{
  "best_metric": 0.44272685050964355,
  "best_model_checkpoint": "data/Llama-31-8B_task-3_120-samples_config-4/checkpoint-214",
  "epoch": 46.0,
  "eval_steps": 500,
  "global_step": 253,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 3.1690733432769775,
      "learning_rate": 1.3333333333333336e-07,
      "loss": 2.8174,
      "step": 1
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 2.785377025604248,
      "learning_rate": 2.666666666666667e-07,
      "loss": 2.5884,
      "step": 2
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 2.4831278324127197,
      "learning_rate": 5.333333333333335e-07,
      "loss": 2.4707,
      "step": 4
    },
    {
      "epoch": 0.9090909090909091,
      "eval_loss": 2.502408742904663,
      "eval_runtime": 20.9991,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 5
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 2.511209726333618,
      "learning_rate": 8.000000000000001e-07,
      "loss": 2.3921,
      "step": 6
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 3.1637580394744873,
      "learning_rate": 1.066666666666667e-06,
      "loss": 2.5814,
      "step": 8
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 2.612966775894165,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 2.3847,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.479367971420288,
      "eval_runtime": 21.0052,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 11
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 2.905508518218994,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 2.6936,
      "step": 12
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 2.6060760021209717,
      "learning_rate": 1.8666666666666669e-06,
      "loss": 2.3757,
      "step": 14
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 3.210188150405884,
      "learning_rate": 2.133333333333334e-06,
      "loss": 2.5822,
      "step": 16
    },
    {
      "epoch": 2.909090909090909,
      "eval_loss": 2.4506218433380127,
      "eval_runtime": 21.0015,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 16
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 3.861948013305664,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 2.6532,
      "step": 18
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 2.5222151279449463,
      "learning_rate": 2.666666666666667e-06,
      "loss": 2.3743,
      "step": 20
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.5924019813537598,
      "learning_rate": 2.9333333333333338e-06,
      "loss": 2.2635,
      "step": 22
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.3902969360351562,
      "eval_runtime": 21.0176,
      "eval_samples_per_second": 1.142,
      "eval_steps_per_second": 1.142,
      "step": 22
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 2.239548921585083,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 2.4045,
      "step": 24
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 2.6937167644500732,
      "learning_rate": 3.4666666666666672e-06,
      "loss": 2.446,
      "step": 26
    },
    {
      "epoch": 4.909090909090909,
      "eval_loss": 2.3183953762054443,
      "eval_runtime": 21.0027,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 27
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 2.26949143409729,
      "learning_rate": 3.7333333333333337e-06,
      "loss": 2.1955,
      "step": 28
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 3.2933804988861084,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.3794,
      "step": 30
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 3.081024169921875,
      "learning_rate": 4.266666666666668e-06,
      "loss": 2.3172,
      "step": 32
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.19378399848938,
      "eval_runtime": 20.9996,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 33
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 3.226388454437256,
      "learning_rate": 4.533333333333334e-06,
