{
  "best_metric": 0.44082728028297424,
  "best_model_checkpoint": "data/Llama-31-8B_task-3_120-samples_config-3/checkpoint-264",
  "epoch": 31.0,
  "eval_steps": 500,
  "global_step": 341,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09090909090909091,
      "grad_norm": 3.931244373321533,
      "learning_rate": 6.060606060606061e-08,
      "loss": 3.095,
      "step": 1
    },
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 2.530280828475952,
      "learning_rate": 1.2121212121212122e-07,
      "loss": 2.5399,
      "step": 2
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 2.5952248573303223,
      "learning_rate": 2.4242424242424244e-07,
      "loss": 2.5861,
      "step": 4
    },
    {
      "epoch": 0.5454545454545454,
      "grad_norm": 2.0955796241760254,
      "learning_rate": 3.6363636363636366e-07,
      "loss": 2.4442,
      "step": 6
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 3.4685494899749756,
      "learning_rate": 4.848484848484849e-07,
      "loss": 2.4938,
      "step": 8
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 3.055572509765625,
      "learning_rate": 6.060606060606061e-07,
      "loss": 2.3729,
      "step": 10
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.4972100257873535,
      "eval_runtime": 21.0088,
      "eval_samples_per_second": 1.142,
      "eval_steps_per_second": 1.142,
      "step": 11
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 2.5200321674346924,
      "learning_rate": 7.272727272727273e-07,
      "loss": 2.4004,
      "step": 12
    },
    {
      "epoch": 1.2727272727272727,
      "grad_norm": 2.5053160190582275,
      "learning_rate": 8.484848484848486e-07,
      "loss": 2.3639,
      "step": 14
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 2.5543558597564697,
      "learning_rate": 9.696969696969698e-07,
      "loss": 2.7736,
      "step": 16
    },
    {
      "epoch": 1.6363636363636362,
      "grad_norm": 2.6235134601593018,
      "learning_rate": 1.090909090909091e-06,
      "loss": 2.3351,
      "step": 18
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 2.2782557010650635,
      "learning_rate": 1.2121212121212122e-06,
      "loss": 2.4073,
      "step": 20
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.364241123199463,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 2.6938,
      "step": 22
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.457120180130005,
      "eval_runtime": 20.9973,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 22
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 2.8939199447631836,
      "learning_rate": 1.4545454545454546e-06,
      "loss": 2.6378,
      "step": 24
    },
    {
      "epoch": 2.3636363636363638,
      "grad_norm": 1.724478006362915,
      "learning_rate": 1.5757575757575759e-06,
      "loss": 2.3463,
      "step": 26
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 3.102917194366455,
      "learning_rate": 1.6969696969696973e-06,
      "loss": 2.3407,
      "step": 28
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 2.207613945007324,
      "learning_rate": 1.8181818181818183e-06,
      "loss": 2.4054,
      "step": 30
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 3.5173332691192627,
      "learning_rate": 1.9393939393939395e-06,
      "loss": 2.6474,
      "step": 32
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.3881404399871826,
      "eval_runtime": 20.9969,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 33
    },
    {
      "epoch": 3.090909090909091,
      "grad_norm": 2.588070869445801,
      "learning_rate": 2.0606060606060607e-06,
      "loss": 2.2888,
      "step": 34
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 3.921119451522827,
      "learning_rate": 2.181818181818182e-06,
      "loss": 2.858,
      "step": 36
    },
    {
      "epoch": 3.4545454545454546,
      "grad_norm": 3.1361024379730225,
      "learning_rate": 2.303030303030303e-06,
      "loss": 2.292,
      "step": 38
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 2.6496403217315674,
      "learning_rate": 2.4242424242424244e-06,
      "loss": 2.2874,
      "step": 40
    },
    {
      "epoch": 3.8181818181818183,
      "grad_norm": 1.7044334411621094,
      "learning_rate": 2.5454545454545456e-06,
      "loss": 2.0497,
      "step": 42
