{
  "best_metric": 1.24324631690979,
  "best_model_checkpoint": "data/Llama-31-8B_task-1_120-samples_config-1/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": 1.947190523147583,
      "learning_rate": 1.818181818181818e-06,
      "loss": 2.2569,
      "step": 1
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 1.950323462486267,
      "learning_rate": 3.636363636363636e-06,
      "loss": 2.3228,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 1.7120546102523804,
      "learning_rate": 7.272727272727272e-06,
      "loss": 2.076,
      "step": 4
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 1.5879558324813843,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 2.0525,
      "step": 6
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 1.8351306915283203,
      "learning_rate": 1.4545454545454545e-05,
      "loss": 2.1059,
      "step": 8
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 1.7258864641189575,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 1.8587,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.8961697816848755,
      "eval_runtime": 9.6373,
      "eval_samples_per_second": 2.49,
      "eval_steps_per_second": 2.49,
      "step": 11
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 1.5399166345596313,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 1.7527,
      "step": 12
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 1.6511270999908447,
      "learning_rate": 2.5454545454545454e-05,
      "loss": 1.8337,
      "step": 14
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 1.104617714881897,
      "learning_rate": 2.909090909090909e-05,
      "loss": 1.7403,
      "step": 16
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 0.8533787131309509,
      "learning_rate": 3.272727272727273e-05,
      "loss": 1.5971,
      "step": 18
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.8006373047828674,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 1.503,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.7819918394088745,
      "learning_rate": 4e-05,
      "loss": 1.5582,
      "step": 22
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.5353227853775024,
      "eval_runtime": 9.6335,
      "eval_samples_per_second": 2.491,
      "eval_steps_per_second": 2.491,
      "step": 22
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 0.713280200958252,
      "learning_rate": 4.3636363636363636e-05,
      "loss": 1.5487,
      "step": 24
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 0.6750282645225525,
      "learning_rate": 4.7272727272727275e-05,
      "loss": 1.4773,
      "step": 26
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.7663286924362183,
      "learning_rate": 5.090909090909091e-05,
      "loss": 1.3698,
      "step": 28
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.6964442133903503,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 1.2677,
      "step": 30
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.7686213850975037,
      "learning_rate": 5.818181818181818e-05,
      "loss": 1.3116,
      "step": 32
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.3440297842025757,
      "eval_runtime": 9.6307,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 33
    },
    {
      "epoch": 3.090909090909091,
      "grad_norm": 0.7271563410758972,
      "learning_rate": 6.181818181818182e-05,
      "loss": 1.3272,
      "step": 34
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 0.7131862640380859,
      "learning_rate": 6.545454545454546e-05,
      "loss": 1.2423,
      "step": 36
    },
    {
      "epoch": 3.4545454545454546,
      "grad_norm": 0.8657882809638977,
      "learning_rate": 6.90909090909091e-05,
      "loss": 1.2197,
      "step": 38
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.8501375317573547,
      "learning_rate": 7.272727272727273e-05,
      "loss": 1.1791,
      "step": 40
    },
    {
      "epoch": 3.8181818181818183,
      "grad_norm": 0.8458476662635803,
      "learning_rate": 7.636363636363637e-05,
      "loss": 1.1082,
      "step": 42
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.8599070906639099,
      "learning_rate": 8e-05,
      "loss": 1.1103,
      "step": 44
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.248623251914978,
      "eval_runtime": 9.6225,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 44
    },
    {
      "epoch": 4.181818181818182,