      "loss": 2.2373,
      "step": 34
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 1.6646192073822021,
      "learning_rate": 4.800000000000001e-06,
      "loss": 2.0122,
      "step": 36
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 2.573935031890869,
      "learning_rate": 5.0666666666666676e-06,
      "loss": 2.0582,
      "step": 38
    },
    {
      "epoch": 6.909090909090909,
      "eval_loss": 2.0559873580932617,
      "eval_runtime": 21.0001,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 38
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 2.028982639312744,
      "learning_rate": 5.333333333333334e-06,
      "loss": 1.9739,
      "step": 40
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 2.90993595123291,
      "learning_rate": 5.600000000000001e-06,
      "loss": 2.0373,
      "step": 42
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.5906777381896973,
      "learning_rate": 5.8666666666666675e-06,
      "loss": 1.9038,
      "step": 44
    },
    {
      "epoch": 8.0,
      "eval_loss": 1.8385127782821655,
      "eval_runtime": 20.9958,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 44
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 2.548511266708374,
      "learning_rate": 6.133333333333334e-06,
      "loss": 1.7707,
      "step": 46
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 2.2620716094970703,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 1.7291,
      "step": 48
    },
    {
      "epoch": 8.909090909090908,
      "eval_loss": 1.625160574913025,
      "eval_runtime": 20.9978,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 49
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 2.2526345252990723,
      "learning_rate": 6.666666666666667e-06,
      "loss": 1.607,
      "step": 50
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 1.2636234760284424,
      "learning_rate": 6.9333333333333344e-06,
      "loss": 1.5173,
      "step": 52
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 2.455042600631714,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 1.3996,
      "step": 54
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.345643162727356,
      "eval_runtime": 21.0044,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 55
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 2.40154767036438,
      "learning_rate": 7.4666666666666675e-06,
      "loss": 1.298,
      "step": 56
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 2.0422167778015137,
      "learning_rate": 7.733333333333334e-06,
      "loss": 1.2468,
      "step": 58
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 2.330064058303833,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.1127,
      "step": 60
    },
    {
      "epoch": 10.909090909090908,
      "eval_loss": 1.1189473867416382,
      "eval_runtime": 21.0042,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 60
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 1.6237120628356934,
      "learning_rate": 8.266666666666667e-06,
      "loss": 1.0346,
      "step": 62
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 1.9251773357391357,
      "learning_rate": 8.533333333333335e-06,
      "loss": 0.9213,
      "step": 64
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.3815380334854126,
      "learning_rate": 8.8e-06,
      "loss": 0.8648,
      "step": 66
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.8626702427864075,
      "eval_runtime": 21.003,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 66
    },
    {
      "epoch": 12.363636363636363,
      "grad_norm": 1.8721394538879395,
      "learning_rate": 9.066666666666667e-06,
      "loss": 0.6295,
      "step": 68
    },
    {
      "epoch": 12.727272727272727,
      "grad_norm": 1.1077913045883179,
      "learning_rate": 9.333333333333334e-06,
      "loss": 0.8247,
      "step": 70
    },
    {
      "epoch": 12.909090909090908,
      "eval_loss": 0.7228104472160339,