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.6162171363830566,
      "learning_rate": 2.666666666666667e-06,
      "loss": 2.2763,
      "step": 44
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.264249801635742,
      "eval_runtime": 21.0004,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 44
    },
    {
      "epoch": 4.181818181818182,
      "grad_norm": 3.5348732471466064,
      "learning_rate": 2.7878787878787885e-06,
      "loss": 2.486,
      "step": 46
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 1.8792097568511963,
      "learning_rate": 2.9090909090909093e-06,
      "loss": 2.0279,
      "step": 48
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 3.1981241703033447,
      "learning_rate": 3.0303030303030305e-06,
      "loss": 2.279,
      "step": 50
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 2.5685250759124756,
      "learning_rate": 3.1515151515151517e-06,
      "loss": 2.1978,
      "step": 52
    },
    {
      "epoch": 4.909090909090909,
      "grad_norm": 2.161813497543335,
      "learning_rate": 3.272727272727273e-06,
      "loss": 2.0268,
      "step": 54
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.069439649581909,
      "eval_runtime": 20.9965,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 55
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 2.084135055541992,
      "learning_rate": 3.3939393939393946e-06,
      "loss": 1.9478,
      "step": 56
    },
    {
      "epoch": 5.2727272727272725,
      "grad_norm": 2.176788806915283,
      "learning_rate": 3.5151515151515154e-06,
      "loss": 1.9753,
      "step": 58
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 4.001251220703125,
      "learning_rate": 3.6363636363636366e-06,
      "loss": 2.1318,
      "step": 60
    },
    {
      "epoch": 5.636363636363637,
      "grad_norm": 3.4586737155914307,
      "learning_rate": 3.757575757575758e-06,
      "loss": 1.9426,
      "step": 62
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 3.11710786819458,
      "learning_rate": 3.878787878787879e-06,
      "loss": 1.9089,
      "step": 64
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.0227549076080322,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.7309,
      "step": 66
    },
    {
      "epoch": 6.0,
      "eval_loss": 1.7871297597885132,
      "eval_runtime": 20.9996,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 66
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 3.2912845611572266,
      "learning_rate": 4.1212121212121215e-06,
      "loss": 1.8532,
      "step": 68
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 2.7779335975646973,
      "learning_rate": 4.242424242424243e-06,
      "loss": 1.7371,
      "step": 70
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 1.3037850856781006,
      "learning_rate": 4.363636363636364e-06,
      "loss": 1.4541,
      "step": 72
    },
    {
      "epoch": 6.7272727272727275,
      "grad_norm": 2.257749557495117,
      "learning_rate": 4.4848484848484855e-06,
      "loss": 1.5641,
      "step": 74
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 2.7067761421203613,
      "learning_rate": 4.606060606060606e-06,
      "loss": 1.4481,
      "step": 76
    },
    {
      "epoch": 7.0,
      "eval_loss": 1.4330077171325684,
      "eval_runtime": 21.0001,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 77
    },
    {
      "epoch": 7.090909090909091,
      "grad_norm": 1.7599331140518188,
      "learning_rate": 4.727272727272728e-06,
      "loss": 1.3604,
      "step": 78
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 1.4922908544540405,
      "learning_rate": 4.848484848484849e-06,
      "loss": 1.3767,
      "step": 80
    },
    {
      "epoch": 7.454545454545454,
      "grad_norm": 2.4537432193756104,
      "learning_rate": 4.9696969696969696e-06,
      "loss": 1.2013,
      "step": 82
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 2.7686843872070312,
      "learning_rate": 5.090909090909091e-06,
      "loss": 1.1578,
      "step": 84
    },
    {
      "epoch": 7.818181818181818,
      "grad_norm": 2.3368959426879883,
      "learning_rate": 5.212121212121213e-06,
      "loss": 1.1423,
      "step": 86
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.5859788656234741,
      "learning_rate": 5.333333333333334e-06,
