eval_runtime": 25.0133,
      "eval_samples_per_second": 0.959,
      "eval_steps_per_second": 0.959,
      "step": 121
    },
    {
      "epoch": 11.090909090909092,
      "grad_norm": 0.37490639090538025,
      "learning_rate": 9.554728301876526e-05,
      "loss": 0.0251,
      "step": 122
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 0.5960229635238647,
      "learning_rate": 9.528180468815155e-05,
      "loss": 0.0236,
      "step": 124
    },
    {
      "epoch": 11.454545454545455,
      "grad_norm": 0.5289472341537476,
      "learning_rate": 9.50090306530454e-05,
      "loss": 0.0222,
      "step": 126
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 0.5612404942512512,
      "learning_rate": 9.472900486219769e-05,
      "loss": 0.0233,
      "step": 128
    },
    {
      "epoch": 11.818181818181818,
      "grad_norm": 0.6438562870025635,
      "learning_rate": 9.444177243274618e-05,
      "loss": 0.0245,
      "step": 130
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.9628269672393799,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0371,
      "step": 132
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.0866872072219849,
      "eval_runtime": 25.0114,
      "eval_samples_per_second": 0.96,
      "eval_steps_per_second": 0.96,
      "step": 132
    },
    {
      "epoch": 12.0,
      "step": 132,
      "total_flos": 1.0451916704605798e+17,
      "train_loss": 0.37940981721674855,
      "train_runtime": 3728.223,
      "train_samples_per_second": 1.18,
      "train_steps_per_second": 0.148
    }
  ],
  "logging_steps": 2,
  "max_steps": 550,
  "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": 1.0451916704605798e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}