      "eval_runtime": 20.9976,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 71
    },
    {
      "epoch": 13.090909090909092,
      "grad_norm": 0.9280027151107788,
      "learning_rate": 9.600000000000001e-06,
      "loss": 0.6643,
      "step": 72
    },
    {
      "epoch": 13.454545454545455,
      "grad_norm": 0.8919649720191956,
      "learning_rate": 9.866666666666668e-06,
      "loss": 0.5425,
      "step": 74
    },
    {
      "epoch": 13.818181818181818,
      "grad_norm": 0.8291063904762268,
      "learning_rate": 9.999945845889795e-06,
      "loss": 0.5681,
      "step": 76
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.6452742218971252,
      "eval_runtime": 21.0019,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 77
    },
    {
      "epoch": 14.181818181818182,
      "grad_norm": 0.8017318844795227,
      "learning_rate": 9.999512620046523e-06,
      "loss": 0.6102,
      "step": 78
    },
    {
      "epoch": 14.545454545454545,
      "grad_norm": 0.48972490429878235,
      "learning_rate": 9.99864620589731e-06,
      "loss": 0.5325,
      "step": 80
    },
    {
      "epoch": 14.909090909090908,
      "grad_norm": 0.6467877626419067,
      "learning_rate": 9.99734667851357e-06,
      "loss": 0.4968,
      "step": 82
    },
    {
      "epoch": 14.909090909090908,
      "eval_loss": 0.6019716858863831,
      "eval_runtime": 21.0041,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 82
    },
    {
      "epoch": 15.272727272727273,
      "grad_norm": 0.4685284495353699,
      "learning_rate": 9.995614150494293e-06,
      "loss": 0.4665,
      "step": 84
    },
    {
      "epoch": 15.636363636363637,
      "grad_norm": 0.6542117595672607,
      "learning_rate": 9.993448771956285e-06,
      "loss": 0.4592,
      "step": 86
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.7055411338806152,
      "learning_rate": 9.99085073052117e-06,
      "loss": 0.589,
      "step": 88
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.5632085800170898,
      "eval_runtime": 21.0058,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 88
    },
    {
      "epoch": 16.363636363636363,
      "grad_norm": 0.5362458825111389,
      "learning_rate": 9.987820251299121e-06,
      "loss": 0.5109,
      "step": 90
    },
    {
      "epoch": 16.727272727272727,
      "grad_norm": 0.3406112492084503,
      "learning_rate": 9.984357596869369e-06,
      "loss": 0.3902,
      "step": 92
    },
    {
      "epoch": 16.90909090909091,
      "eval_loss": 0.5393884778022766,
      "eval_runtime": 21.0005,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 93
    },
    {
      "epoch": 17.09090909090909,
      "grad_norm": 0.5570920705795288,
      "learning_rate": 9.980463067257437e-06,
      "loss": 0.5325,
      "step": 94
    },
    {
      "epoch": 17.454545454545453,
      "grad_norm": 0.40238556265830994,
      "learning_rate": 9.976136999909156e-06,
      "loss": 0.4039,
      "step": 96
    },
    {
      "epoch": 17.818181818181817,
      "grad_norm": 0.6511011719703674,
      "learning_rate": 9.971379769661422e-06,
      "loss": 0.4795,
      "step": 98
    },
    {
      "epoch": 18.0,
      "eval_loss": 0.5253307819366455,
      "eval_runtime": 21.0047,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 99
    },
    {
      "epoch": 18.181818181818183,
      "grad_norm": 0.4080711305141449,
      "learning_rate": 9.966191788709716e-06,
      "loss": 0.4767,
      "step": 100
    },
    {
      "epoch": 18.545454545454547,
      "grad_norm": 0.35794946551322937,
      "learning_rate": 9.960573506572391e-06,
      "loss": 0.4024,
      "step": 102
    },
    {
      "epoch": 18.90909090909091,
      "grad_norm": 0.2140028029680252,
      "learning_rate": 9.95452541005172e-06,
      "loss": 0.3937,
      "step": 104
    },
    {
      "epoch": 18.90909090909091,
      "eval_loss": 0.5187709927558899,
      "eval_runtime": 20.9985,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 104