      "loss": 1.0554,
      "step": 88
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.0675058364868164,
      "eval_runtime": 21.0113,
      "eval_samples_per_second": 1.142,
      "eval_steps_per_second": 1.142,
      "step": 88
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 1.8273411989212036,
      "learning_rate": 5.4545454545454545e-06,
      "loss": 0.9927,
      "step": 90
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 1.7687430381774902,
      "learning_rate": 5.575757575757577e-06,
      "loss": 0.8886,
      "step": 92
    },
    {
      "epoch": 8.545454545454545,
      "grad_norm": 1.9781934022903442,
      "learning_rate": 5.696969696969698e-06,
      "loss": 0.7076,
      "step": 94
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 1.8124470710754395,
      "learning_rate": 5.8181818181818185e-06,
      "loss": 0.8414,
      "step": 96
    },
    {
      "epoch": 8.909090909090908,
      "grad_norm": 1.2052216529846191,
      "learning_rate": 5.93939393939394e-06,
      "loss": 0.8392,
      "step": 98
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.756293773651123,
      "eval_runtime": 21.002,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 99
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 1.1055922508239746,
      "learning_rate": 6.060606060606061e-06,
      "loss": 0.7587,
      "step": 100
    },
    {
      "epoch": 9.272727272727273,
      "grad_norm": 1.4254761934280396,
      "learning_rate": 6.181818181818182e-06,
      "loss": 0.333,
      "step": 102
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 1.636863350868225,
      "learning_rate": 6.303030303030303e-06,
      "loss": 1.0304,
      "step": 104
    },
    {
      "epoch": 9.636363636363637,
      "grad_norm": 1.3520318269729614,
      "learning_rate": 6.424242424242425e-06,
      "loss": 0.5626,
      "step": 106
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 0.739101231098175,
      "learning_rate": 6.545454545454546e-06,
      "loss": 0.4562,
      "step": 108
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.719274640083313,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.4685,
      "step": 110
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.6436794400215149,
      "eval_runtime": 20.9986,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 110
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 0.5967914462089539,
      "learning_rate": 6.787878787878789e-06,
      "loss": 0.4207,
      "step": 112
    },
    {
      "epoch": 10.363636363636363,
      "grad_norm": 0.7981317639350891,
      "learning_rate": 6.90909090909091e-06,
      "loss": 0.7027,
      "step": 114
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 0.648346483707428,
      "learning_rate": 7.030303030303031e-06,
      "loss": 0.48,
      "step": 116
    },
    {
      "epoch": 10.727272727272727,
      "grad_norm": 1.0265989303588867,
      "learning_rate": 7.151515151515152e-06,
      "loss": 0.6033,
      "step": 118
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 0.6349014639854431,
      "learning_rate": 7.272727272727273e-06,
      "loss": 0.3588,
      "step": 120
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.5850974321365356,
      "eval_runtime": 21.004,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 121
    },
    {
      "epoch": 11.090909090909092,
      "grad_norm": 0.530517041683197,
      "learning_rate": 7.393939393939395e-06,
      "loss": 0.4937,
      "step": 122
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 0.9633344411849976,
      "learning_rate": 7.515151515151516e-06,
      "loss": 0.542,
      "step": 124
    },
    {
      "epoch": 11.454545454545455,
      "grad_norm": 0.6681090593338013,
      "learning_rate": 7.636363636363638e-06,
      "loss": 0.4627,
      "step": 126
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 0.4449699819087982,
      "learning_rate": 7.757575757575758e-06,
      "loss": 0.4319,
      "step": 128
    },
    {
      "epoch": 11.818181818181818,
      "grad_norm": 0.5469127297401428,
      "learning_rate": 7.87878787878788e-06,
      "loss": 0.3578,
      "step": 130
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.9779024720191956,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.6319,
