{
  "best_metric": 0.7883281707763672,
  "best_model_checkpoint": "data/Mistral-7B_task-2_120-samples_config-1_full/checkpoint-66",
  "epoch": 13.0,
  "eval_steps": 500,
  "global_step": 143,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09090909090909091,
      "grad_norm": 1.1011921167373657,
      "learning_rate": 1.818181818181818e-06,
      "loss": 1.1337,
      "step": 1
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 1.392738699913025,
      "learning_rate": 3.636363636363636e-06,
      "loss": 1.3187,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 1.102772831916809,
      "learning_rate": 7.272727272727272e-06,
      "loss": 1.1577,
      "step": 4
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 1.1216496229171753,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 1.1662,
      "step": 6
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 1.0292140245437622,
      "learning_rate": 1.4545454545454545e-05,
      "loss": 1.2153,
      "step": 8
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 0.77543705701828,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 1.0983,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.0808161497116089,
      "eval_runtime": 30.0166,
      "eval_samples_per_second": 0.8,
      "eval_steps_per_second": 0.8,
      "step": 11
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 0.6813668608665466,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 1.1177,
      "step": 12
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 0.435611367225647,
      "learning_rate": 2.5454545454545454e-05,
      "loss": 1.0331,
      "step": 14
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 0.42835795879364014,
      "learning_rate": 2.909090909090909e-05,
      "loss": 1.0366,
      "step": 16
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 0.40383854508399963,
      "learning_rate": 3.272727272727273e-05,
      "loss": 1.047,
      "step": 18
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.37422505021095276,
      "learning_rate": 3.6363636363636364e-05,
      "loss": 0.9824,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3606886863708496,
      "learning_rate": 4e-05,
      "loss": 0.8983,
      "step": 22
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.953758180141449,
      "eval_runtime": 30.0246,
      "eval_samples_per_second": 0.799,
      "eval_steps_per_second": 0.799,
      "step": 22
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 0.30870339274406433,
      "learning_rate": 4.3636363636363636e-05,
      "loss": 0.9373,
      "step": 24
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 0.4794676899909973,
      "learning_rate": 4.7272727272727275e-05,
      "loss": 0.919,
      "step": 26
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 0.39165472984313965,
      "learning_rate": 5.090909090909091e-05,
      "loss": 0.8539,
      "step": 28
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 0.3516630232334137,
      "learning_rate": 5.4545454545454546e-05,
      "loss": 0.8662,
      "step": 30
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 0.3009703755378723,
      "learning_rate": 5.818181818181818e-05,
      "loss": 0.8093,
      "step": 32
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.8323513865470886,
      "eval_runtime": 30.0119,
      "eval_samples_per_second": 0.8,
      "eval_steps_per_second": 0.8,
      "step": 33
    },
    {
      "epoch": 3.090909090909091,
      "grad_norm": 0.2687840759754181,
      "learning_rate": 6.181818181818182e-05,
      "loss": 0.8227,
      "step": 34
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 0.2438051998615265,
      "learning_rate": 6.545454545454546e-05,
      "loss": 0.7844,
      "step": 36
    },
    {
      "epoch": 3.4545454545454546,
      "grad_norm": 0.2567192316055298,
      "learning_rate": 6.90909090909091e-05,
      "loss": 0.7855,
      "step": 38
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 0.2168852835893631,
      "learning_rate": 7.272727272727273e-05,
      "loss": 0.7365,
      "step": 40
    },
    {
      "epoch": 3.8181818181818183,
      "grad_norm": 0.3096124231815338,
      "learning_rate": 7.636363636363637e-05,
      "loss": 0.7191,
      "step": 42
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.24166655540466309,
      "learning_rate": 8e-05,
      "loss": 0.812,
      "step": 44
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.8044827580451965,
      "eval_runtime": 30.0135,
      "eval_samples_per_second": 0.8,
      "eval_steps_per_second": 0.8,
      "step": 44
    },
    {
      "epoch": 4.181818181818182,
      "grad_norm": 0.2755947411060333,
      "learning_rate": 8.363636363636364e-05,
      "loss": 0.7234,
      "step": 46
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 0.2453605979681015,
      "learning_rate": 8.727272727272727e-05,
      "loss": 0.7506,
      "step": 48