    },
    {
      "epoch": 19.272727272727273,
      "grad_norm": 0.26322704553604126,
      "learning_rate": 9.948048023191728e-06,
      "loss": 0.3845,
      "step": 106
    },
    {
      "epoch": 19.636363636363637,
      "grad_norm": 0.29932188987731934,
      "learning_rate": 9.941141907232766e-06,
      "loss": 0.5589,
      "step": 108
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.3077118396759033,
      "learning_rate": 9.933807660562898e-06,
      "loss": 0.3482,
      "step": 110
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.5148636698722839,
      "eval_runtime": 21.0007,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 110
    },
    {
      "epoch": 20.363636363636363,
      "grad_norm": 0.2408452033996582,
      "learning_rate": 9.926045918666045e-06,
      "loss": 0.3168,
      "step": 112
    },
    {
      "epoch": 20.727272727272727,
      "grad_norm": 0.24809201061725616,
      "learning_rate": 9.91785735406693e-06,
      "loss": 0.4633,
      "step": 114
    },
    {
      "epoch": 20.90909090909091,
      "eval_loss": 0.5076439380645752,
      "eval_runtime": 21.008,
      "eval_samples_per_second": 1.142,
      "eval_steps_per_second": 1.142,
      "step": 115
    },
    {
      "epoch": 21.09090909090909,
      "grad_norm": 0.4201977550983429,
      "learning_rate": 9.909242676272797e-06,
      "loss": 0.5529,
      "step": 116
    },
    {
      "epoch": 21.454545454545453,
      "grad_norm": 0.30744773149490356,
      "learning_rate": 9.90020263171194e-06,
      "loss": 0.3469,
      "step": 118
    },
    {
      "epoch": 21.818181818181817,
      "grad_norm": 0.2413516491651535,
      "learning_rate": 9.890738003669029e-06,
      "loss": 0.4324,
      "step": 120
    },
    {
      "epoch": 22.0,
      "eval_loss": 0.507264256477356,
      "eval_runtime": 20.9993,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 121
    },
    {
      "epoch": 22.181818181818183,
      "grad_norm": 0.291746586561203,
      "learning_rate": 9.880849612217238e-06,
      "loss": 0.325,
      "step": 122
    },
    {
      "epoch": 22.545454545454547,
      "grad_norm": 0.22690318524837494,
      "learning_rate": 9.870538314147194e-06,
      "loss": 0.3262,
      "step": 124
    },
    {
      "epoch": 22.90909090909091,
      "grad_norm": 0.28505831956863403,
      "learning_rate": 9.859805002892733e-06,
      "loss": 0.5268,
      "step": 126
    },
    {
      "epoch": 22.90909090909091,
      "eval_loss": 0.5014194846153259,
      "eval_runtime": 21.0084,
      "eval_samples_per_second": 1.142,
      "eval_steps_per_second": 1.142,
      "step": 126
    },
    {
      "epoch": 23.272727272727273,
      "grad_norm": 0.33074063062667847,
      "learning_rate": 9.84865060845349e-06,
      "loss": 0.4705,
      "step": 128
    },
    {
      "epoch": 23.636363636363637,
      "grad_norm": 0.24479380249977112,
      "learning_rate": 9.83707609731432e-06,
      "loss": 0.3475,
      "step": 130
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.2692837715148926,
      "learning_rate": 9.825082472361558e-06,
      "loss": 0.3829,
      "step": 132
    },
    {
      "epoch": 24.0,
      "eval_loss": 0.49462565779685974,
      "eval_runtime": 21.0006,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 132
    },
    {
      "epoch": 24.363636363636363,
      "grad_norm": 0.2481246441602707,
      "learning_rate": 9.812670772796113e-06,
      "loss": 0.3245,
      "step": 134
    },
    {
      "epoch": 24.727272727272727,
      "grad_norm": 0.2520200312137604,
      "learning_rate": 9.799842074043438e-06,
      "loss": 0.3884,
      "step": 136
    },
    {
      "epoch": 24.90909090909091,
      "eval_loss": 0.4874545633792877,
      "eval_runtime": 21.0022,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 137
    },
    {
      "epoch": 25.09090909090909,
      "grad_norm": 0.34048882126808167,
      "learning_rate": 9.786597487660336e-06,
