{
  "best_metric": 1.0298420190811157,
  "best_model_checkpoint": "data/Llama-31-8B_task-2_120-samples_config-1_full/checkpoint-88",
  "epoch": 15.0,
  "eval_steps": 500,
  "global_step": 165,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09090909090909091,
      "grad_norm": 0.49713072180747986,
      "learning_rate": 1.818181818181818e-06,
      "loss": 1.5048,
      "step": 1
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 0.6580377817153931,
      "learning_rate": 3.636363636363636e-06,
      "loss": 1.7142,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 0.4897933304309845,
      "learning_rate": 7.272727272727272e-06,
      "loss": 1.5422,
      "step": 4
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 0.507178783416748,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 1.5741,
      "step": 6
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 0.48809972405433655,
      "learning_rate": 1.4545454545454545e-05,
      "loss": 1.6288,
      "step": 8
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.4498646557331085,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 1.5352,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.53848397731781,
      "eval_runtime": 24.3714,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 11
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 0.5035683512687683,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 1.5785,
      "step": 12
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 0.37113550305366516,
      "learning_rate": 2.5454545454545454e-05,
      "loss": 1.4958,
      "step": 14
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 0.470533162355423,
      "learning_rate": 2.909090909090909e-05,
      "loss": 1.492,
      "step": 16
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 0.3534870445728302,
      "learning_rate": 3.272727272727273e-05,
      "loss": 1.5078,
      "step": 18
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.37165534496307373,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 1.4209,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.2557329833507538,
      "learning_rate": 4e-05,
      "loss": 1.3167,
      "step": 22
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.3841285705566406,
      "eval_runtime": 24.3704,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 22
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 0.225205197930336,
      "learning_rate": 4.3636363636363636e-05,
      "loss": 1.3669,
      "step": 24
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 0.2652900218963623,
      "learning_rate": 4.7272727272727275e-05,
      "loss": 1.3475,
      "step": 26
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.23897986114025116,
      "learning_rate": 5.090909090909091e-05,
      "loss": 1.2816,
      "step": 28
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.2523474395275116,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 1.2912,
      "step": 30
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.2484261691570282,
      "learning_rate": 5.818181818181818e-05,
      "loss": 1.226,
      "step": 32
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.2298388481140137,
      "eval_runtime": 24.3715,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 33
    },
    {
      "epoch": 3.090909090909091,
      "grad_norm": 0.2355325073003769,
      "learning_rate": 6.181818181818182e-05,
      "loss": 1.2254,
      "step": 34
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 0.23926451802253723,
      "learning_rate": 6.545454545454546e-05,
      "loss": 1.1508,
      "step": 36
    },
    {
      "epoch": 3.4545454545454546,
      "grad_norm": 0.36814334988594055,
      "learning_rate": 6.90909090909091e-05,
      "loss": 1.142,
      "step": 38
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.25748685002326965,
      "learning_rate": 7.272727272727273e-05,
      "loss": 1.0562,
      "step": 40
    },
    {
      "epoch": 3.8181818181818183,
      "grad_norm": 0.4333396852016449,
      "learning_rate": 7.636363636363637e-05,
      "loss": 1.0066,
      "step": 42
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.16409271955490112,
      "learning_rate": 8e-05,
      "loss": 1.1287,
      "step": 44
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.0978490114212036,
      "eval_runtime": 24.3666,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 44
    },
    {
      "epoch": 4.181818181818182,
      "grad_norm": 0.16672073304653168,
      "learning_rate": 8.363636363636364e-05,
      "loss": 1.0102,
      "step": 46
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 0.16206879913806915,
      "learning_rate": 8.727272727272727e-05,
      "loss": 1.0474,
      "step": 48
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 0.1528177112340927,
      "learning_rate": 9.090909090909092e-05,
      "loss": 1.0141,
      "step": 50
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 0.1910155862569809,
      "learning_rate": 9.454545454545455e-05,
      "loss": 1.0145,
      "step": 52
    },
    {
      "epoch": 4.909090909090909,
      "grad_norm": 0.1659548580646515,
      "learning_rate": 9.818181818181818e-05,
      "loss": 1.06,
      "step": 54
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.0666165351867676,
      "eval_runtime": 24.3738,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 55
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 0.16574902832508087,
      "learning_rate": 9.999899300364532e-05,