      "step": 132
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.5406798720359802,
      "eval_runtime": 20.9992,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 132
    },
    {
      "epoch": 12.181818181818182,
      "grad_norm": 0.5859923958778381,
      "learning_rate": 8.121212121212121e-06,
      "loss": 0.3894,
      "step": 134
    },
    {
      "epoch": 12.363636363636363,
      "grad_norm": 0.1749761551618576,
      "learning_rate": 8.242424242424243e-06,
      "loss": 0.2537,
      "step": 136
    },
    {
      "epoch": 12.545454545454545,
      "grad_norm": 1.400254487991333,
      "learning_rate": 8.363636363636365e-06,
      "loss": 0.6108,
      "step": 138
    },
    {
      "epoch": 12.727272727272727,
      "grad_norm": 0.556860089302063,
      "learning_rate": 8.484848484848486e-06,
      "loss": 0.5365,
      "step": 140
    },
    {
      "epoch": 12.909090909090908,
      "grad_norm": 0.4626636207103729,
      "learning_rate": 8.606060606060606e-06,
      "loss": 0.4211,
      "step": 142
    },
    {
      "epoch": 13.0,
      "eval_loss": 0.5247967839241028,
      "eval_runtime": 21.0033,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 143
    },
    {
      "epoch": 13.090909090909092,
      "grad_norm": 0.42664870619773865,
      "learning_rate": 8.727272727272728e-06,
      "loss": 0.4939,
      "step": 144
    },
    {
      "epoch": 13.272727272727273,
      "grad_norm": 0.38396334648132324,
      "learning_rate": 8.84848484848485e-06,
      "loss": 0.3484,
      "step": 146
    },
    {
      "epoch": 13.454545454545455,
      "grad_norm": 0.43991151452064514,
      "learning_rate": 8.969696969696971e-06,
      "loss": 0.4156,
      "step": 148
    },
    {
      "epoch": 13.636363636363637,
      "grad_norm": 0.4237063229084015,
      "learning_rate": 9.090909090909091e-06,
      "loss": 0.3955,
      "step": 150
    },
    {
      "epoch": 13.818181818181818,
      "grad_norm": 0.34925854206085205,
      "learning_rate": 9.212121212121213e-06,
      "loss": 0.446,
      "step": 152
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.8939254879951477,
      "learning_rate": 9.333333333333334e-06,
      "loss": 0.495,
      "step": 154
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.5127188563346863,
      "eval_runtime": 20.9989,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 154
    },
    {
      "epoch": 14.181818181818182,
      "grad_norm": 0.4808824062347412,
      "learning_rate": 9.454545454545456e-06,
      "loss": 0.4217,
      "step": 156
    },
    {
      "epoch": 14.363636363636363,
      "grad_norm": 0.3635479807853699,
      "learning_rate": 9.575757575757576e-06,
      "loss": 0.4738,
      "step": 158
    },
    {
      "epoch": 14.545454545454545,
      "grad_norm": 0.2080135941505432,
      "learning_rate": 9.696969696969698e-06,
      "loss": 0.3693,
      "step": 160
    },
    {
      "epoch": 14.727272727272727,
      "grad_norm": 2.0220720767974854,
      "learning_rate": 9.81818181818182e-06,
      "loss": 0.3737,
      "step": 162
    },
    {
      "epoch": 14.909090909090908,
      "grad_norm": 0.3454424738883972,
      "learning_rate": 9.939393939393939e-06,
      "loss": 0.4232,
      "step": 164
    },
    {
      "epoch": 15.0,
      "eval_loss": 0.5019441246986389,
      "eval_runtime": 21.0036,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 165
    },
    {
      "epoch": 15.090909090909092,
      "grad_norm": 0.4212585389614105,
      "learning_rate": 9.999988811118232e-06,
      "loss": 0.4735,
      "step": 166
    },
    {
      "epoch": 15.272727272727273,
      "grad_norm": 0.22414664924144745,
      "learning_rate": 9.999899300364534e-06,
      "loss": 0.2503,
      "step": 168
    },
    {
      "epoch": 15.454545454545455,
      "grad_norm": 1.3229994773864746,
      "learning_rate": 9.999720280459576e-06,
      "loss": 0.4071,
      "step": 170
    },
    {
      "epoch": 15.636363636363637,
      "grad_norm": 0.29747679829597473,
      "learning_rate": 9.999451754608208e-06,
      "loss": 0.3427,
      "step": 172
    },
    {
      "epoch": 15.818181818181818,
      "grad_norm": 0.3459671437740326,
      "learning_rate": 9.99909372761763e-06,
      "loss": 0.4779,
      "step": 174
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.31325361132621765,