      "loss": 0.5025,
      "step": 138
    },
    {
      "epoch": 25.454545454545453,
      "grad_norm": 0.2133476287126541,
      "learning_rate": 9.77293816123866e-06,
      "loss": 0.3012,
      "step": 140
    },
    {
      "epoch": 25.818181818181817,
      "grad_norm": 0.3172047734260559,
      "learning_rate": 9.75886527830587e-06,
      "loss": 0.3955,
      "step": 142
    },
    {
      "epoch": 26.0,
      "eval_loss": 0.4859234094619751,
      "eval_runtime": 21.0019,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 143
    },
    {
      "epoch": 26.181818181818183,
      "grad_norm": 0.299968957901001,
      "learning_rate": 9.744380058222483e-06,
      "loss": 0.471,
      "step": 144
    },
    {
      "epoch": 26.545454545454547,
      "grad_norm": 0.22688376903533936,
      "learning_rate": 9.729483756076436e-06,
      "loss": 0.361,
      "step": 146
    },
    {
      "epoch": 26.90909090909091,
      "grad_norm": 0.20205911993980408,
      "learning_rate": 9.714177662574316e-06,
      "loss": 0.3296,
      "step": 148
    },
    {
      "epoch": 26.90909090909091,
      "eval_loss": 0.482715368270874,
      "eval_runtime": 21.0014,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 148
    },
    {
      "epoch": 27.272727272727273,
      "grad_norm": 0.2501104772090912,
      "learning_rate": 9.698463103929542e-06,
      "loss": 0.3371,
      "step": 150
    },
    {
      "epoch": 27.636363636363637,
      "grad_norm": 0.2598528265953064,
      "learning_rate": 9.682341441747446e-06,
      "loss": 0.439,
      "step": 152
    },
    {
      "epoch": 28.0,
      "grad_norm": 0.25387951731681824,
      "learning_rate": 9.665814072907293e-06,
      "loss": 0.364,
      "step": 154
    },
    {
      "epoch": 28.0,
      "eval_loss": 0.4805121123790741,
      "eval_runtime": 21.002,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 154
    },
    {
      "epoch": 28.363636363636363,
      "grad_norm": 0.2787465751171112,
      "learning_rate": 9.648882429441258e-06,
      "loss": 0.4797,
      "step": 156
    },
    {
      "epoch": 28.727272727272727,
      "grad_norm": 0.25384321808815,
      "learning_rate": 9.63154797841033e-06,
      "loss": 0.3218,
      "step": 158
    },
    {
      "epoch": 28.90909090909091,
      "eval_loss": 0.4764772951602936,
      "eval_runtime": 21.0046,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 159
    },
    {
      "epoch": 29.09090909090909,
      "grad_norm": 0.2499808818101883,
      "learning_rate": 9.613812221777212e-06,
      "loss": 0.302,
      "step": 160
    },
    {
      "epoch": 29.454545454545453,
      "grad_norm": 0.24313043057918549,
      "learning_rate": 9.595676696276173e-06,
      "loss": 0.4349,
      "step": 162
    },
    {
      "epoch": 29.818181818181817,
      "grad_norm": 0.26249364018440247,
      "learning_rate": 9.577142973279896e-06,
      "loss": 0.2995,
      "step": 164
    },
    {
      "epoch": 30.0,
      "eval_loss": 0.4727340638637543,
      "eval_runtime": 21.0056,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 165
    },
    {
      "epoch": 30.181818181818183,
      "grad_norm": 0.3164134621620178,
      "learning_rate": 9.55821265866333e-06,
      "loss": 0.3373,
      "step": 166
    },
    {
      "epoch": 30.545454545454547,
      "grad_norm": 0.23215709626674652,
      "learning_rate": 9.538887392664544e-06,
      "loss": 0.3434,
      "step": 168
    },
    {
      "epoch": 30.90909090909091,
      "grad_norm": 0.24047435820102692,
      "learning_rate": 9.519168849742603e-06,
      "loss": 0.3728,
      "step": 170
    },
    {
      "epoch": 30.90909090909091,
      "eval_loss": 0.4668484032154083,
      "eval_runtime": 21.0083,
      "eval_samples_per_second": 1.142,
      "eval_steps_per_second": 1.142,
      "step": 170
    },
    {
      "epoch": 31.272727272727273,
      "grad_norm": 0.38881421089172363,
      "learning_rate": 9.499058738432492e-06,