      "grad_norm": 0.9287438988685608,
      "learning_rate": 8.363636363636364e-05,
      "loss": 0.991,
      "step": 46
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 1.0513927936553955,
      "learning_rate": 8.727272727272727e-05,
      "loss": 0.9697,
      "step": 48
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 1.1929577589035034,
      "learning_rate": 9.090909090909092e-05,
      "loss": 1.0229,
      "step": 50
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 1.1892277002334595,
      "learning_rate": 9.454545454545455e-05,
      "loss": 0.9632,
      "step": 52
    },
    {
      "epoch": 4.909090909090909,
      "grad_norm": 1.2755712270736694,
      "learning_rate": 9.818181818181818e-05,
      "loss": 0.9015,
      "step": 54
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.24324631690979,
      "eval_runtime": 9.6237,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 55
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 1.26650071144104,
      "learning_rate": 9.999899300364532e-05,
      "loss": 0.871,
      "step": 56
    },
    {
      "epoch": 5.2727272727272725,
      "grad_norm": 1.2812609672546387,
      "learning_rate": 9.99909372761763e-05,
      "loss": 0.7046,
      "step": 58
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 1.5597699880599976,
      "learning_rate": 9.997482711915927e-05,
      "loss": 0.6681,
      "step": 60
    },
    {
      "epoch": 5.636363636363637,
      "grad_norm": 1.8858792781829834,
      "learning_rate": 9.99506651282272e-05,
      "loss": 0.7369,
      "step": 62
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 1.7140628099441528,
      "learning_rate": 9.991845519630678e-05,
      "loss": 0.6761,
      "step": 64
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.500331163406372,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.6339,
      "step": 66
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.3448349237442017,
      "eval_runtime": 9.633,
      "eval_samples_per_second": 2.491,
      "eval_steps_per_second": 2.491,
      "step": 66
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 1.3713549375534058,
      "learning_rate": 9.982991356370404e-05,
      "loss": 0.527,
      "step": 68
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 1.5518769025802612,
      "learning_rate": 9.977359612865423e-05,
      "loss": 0.4046,
      "step": 70
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 2.7726385593414307,
      "learning_rate": 9.970925928158274e-05,
      "loss": 0.3872,
      "step": 72
    },
    {
      "epoch": 6.7272727272727275,
      "grad_norm": 3.0638201236724854,
      "learning_rate": 9.963691338830044e-05,
      "loss": 0.4758,
      "step": 74
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 1.8400838375091553,
      "learning_rate": 9.955657010501806e-05,
      "loss": 0.2953,
      "step": 76
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.664857268333435,
      "eval_runtime": 9.6249,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 77
    },
    {
      "epoch": 7.090909090909091,
      "grad_norm": 1.4201287031173706,
      "learning_rate": 9.946824237646824e-05,
      "loss": 0.2398,
      "step": 78
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 1.569122314453125,
      "learning_rate": 9.937194443381972e-05,
      "loss": 0.2232,
      "step": 80
    },
    {
      "epoch": 7.454545454545454,
      "grad_norm": 2.7534537315368652,
      "learning_rate": 9.926769179238466e-05,
      "loss": 0.1809,
      "step": 82
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 3.2907912731170654,
      "learning_rate": 9.915550124911866e-05,
      "loss": 0.2009,
      "step": 84
    },
    {
      "epoch": 7.818181818181818,
      "grad_norm": 2.7623965740203857,
      "learning_rate": 9.903539087991462e-05,
      "loss": 0.2162,
      "step": 86
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.723360776901245,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.2611,
      "step": 88
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.9194618463516235,
      "eval_runtime": 9.6254,
      "eval_samples_per_second": 2.493,
      "eval_steps_per_second": 2.493,
      "step": 88
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 1.1859649419784546,
      "learning_rate": 9.877148934427037e-05,
      "loss": 0.13,
      "step": 90
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 1.631119966506958,