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 0.2451571524143219,
      "learning_rate": 9.090909090909092e-05,
      "loss": 0.722,
      "step": 50
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 0.3140670359134674,
      "learning_rate": 9.454545454545455e-05,
      "loss": 0.7278,
      "step": 52
    },
    {
      "epoch": 4.909090909090909,
      "grad_norm": 0.36974820494651794,
      "learning_rate": 9.818181818181818e-05,
      "loss": 0.7556,
      "step": 54
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.7907748222351074,
      "eval_runtime": 30.0136,
      "eval_samples_per_second": 0.8,
      "eval_steps_per_second": 0.8,
      "step": 55
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 0.2563894987106323,
      "learning_rate": 9.999899300364532e-05,
      "loss": 0.7018,
      "step": 56
    },
    {
      "epoch": 5.2727272727272725,
      "grad_norm": 0.2989403307437897,
      "learning_rate": 9.99909372761763e-05,
      "loss": 0.717,
      "step": 58
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 0.32498207688331604,
      "learning_rate": 9.997482711915927e-05,
      "loss": 0.6604,
      "step": 60
    },
    {
      "epoch": 5.636363636363637,
      "grad_norm": 0.32759106159210205,
      "learning_rate": 9.99506651282272e-05,
      "loss": 0.6733,
      "step": 62
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 0.33633723855018616,
      "learning_rate": 9.991845519630678e-05,
      "loss": 0.6688,
      "step": 64
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.3074813187122345,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.7048,
      "step": 66
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.7883281707763672,
      "eval_runtime": 30.0108,
      "eval_samples_per_second": 0.8,
      "eval_steps_per_second": 0.8,
      "step": 66
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 0.3761835992336273,
      "learning_rate": 9.982991356370404e-05,
      "loss": 0.6087,
      "step": 68
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 0.39026913046836853,
      "learning_rate": 9.977359612865423e-05,
      "loss": 0.6425,
      "step": 70
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 0.4483635127544403,
      "learning_rate": 9.970925928158274e-05,
      "loss": 0.5776,
      "step": 72
    },
    {
      "epoch": 6.7272727272727275,
      "grad_norm": 0.47310715913772583,
      "learning_rate": 9.963691338830044e-05,
      "loss": 0.6171,
      "step": 74
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 0.4454406797885895,
      "learning_rate": 9.955657010501806e-05,
      "loss": 0.6515,
      "step": 76
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.7959274649620056,
      "eval_runtime": 30.0186,
      "eval_samples_per_second": 0.8,
      "eval_steps_per_second": 0.8,
      "step": 77
    },
    {
      "epoch": 7.090909090909091,
      "grad_norm": 0.4192354083061218,
      "learning_rate": 9.946824237646824e-05,
      "loss": 0.6557,
      "step": 78
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 0.5135143399238586,
      "learning_rate": 9.937194443381972e-05,
      "loss": 0.5054,
      "step": 80
    },
    {
      "epoch": 7.454545454545454,
      "grad_norm": 0.6870741844177246,
      "learning_rate": 9.926769179238466e-05,
      "loss": 0.5194,
      "step": 82
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 0.5848849415779114,
      "learning_rate": 9.915550124911866e-05,
      "loss": 0.5496,
      "step": 84
    },
    {
      "epoch": 7.818181818181818,
      "grad_norm": 0.5785425305366516,
      "learning_rate": 9.903539087991462e-05,
      "loss": 0.5373,
      "step": 86
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.6328079104423523,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.5883,
      "step": 88
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.8303840160369873,
      "eval_runtime": 30.0128,
      "eval_samples_per_second": 0.8,
      "eval_steps_per_second": 0.8,
      "step": 88
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 0.6976420879364014,
      "learning_rate": 9.877148934427037e-05,
      "loss": 0.505,
      "step": 90
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 1.2223607301712036,
      "learning_rate": 9.862774069706346e-05,
      "loss": 0.4639,
      "step": 92
    },
    {
      "epoch": 8.545454545454545,
      "grad_norm": 0.7318143844604492,
      "learning_rate": 9.847615725553456e-05,
      "loss": 0.4862,
      "step": 94
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 0.6903765201568604,
      "learning_rate": 9.831676344247342e-05,
      "loss": 0.4106,
      "step": 96
    },
    {
      "epoch": 8.909090909090908,
      "grad_norm": 0.8092580437660217,
      "learning_rate": 9.814958493905963e-05,
      "loss": 0.4745,
      "step": 98
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.8685193061828613,
      "eval_runtime": 30.0124,