      "learning_rate": 9.99864620589731e-06,
      "loss": 0.496,
      "step": 176
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.5102820992469788,
      "eval_runtime": 21.0039,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 176
    },
    {
      "epoch": 16.181818181818183,
      "grad_norm": 0.35442739725112915,
      "learning_rate": 9.998109197458865e-06,
      "loss": 0.2946,
      "step": 178
    },
    {
      "epoch": 16.363636363636363,
      "grad_norm": 1.0050742626190186,
      "learning_rate": 9.997482711915926e-06,
      "loss": 0.5689,
      "step": 180
    },
    {
      "epoch": 16.545454545454547,
      "grad_norm": 0.3000654876232147,
      "learning_rate": 9.996766760483955e-06,
      "loss": 0.3062,
      "step": 182
    },
    {
      "epoch": 16.727272727272727,
      "grad_norm": 0.33660659193992615,
      "learning_rate": 9.995961355980052e-06,
      "loss": 0.3404,
      "step": 184
    },
    {
      "epoch": 16.90909090909091,
      "grad_norm": 0.32371142506599426,
      "learning_rate": 9.99506651282272e-06,
      "loss": 0.3903,
      "step": 186
    },
    {
      "epoch": 17.0,
      "eval_loss": 0.4814320504665375,
      "eval_runtime": 21.0016,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 187
    },
    {
      "epoch": 17.09090909090909,
      "grad_norm": 0.7857639193534851,
      "learning_rate": 9.994082247031613e-06,
      "loss": 0.5145,
      "step": 188
    },
    {
      "epoch": 17.272727272727273,
      "grad_norm": 0.007223070599138737,
      "learning_rate": 9.993008576227248e-06,
      "loss": 0.2366,
      "step": 190
    },
    {
      "epoch": 17.454545454545453,
      "grad_norm": 0.5182167887687683,
      "learning_rate": 9.991845519630679e-06,
      "loss": 0.4549,
      "step": 192
    },
    {
      "epoch": 17.636363636363637,
      "grad_norm": 0.2884611487388611,
      "learning_rate": 9.99059309806317e-06,
      "loss": 0.3861,
      "step": 194
    },
    {
      "epoch": 17.818181818181817,
      "grad_norm": 0.49307528138160706,
      "learning_rate": 9.989251333945813e-06,
      "loss": 0.4248,
      "step": 196
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.48076504468917847,
      "learning_rate": 9.987820251299121e-06,
      "loss": 0.331,
      "step": 198
    },
    {
      "epoch": 18.0,
      "eval_loss": 0.49132904410362244,
      "eval_runtime": 21.0024,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 198
    },
    {
      "epoch": 18.181818181818183,
      "grad_norm": 0.5222908854484558,
      "learning_rate": 9.986299875742612e-06,
      "loss": 0.4817,
      "step": 200
    },
    {
      "epoch": 18.363636363636363,
      "grad_norm": 0.3346542418003082,
      "learning_rate": 9.984690234494338e-06,
      "loss": 0.2685,
      "step": 202
    },
    {
      "epoch": 18.545454545454547,
      "grad_norm": 0.48757171630859375,
      "learning_rate": 9.982991356370404e-06,
      "loss": 0.4127,
      "step": 204
    },
    {
      "epoch": 18.727272727272727,
      "grad_norm": 0.5337949991226196,
      "learning_rate": 9.98120327178445e-06,
      "loss": 0.4215,
      "step": 206
    },
    {
      "epoch": 18.90909090909091,
      "grad_norm": 0.26037395000457764,
      "learning_rate": 9.979326012747106e-06,
      "loss": 0.2403,
      "step": 208
    },
    {
      "epoch": 19.0,
      "eval_loss": 0.4869350492954254,
      "eval_runtime": 21.0006,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 209
    },
    {
      "epoch": 19.09090909090909,
      "grad_norm": 0.31072062253952026,
      "learning_rate": 9.977359612865424e-06,
      "loss": 0.2671,
      "step": 210
    },
    {
      "epoch": 19.272727272727273,
      "grad_norm": 0.5128066539764404,
      "learning_rate": 9.975304107342268e-06,
      "loss": 0.3695,
      "step": 212
    },
    {
      "epoch": 19.454545454545453,
      "grad_norm": 0.585717499256134,
      "learning_rate": 9.973159532975691e-06,
      "loss": 0.5722,
      "step": 214
    },
    {
      "epoch": 19.636363636363637,
      "grad_norm": 0.4827703833580017,
      "learning_rate": 9.970925928158275e-06,
      "loss": 0.3577,
      "step": 216
    },
    {
      "epoch": 19.818181818181817,
      "grad_norm": 0.3153725266456604,
      "learning_rate": 9.968603332876435e-06,