      "learning_rate": 9.862774069706346e-05,
      "loss": 0.0969,
      "step": 92
    },
    {
      "epoch": 8.545454545454545,
      "grad_norm": 1.6909385919570923,
      "learning_rate": 9.847615725553456e-05,
      "loss": 0.0749,
      "step": 94
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 2.5971972942352295,
      "learning_rate": 9.831676344247342e-05,
      "loss": 0.0947,
      "step": 96
    },
    {
      "epoch": 8.909090909090908,
      "grad_norm": 2.3750057220458984,
      "learning_rate": 9.814958493905963e-05,
      "loss": 0.107,
      "step": 98
    },
    {
      "epoch": 9.0,
      "eval_loss": 2.4669106006622314,
      "eval_runtime": 9.6325,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 99
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 1.5737000703811646,
      "learning_rate": 9.797464868072488e-05,
      "loss": 0.0641,
      "step": 100
    },
    {
      "epoch": 9.272727272727273,
      "grad_norm": 1.2060961723327637,
      "learning_rate": 9.779198285281325e-05,
      "loss": 0.0515,
      "step": 102
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 0.8381581902503967,
      "learning_rate": 9.760161688604008e-05,
      "loss": 0.04,
      "step": 104
    },
    {
      "epoch": 9.636363636363637,
      "grad_norm": 2.207648754119873,
      "learning_rate": 9.740358145174998e-05,
      "loss": 0.0481,
      "step": 106
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 2.044506311416626,
      "learning_rate": 9.719790845697533e-05,
      "loss": 0.0573,
      "step": 108
    },
    {
      "epoch": 10.0,
      "grad_norm": 2.117966413497925,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0592,
      "step": 110
    },
    {
      "epoch": 10.0,
      "eval_loss": 2.5538697242736816,
      "eval_runtime": 9.6217,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 110
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 1.4448171854019165,
      "learning_rate": 9.676378356149734e-05,
      "loss": 0.021,
      "step": 112
    },
    {
      "epoch": 10.363636363636363,
      "grad_norm": 1.2305335998535156,
      "learning_rate": 9.653540160603956e-05,
      "loss": 0.0286,
      "step": 114
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 1.0550709962844849,
      "learning_rate": 9.629952196931901e-05,
      "loss": 0.0274,
      "step": 116
    },
    {
      "epoch": 10.727272727272727,
      "grad_norm": 1.6463170051574707,
      "learning_rate": 9.60561826557425e-05,
      "loss": 0.0486,
      "step": 118
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 1.7919278144836426,
      "learning_rate": 9.580542287160348e-05,
      "loss": 0.06,
      "step": 120
    },
    {
      "epoch": 11.0,
      "eval_loss": 2.47693133354187,
      "eval_runtime": 9.6239,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 121
    },
    {
      "epoch": 11.090909090909092,
      "grad_norm": 0.7210983037948608,
      "learning_rate": 9.554728301876526e-05,
      "loss": 0.0226,
      "step": 122
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 0.9753442406654358,
      "learning_rate": 9.528180468815155e-05,
      "loss": 0.0168,
      "step": 124
    },
    {
      "epoch": 11.454545454545455,
      "grad_norm": 1.1973837614059448,
      "learning_rate": 9.50090306530454e-05,
      "loss": 0.0248,
      "step": 126
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 0.9973687529563904,
      "learning_rate": 9.472900486219769e-05,
      "loss": 0.0107,
      "step": 128
    },
    {
      "epoch": 11.818181818181818,
      "grad_norm": 1.4094384908676147,
      "learning_rate": 9.444177243274618e-05,
      "loss": 0.0223,
      "step": 130
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.581376552581787,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0548,
      "step": 132
    },
    {
      "epoch": 12.0,
      "eval_loss": 2.9299166202545166,
      "eval_runtime": 9.6267,
      "eval_samples_per_second": 2.493,
      "eval_steps_per_second": 2.493,
      "step": 132
    },
    {
      "epoch": 12.0,
      "step": 132,
      "total_flos": 1.6209403933884416e+16,
      "train_loss": 0.7309455393926438,
      "train_runtime": 1236.3069,
      "train_samples_per_second": 3.559,
      "train_steps_per_second": 0.445
    }
  ],
  "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": 1.6209403933884416e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}