      "eval_samples_per_second": 0.8,
      "eval_steps_per_second": 0.8,
      "step": 99
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 0.6964694857597351,
      "learning_rate": 9.797464868072488e-05,
      "loss": 0.4666,
      "step": 100
    },
    {
      "epoch": 9.272727272727273,
      "grad_norm": 0.9236999154090881,
      "learning_rate": 9.779198285281325e-05,
      "loss": 0.3613,
      "step": 102
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 0.9183457493782043,
      "learning_rate": 9.760161688604008e-05,
      "loss": 0.3662,
      "step": 104
    },
    {
      "epoch": 9.636363636363637,
      "grad_norm": 0.9439888000488281,
      "learning_rate": 9.740358145174998e-05,
      "loss": 0.4229,
      "step": 106
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 0.8765619993209839,
      "learning_rate": 9.719790845697533e-05,
      "loss": 0.3876,
      "step": 108
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.483403205871582,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.3673,
      "step": 110
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.927913248538971,
      "eval_runtime": 30.007,
      "eval_samples_per_second": 0.8,
      "eval_steps_per_second": 0.8,
      "step": 110
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 0.7813490033149719,
      "learning_rate": 9.676378356149734e-05,
      "loss": 0.2983,
      "step": 112
    },
    {
      "epoch": 10.363636363636363,
      "grad_norm": 1.1128252744674683,
      "learning_rate": 9.653540160603956e-05,
      "loss": 0.3078,
      "step": 114
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 0.850096583366394,
      "learning_rate": 9.629952196931901e-05,
      "loss": 0.3013,
      "step": 116
    },
    {
      "epoch": 10.727272727272727,
      "grad_norm": 0.9651398062705994,
      "learning_rate": 9.60561826557425e-05,
      "loss": 0.3184,
      "step": 118
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 1.1538136005401611,
      "learning_rate": 9.580542287160348e-05,
      "loss": 0.3406,
      "step": 120
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.9977002143859863,
      "eval_runtime": 30.0285,
      "eval_samples_per_second": 0.799,
      "eval_steps_per_second": 0.799,
      "step": 121
    },
    {
      "epoch": 11.090909090909092,
      "grad_norm": 0.910765528678894,
      "learning_rate": 9.554728301876526e-05,
      "loss": 0.2753,
      "step": 122
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 1.1856135129928589,
      "learning_rate": 9.528180468815155e-05,
      "loss": 0.2395,
      "step": 124
    },
    {
      "epoch": 11.454545454545455,
      "grad_norm": 1.0970741510391235,
      "learning_rate": 9.50090306530454e-05,
      "loss": 0.2535,
      "step": 126
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 0.8762866258621216,
      "learning_rate": 9.472900486219769e-05,
      "loss": 0.2833,
      "step": 128
    },
    {
      "epoch": 11.818181818181818,
      "grad_norm": 1.1564828157424927,
      "learning_rate": 9.444177243274618e-05,
      "loss": 0.2489,
      "step": 130
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.9118844270706177,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.2499,
      "step": 132
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.029537558555603,
      "eval_runtime": 30.0146,
      "eval_samples_per_second": 0.8,
      "eval_steps_per_second": 0.8,
      "step": 132
    },
    {
      "epoch": 12.181818181818182,
      "grad_norm": 0.7883229851722717,
      "learning_rate": 9.384587392471515e-05,
      "loss": 0.2215,
      "step": 134
    },
    {
      "epoch": 12.363636363636363,
      "grad_norm": 1.2054721117019653,
      "learning_rate": 9.353730385598887e-05,
      "loss": 0.1921,
      "step": 136
    },
    {
      "epoch": 12.545454545454545,
      "grad_norm": 0.9445869326591492,
      "learning_rate": 9.322171915289635e-05,
      "loss": 0.2046,
      "step": 138
    },
    {
      "epoch": 12.727272727272727,
      "grad_norm": 0.9175389409065247,
      "learning_rate": 9.289917066174886e-05,
      "loss": 0.1948,
      "step": 140
    },
    {
      "epoch": 12.909090909090908,
      "grad_norm": 1.1180704832077026,
      "learning_rate": 9.256971035084785e-05,
      "loss": 0.2273,
      "step": 142
    },
    {
      "epoch": 13.0,
      "eval_loss": 1.1281015872955322,
      "eval_runtime": 30.0147,
      "eval_samples_per_second": 0.8,
      "eval_steps_per_second": 0.8,
      "step": 143
    },
    {
      "epoch": 13.0,
      "step": 143,
      "total_flos": 1.3485981883275674e+17,
      "train_loss": 0.618694863744549,
      "train_runtime": 4908.8927,
      "train_samples_per_second": 0.896,
      "train_steps_per_second": 0.112
    }
  ],
  "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": 1.3485981883275674e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}