      "loss": 0.2262,
      "step": 218
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.4635719954967499,
      "learning_rate": 9.966191788709716e-06,
      "loss": 0.3563,
      "step": 220
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.471805602312088,
      "eval_runtime": 21.0011,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 220
    },
    {
      "epoch": 20.181818181818183,
      "grad_norm": 0.4292517304420471,
      "learning_rate": 9.963691338830045e-06,
      "loss": 0.2494,
      "step": 222
    },
    {
      "epoch": 20.363636363636363,
      "grad_norm": 0.2989480495452881,
      "learning_rate": 9.961102028000948e-06,
      "loss": 0.2761,
      "step": 224
    },
    {
      "epoch": 20.545454545454547,
      "grad_norm": 0.4395551383495331,
      "learning_rate": 9.958423902576764e-06,
      "loss": 0.5072,
      "step": 226
    },
    {
      "epoch": 20.727272727272727,
      "grad_norm": 0.47977691888809204,
      "learning_rate": 9.955657010501807e-06,
      "loss": 0.2494,
      "step": 228
    },
    {
      "epoch": 20.90909090909091,
      "grad_norm": 0.4911171793937683,
      "learning_rate": 9.952801401309504e-06,
      "loss": 0.4107,
      "step": 230
    },
    {
      "epoch": 21.0,
      "eval_loss": 0.4596048593521118,
      "eval_runtime": 20.9979,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 231
    },
    {
      "epoch": 21.09090909090909,
      "grad_norm": 0.5808133482933044,
      "learning_rate": 9.949857126121519e-06,
      "loss": 0.4717,
      "step": 232
    },
    {
      "epoch": 21.272727272727273,
      "grad_norm": 0.19762107729911804,
      "learning_rate": 9.946824237646823e-06,
      "loss": 0.165,
      "step": 234
    },
    {
      "epoch": 21.454545454545453,
      "grad_norm": 0.5478127598762512,
      "learning_rate": 9.94370279018077e-06,
      "loss": 0.383,
      "step": 236
    },
    {
      "epoch": 21.636363636363637,
      "grad_norm": 0.3956260085105896,
      "learning_rate": 9.940492839604103e-06,
      "loss": 0.3194,
      "step": 238
    },
    {
      "epoch": 21.818181818181817,
      "grad_norm": 0.608517050743103,
      "learning_rate": 9.937194443381972e-06,
      "loss": 0.3811,
      "step": 240
    },
    {
      "epoch": 22.0,
      "grad_norm": 0.35069647431373596,
      "learning_rate": 9.933807660562898e-06,
      "loss": 0.2631,
      "step": 242
    },
    {
      "epoch": 22.0,
      "eval_loss": 0.4477730691432953,
      "eval_runtime": 20.9988,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 242
    },
    {
      "epoch": 22.181818181818183,
      "grad_norm": 0.5358024835586548,
      "learning_rate": 9.930332551777709e-06,
      "loss": 0.24,
      "step": 244
    },
    {
      "epoch": 22.363636363636363,
      "grad_norm": 0.5837539434432983,
      "learning_rate": 9.926769179238467e-06,
      "loss": 0.312,
      "step": 246
    },
    {
      "epoch": 22.545454545454547,
      "grad_norm": 0.4202631711959839,
      "learning_rate": 9.923117606737347e-06,
      "loss": 0.1839,
      "step": 248
    },
    {
      "epoch": 22.727272727272727,
      "grad_norm": 0.7312754988670349,
      "learning_rate": 9.919377899645497e-06,
      "loss": 0.3982,
      "step": 250
    },
    {
      "epoch": 22.90909090909091,
      "grad_norm": 0.6649301648139954,
      "learning_rate": 9.915550124911866e-06,
      "loss": 0.4212,
      "step": 252
    },
    {
      "epoch": 23.0,
      "eval_loss": 0.4496145248413086,
      "eval_runtime": 21.002,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 253
    },
    {
      "epoch": 23.09090909090909,
      "grad_norm": 0.4475117027759552,
      "learning_rate": 9.91163435106201e-06,
      "loss": 0.2837,
      "step": 254
    },
    {
      "epoch": 23.272727272727273,
      "grad_norm": 0.7331680059432983,
      "learning_rate": 9.907630648196857e-06,
      "loss": 0.4256,
      "step": 256
    },
    {
      "epoch": 23.454545454545453,
      "grad_norm": 0.5951094627380371,
      "learning_rate": 9.903539087991462e-06,
      "loss": 0.2876,
      "step": 258
    },
    {
      "epoch": 23.636363636363637,
      "grad_norm": 0.5116230845451355,
      "learning_rate": 9.899359743693715e-06,
      "loss": 0.2226,
      "step": 260
    },
    {