      "loss": 0.9889,
      "step": 56
    },
    {
      "epoch": 5.2727272727272725,
      "grad_norm": 0.18105646967887878,
      "learning_rate": 9.99909372761763e-05,
      "loss": 1.0223,
      "step": 58
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 0.176856130361557,
      "learning_rate": 9.997482711915927e-05,
      "loss": 0.9484,
      "step": 60
    },
    {
      "epoch": 5.636363636363637,
      "grad_norm": 0.18571056425571442,
      "learning_rate": 9.99506651282272e-05,
      "loss": 0.9777,
      "step": 62
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 0.18585607409477234,
      "learning_rate": 9.991845519630678e-05,
      "loss": 0.9662,
      "step": 64
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.1818024069070816,
      "learning_rate": 9.987820251299122e-05,
      "loss": 1.019,
      "step": 66
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.0471864938735962,
      "eval_runtime": 24.3649,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 66
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 0.2027784138917923,
      "learning_rate": 9.982991356370404e-05,
      "loss": 0.9194,
      "step": 68
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 0.2038961946964264,
      "learning_rate": 9.977359612865423e-05,
      "loss": 0.9722,
      "step": 70
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 0.19705639779567719,
      "learning_rate": 9.970925928158274e-05,
      "loss": 0.8931,
      "step": 72
    },
    {
      "epoch": 6.7272727272727275,
      "grad_norm": 0.244102343916893,
      "learning_rate": 9.963691338830044e-05,
      "loss": 0.9479,
      "step": 74
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 0.18375569581985474,
      "learning_rate": 9.955657010501806e-05,
      "loss": 0.9816,
      "step": 76
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.0346657037734985,
      "eval_runtime": 24.3789,
      "eval_samples_per_second": 0.984,
      "eval_steps_per_second": 0.984,
      "step": 77
    },
    {
      "epoch": 7.090909090909091,
      "grad_norm": 0.1986258625984192,
      "learning_rate": 9.946824237646824e-05,
      "loss": 1.0092,
      "step": 78
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 0.23222436010837555,
      "learning_rate": 9.937194443381972e-05,
      "loss": 0.8659,
      "step": 80
    },
    {
      "epoch": 7.454545454545454,
      "grad_norm": 0.24806098639965057,
      "learning_rate": 9.926769179238466e-05,
      "loss": 0.8708,
      "step": 82
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 0.24764443933963776,
      "learning_rate": 9.915550124911866e-05,
      "loss": 0.9041,
      "step": 84
    },
    {
      "epoch": 7.818181818181818,
      "grad_norm": 0.268614798784256,
      "learning_rate": 9.903539087991462e-05,
      "loss": 0.8901,
      "step": 86
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.26300254464149475,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.9461,
      "step": 88
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.0298420190811157,
      "eval_runtime": 24.3698,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 88
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 0.25266140699386597,
      "learning_rate": 9.877148934427037e-05,
      "loss": 0.8964,
      "step": 90
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 1.057136058807373,
      "learning_rate": 9.862774069706346e-05,
      "loss": 0.8322,
      "step": 92
    },
    {
      "epoch": 8.545454545454545,
      "grad_norm": 0.33643069863319397,
      "learning_rate": 9.847615725553456e-05,
      "loss": 0.8889,
      "step": 94
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 0.30312344431877136,
      "learning_rate": 9.831676344247342e-05,
      "loss": 0.789,
      "step": 96
    },
    {
      "epoch": 8.909090909090908,
      "grad_norm": 0.3003259301185608,
      "learning_rate": 9.814958493905963e-05,
      "loss": 0.8633,
      "step": 98
    },
    {
      "epoch": 9.0,
      "eval_loss": 1.0388423204421997,
      "eval_runtime": 24.3676,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 99
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 0.3024749159812927,
      "learning_rate": 9.797464868072488e-05,
      "loss": 0.873,
      "step": 100
    },
    {
      "epoch": 9.272727272727273,
      "grad_norm": 0.3733336627483368,
      "learning_rate": 9.779198285281325e-05,
      "loss": 0.7714,
      "step": 102
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 0.420226126909256,
      "learning_rate": 9.760161688604008e-05,
      "loss": 0.7677,
      "step": 104
    },
    {
      "epoch": 9.636363636363637,
      "grad_norm": 0.3949902653694153,
      "learning_rate": 9.740358145174998e-05,
      "loss": 0.8519,
      "step": 106
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 0.41283905506134033,
      "learning_rate": 9.719790845697533e-05,
      "loss": 0.7942,
      "step": 108
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.4529339075088501,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.7599,
      "step": 110
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.059997797012329,
      "eval_runtime": 24.3707,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 110
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 0.4940122663974762,
      "learning_rate": 9.676378356149734e-05,
      "loss": 0.7012,
      "step": 112
    },
    {
      "epoch": 10.363636363636363,
      "grad_norm": 0.4954428970813751,