      "epoch": 23.818181818181817,
      "grad_norm": 0.4971112310886383,
      "learning_rate": 9.895092690123036e-06,
      "loss": 0.2511,
      "step": 262
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.0066973976790905,
      "learning_rate": 9.890738003669029e-06,
      "loss": 0.3304,
      "step": 264
    },
    {
      "epoch": 24.0,
      "eval_loss": 0.44082728028297424,
      "eval_runtime": 20.9935,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 264
    },
    {
      "epoch": 24.181818181818183,
      "grad_norm": 0.5074991583824158,
      "learning_rate": 9.886295762290125e-06,
      "loss": 0.1619,
      "step": 266
    },
    {
      "epoch": 24.363636363636363,
      "grad_norm": 0.803401529788971,
      "learning_rate": 9.881766045512176e-06,
      "loss": 0.3071,
      "step": 268
    },
    {
      "epoch": 24.545454545454547,
      "grad_norm": 0.7890399694442749,
      "learning_rate": 9.877148934427037e-06,
      "loss": 0.2939,
      "step": 270
    },
    {
      "epoch": 24.727272727272727,
      "grad_norm": 0.5034437775611877,
      "learning_rate": 9.872444511691108e-06,
      "loss": 0.2662,
      "step": 272
    },
    {
      "epoch": 24.90909090909091,
      "grad_norm": 0.7107832431793213,
      "learning_rate": 9.867652861523866e-06,
      "loss": 0.3296,
      "step": 274
    },
    {
      "epoch": 25.0,
      "eval_loss": 0.4437350928783417,
      "eval_runtime": 21.0123,
      "eval_samples_per_second": 1.142,
      "eval_steps_per_second": 1.142,
      "step": 275
    },
    {
      "epoch": 25.09090909090909,
      "grad_norm": 0.8815945386886597,
      "learning_rate": 9.862774069706346e-06,
      "loss": 0.3628,
      "step": 276
    },
    {
      "epoch": 25.272727272727273,
      "grad_norm": 0.6133561134338379,
      "learning_rate": 9.85780822357961e-06,
      "loss": 0.2528,
      "step": 278
    },
    {
      "epoch": 25.454545454545453,
      "grad_norm": 0.7392721772193909,
      "learning_rate": 9.85275541204318e-06,
      "loss": 0.166,
      "step": 280
    },
    {
      "epoch": 25.636363636363637,
      "grad_norm": 0.7326251268386841,
      "learning_rate": 9.847615725553457e-06,
      "loss": 0.1924,
      "step": 282
    },
    {
      "epoch": 25.818181818181817,
      "grad_norm": 0.9658777117729187,
      "learning_rate": 9.842389256122086e-06,
      "loss": 0.3393,
      "step": 284
    },
    {
      "epoch": 26.0,
      "grad_norm": 0.9744178056716919,
      "learning_rate": 9.83707609731432e-06,
      "loss": 0.3266,
      "step": 286
    },
    {
      "epoch": 26.0,
      "eval_loss": 0.44414618611335754,
      "eval_runtime": 20.9954,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 286
    },
    {
      "epoch": 26.181818181818183,
      "grad_norm": 1.009691834449768,
      "learning_rate": 9.831676344247343e-06,
      "loss": 0.2983,
      "step": 288
    },
    {
      "epoch": 26.363636363636363,
      "grad_norm": 0.9881241917610168,
      "learning_rate": 9.826190093588564e-06,
      "loss": 0.3114,
      "step": 290
    },
    {
      "epoch": 26.545454545454547,
      "grad_norm": 0.6508401036262512,
      "learning_rate": 9.820617443553889e-06,
      "loss": 0.1722,
      "step": 292
    },
    {
      "epoch": 26.727272727272727,
      "grad_norm": 1.0022698640823364,
      "learning_rate": 9.814958493905962e-06,
      "loss": 0.2775,
      "step": 294
    },
    {
      "epoch": 26.90909090909091,
      "grad_norm": 0.8220618963241577,
      "learning_rate": 9.80921334595238e-06,
      "loss": 0.1403,
      "step": 296
    },
    {
      "epoch": 27.0,
      "eval_loss": 0.4495992660522461,
      "eval_runtime": 21.0004,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 297
    },
    {
      "epoch": 27.09090909090909,
      "grad_norm": 0.6972556114196777,
      "learning_rate": 9.80338210254388e-06,
      "loss": 0.2361,
      "step": 298
    },
    {
      "epoch": 27.272727272727273,
      "grad_norm": 1.3660632371902466,
      "learning_rate": 9.797464868072489e-06,
      "loss": 0.1793,
      "step": 300
    },
    {
      "epoch": 27.454545454545453,
      "grad_norm": 0.9514533281326294,
      "learning_rate": 9.791461748469669e-06,
      "loss": 0.346,
      "step": 302
    },
    {
      "epoch": 27.636363636363637,
      "grad_norm": 1.0539543628692627,
      "learning_rate": 9.785372851204415e-06,
      "loss": 0.201,
      "step": 304
    },
    {
      "epoch": 27.818181818181817,
      "grad_norm": 1.111655592918396,
      "learning_rate": 9.779198285281326e-06,
      "loss": 0.2216,
      "step": 306
    },
    {
      "epoch": 28.0,
      "grad_norm": 0.7238084077835083,
      "learning_rate": 9.77293816123866e-06,
      "loss": 0.1732,
      "step": 308
    },
    {
      "epoch": 28.0,
      "eval_loss": 0.45740625262260437,
      "eval_runtime": 21.0057,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 308
    },
    {
      "epoch": 28.181818181818183,
      "grad_norm": 1.3565870523452759,
      "learning_rate": 9.766592591146353e-06,
      "loss": 0.3174,
      "step": 310
    },
    {
      "epoch": 28.363636363636363,
      "grad_norm": 1.3067699670791626,
      "learning_rate": 9.760161688604008e-06,
      "loss": 0.2076,
      "step": 312
    },
    {
      "epoch": 28.545454545454547,
      "grad_norm": 0.9948272705078125,
      "learning_rate": 9.753645568738872e-06,
      "loss": 0.1527,
      "step": 314
    },
    {
      "epoch": 28.727272727272727,
      "grad_norm": 1.3128433227539062,
      "learning_rate": 9.747044348203766e-06,
      "loss": 0.1789,
      "step": 316
    },
    {
      "epoch": 28.90909090909091,
      "grad_norm": 0.9066749215126038,
      "learning_rate": 9.740358145174999e-06,
      "loss": 0.1797,
      "step": 318
    },
    {
      "epoch": 29.0,
      "eval_loss": 0.4808818995952606,
      "eval_runtime": 20.9997,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 319
    },
    {
      "epoch": 29.09090909090909,
      "grad_norm": 1.0640830993652344,
      "learning_rate": 9.733587079350254e-06,
      "loss": 0.1481,
      "step": 320
    },
    {
      "epoch": 29.272727272727273,
      "grad_norm": 1.490294337272644,
      "learning_rate": 9.72673127194644e-06,
      "loss": 0.355,
      "step": 322
    },
    {
      "epoch": 29.454545454545453,
      "grad_norm": 0.6777256727218628,
      "learning_rate": 9.719790845697534e-06,
      "loss": 0.1048,
      "step": 324
    },
    {
      "epoch": 29.636363636363637,
      "grad_norm": 1.5173866748809814,
      "learning_rate": 9.71276592485237e-06,
      "loss": 0.1738,
      "step": 326
    },
    {
      "epoch": 29.818181818181817,
      "grad_norm": 1.211938500404358,
      "learning_rate": 9.705656635172418e-06,
      "loss": 0.0788,
      "step": 328
    },
    {
      "epoch": 30.0,
      "grad_norm": 1.4657930135726929,
      "learning_rate": 9.698463103929542e-06,
      "loss": 0.1355,
      "step": 330
    },
    {
      "epoch": 30.0,
      "eval_loss": 0.49904152750968933,
      "eval_runtime": 21.0013,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 330
    },
    {
      "epoch": 30.181818181818183,
      "grad_norm": 1.3539866209030151,
      "learning_rate": 9.69118545990371e-06,
      "loss": 0.1336,
      "step": 332
    },
    {
      "epoch": 30.363636363636363,
      "grad_norm": 1.1103911399841309,
      "learning_rate": 9.683823833380692e-06,
      "loss": 0.1738,
      "step": 334
    },
    {
      "epoch": 30.545454545454547,
      "grad_norm": 1.4360435009002686,
      "learning_rate": 9.676378356149733e-06,
      "loss": 0.0891,
      "step": 336
    },
    {
      "epoch": 30.727272727272727,
      "grad_norm": 1.7939261198043823,
      "learning_rate": 9.668849161501186e-06,
      "loss": 0.2107,
      "step": 338
    },
    {
      "epoch": 30.90909090909091,
      "grad_norm": 1.3665426969528198,
      "learning_rate": 9.66123638422413e-06,
      "loss": 0.1346,
      "step": 340
    },
    {
      "epoch": 31.0,
      "eval_loss": 0.5312966704368591,
      "eval_runtime": 21.0009,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 341
    },
    {
      "epoch": 31.0,
      "step": 341,
      "total_flos": 2.1292989192077312e+17,
      "train_loss": 0.8120344140050698,
      "train_runtime": 7851.2602,
      "train_samples_per_second": 1.681,
      "train_steps_per_second": 0.21
    }
  ],
  "logging_steps": 2,
  "max_steps": 1650,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 150,
  "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": 2.1292989192077312e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}