      "learning_rate": 9.653540160603956e-05,
      "loss": 0.7028,
      "step": 114
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 0.49719637632369995,
      "learning_rate": 9.629952196931901e-05,
      "loss": 0.711,
      "step": 116
    },
    {
      "epoch": 10.727272727272727,
      "grad_norm": 0.6597090363502502,
      "learning_rate": 9.60561826557425e-05,
      "loss": 0.6915,
      "step": 118
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 0.5840014219284058,
      "learning_rate": 9.580542287160348e-05,
      "loss": 0.7543,
      "step": 120
    },
    {
      "epoch": 11.0,
      "eval_loss": 1.1064456701278687,
      "eval_runtime": 24.3667,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 121
    },
    {
      "epoch": 11.090909090909092,
      "grad_norm": 0.6744766235351562,
      "learning_rate": 9.554728301876526e-05,
      "loss": 0.684,
      "step": 122
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 0.7541329860687256,
      "learning_rate": 9.528180468815155e-05,
      "loss": 0.6297,
      "step": 124
    },
    {
      "epoch": 11.454545454545455,
      "grad_norm": 0.6155569553375244,
      "learning_rate": 9.50090306530454e-05,
      "loss": 0.6376,
      "step": 126
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 0.5912525653839111,
      "learning_rate": 9.472900486219769e-05,
      "loss": 0.6693,
      "step": 128
    },
    {
      "epoch": 11.818181818181818,
      "grad_norm": 0.7152079939842224,
      "learning_rate": 9.444177243274618e-05,
      "loss": 0.6336,
      "step": 130
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.6318918466567993,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.5739,
      "step": 132
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.159354567527771,
      "eval_runtime": 24.3607,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 132
    },
    {
      "epoch": 12.181818181818182,
      "grad_norm": 0.6897273659706116,
      "learning_rate": 9.384587392471515e-05,
      "loss": 0.5698,
      "step": 134
    },
    {
      "epoch": 12.363636363636363,
      "grad_norm": 0.8041525483131409,
      "learning_rate": 9.353730385598887e-05,
      "loss": 0.5094,
      "step": 136
    },
    {
      "epoch": 12.545454545454545,
      "grad_norm": 0.6993588209152222,
      "learning_rate": 9.322171915289635e-05,
      "loss": 0.5329,
      "step": 138
    },
    {
      "epoch": 12.727272727272727,
      "grad_norm": 0.8220525979995728,
      "learning_rate": 9.289917066174886e-05,
      "loss": 0.5145,
      "step": 140
    },
    {
      "epoch": 12.909090909090908,
      "grad_norm": 0.6971898078918457,
      "learning_rate": 9.256971035084785e-05,
      "loss": 0.5887,
      "step": 142
    },
    {
      "epoch": 13.0,
      "eval_loss": 1.194595456123352,
      "eval_runtime": 24.3612,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 143
    },
    {
      "epoch": 13.090909090909092,
      "grad_norm": 0.9051700830459595,
      "learning_rate": 9.223339130211192e-05,
      "loss": 0.4197,
      "step": 144
    },
    {
      "epoch": 13.272727272727273,
      "grad_norm": 1.2017525434494019,
      "learning_rate": 9.189026770252436e-05,
      "loss": 0.4788,
      "step": 146
    },
    {
      "epoch": 13.454545454545455,
      "grad_norm": 0.8263213634490967,
      "learning_rate": 9.154039483540273e-05,
      "loss": 0.5146,
      "step": 148
    },
    {
      "epoch": 13.636363636363637,
      "grad_norm": 0.9688396453857422,
      "learning_rate": 9.118382907149165e-05,
      "loss": 0.459,
      "step": 150
    },
    {
      "epoch": 13.818181818181818,
      "grad_norm": 0.8769497871398926,
      "learning_rate": 9.082062785988049e-05,
      "loss": 0.425,
      "step": 152
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.868798017501831,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.3635,
      "step": 154
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.3156319856643677,
      "eval_runtime": 24.3601,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 154
    },
    {
      "epoch": 14.181818181818182,
      "grad_norm": 0.8629772067070007,
      "learning_rate": 9.007455422593077e-05,
      "loss": 0.3637,
      "step": 156
    },
    {
      "epoch": 14.363636363636363,
      "grad_norm": 1.1226879358291626,
      "learning_rate": 8.969180200933047e-05,
      "loss": 0.3725,
      "step": 158
    },
    {
      "epoch": 14.545454545454545,
      "grad_norm": 0.8470020294189453,
      "learning_rate": 8.930265473713938e-05,
      "loss": 0.4403,
      "step": 160
    },
    {
      "epoch": 14.727272727272727,
      "grad_norm": 1.1817563772201538,
      "learning_rate": 8.890717510790763e-05,
      "loss": 0.3476,
      "step": 162
    },
    {
      "epoch": 14.909090909090908,
      "grad_norm": 0.9473496675491333,
      "learning_rate": 8.850542684044078e-05,
      "loss": 0.2782,
      "step": 164
    },
    {
      "epoch": 15.0,
      "eval_loss": 1.3628578186035156,
      "eval_runtime": 24.3663,
      "eval_samples_per_second": 0.985,
      "eval_steps_per_second": 0.985,
      "step": 165
    },
    {
      "epoch": 15.0,
      "step": 165,
      "total_flos": 1.2546788345407078e+17,
      "train_loss": 0.9098414410244334,
      "train_runtime": 4509.3746,
      "train_samples_per_second": 0.976,
      "train_steps_per_second": 0.122
    }
  ],
  "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": 1.2546788345407078e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}