      "loss": 0.4329,
      "step": 172
    },
    {
      "epoch": 31.636363636363637,
      "grad_norm": 0.38587087392807007,
      "learning_rate": 9.478558801197065e-06,
      "loss": 0.3596,
      "step": 174
    },
    {
      "epoch": 32.0,
      "grad_norm": 0.2352934628725052,
      "learning_rate": 9.457670814276083e-06,
      "loss": 0.2413,
      "step": 176
    },
    {
      "epoch": 32.0,
      "eval_loss": 0.46528127789497375,
      "eval_runtime": 20.9994,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 176
    },
    {
      "epoch": 32.36363636363637,
      "grad_norm": 0.32384243607521057,
      "learning_rate": 9.436396587532297e-06,
      "loss": 0.3098,
      "step": 178
    },
    {
      "epoch": 32.72727272727273,
      "grad_norm": 0.3551454544067383,
      "learning_rate": 9.414737964294636e-06,
      "loss": 0.4141,
      "step": 180
    },
    {
      "epoch": 32.90909090909091,
      "eval_loss": 0.46264728903770447,
      "eval_runtime": 21.0065,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 181
    },
    {
      "epoch": 33.09090909090909,
      "grad_norm": 0.365192174911499,
      "learning_rate": 9.392696821198488e-06,
      "loss": 0.2879,
      "step": 182
    },
    {
      "epoch": 33.45454545454545,
      "grad_norm": 0.38976219296455383,
      "learning_rate": 9.370275068023097e-06,
      "loss": 0.3467,
      "step": 184
    },
    {
      "epoch": 33.81818181818182,
      "grad_norm": 0.3077632486820221,
      "learning_rate": 9.347474647526095e-06,
      "loss": 0.3693,
      "step": 186
    },
    {
      "epoch": 34.0,
      "eval_loss": 0.4562326967716217,
      "eval_runtime": 21.0019,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 187
    },
    {
      "epoch": 34.18181818181818,
      "grad_norm": 0.209530308842659,
      "learning_rate": 9.324297535275156e-06,
      "loss": 0.1788,
      "step": 188
    },
    {
      "epoch": 34.54545454545455,
      "grad_norm": 0.36394768953323364,
      "learning_rate": 9.30074573947683e-06,
      "loss": 0.3143,
      "step": 190
    },
    {
      "epoch": 34.90909090909091,
      "grad_norm": 0.41356202960014343,
      "learning_rate": 9.276821300802535e-06,
      "loss": 0.3666,
      "step": 192
    },
    {
      "epoch": 34.90909090909091,
      "eval_loss": 0.45260027050971985,
      "eval_runtime": 21.0113,
      "eval_samples_per_second": 1.142,
      "eval_steps_per_second": 1.142,
      "step": 192
    },
    {
      "epoch": 35.27272727272727,
      "grad_norm": 0.42078325152397156,
      "learning_rate": 9.25252629221175e-06,
      "loss": 0.468,
      "step": 194
    },
    {
      "epoch": 35.63636363636363,
      "grad_norm": 0.30819612741470337,
      "learning_rate": 9.227862818772392e-06,
      "loss": 0.1892,
      "step": 196
    },
    {
      "epoch": 36.0,
      "grad_norm": 0.3855637013912201,
      "learning_rate": 9.202833017478421e-06,
      "loss": 0.3605,
      "step": 198
    },
    {
      "epoch": 36.0,
      "eval_loss": 0.4506441354751587,
      "eval_runtime": 20.9992,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 198
    },
    {
      "epoch": 36.36363636363637,
      "grad_norm": 0.39254796504974365,
      "learning_rate": 9.177439057064684e-06,
      "loss": 0.3432,
      "step": 200
    },
    {
      "epoch": 36.72727272727273,
      "grad_norm": 0.3540774881839752,
      "learning_rate": 9.151683137818989e-06,
      "loss": 0.2923,
      "step": 202
    },
    {
      "epoch": 36.90909090909091,
      "eval_loss": 0.44742465019226074,
      "eval_runtime": 21.0039,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 203
    },
    {
      "epoch": 37.09090909090909,
      "grad_norm": 0.4445912539958954,
      "learning_rate": 9.125567491391476e-06,
      "loss": 0.3135,
      "step": 204
    },
    {
      "epoch": 37.45454545454545,
      "grad_norm": 0.39482685923576355,
      "learning_rate": 9.099094380601244e-06,
      "loss": 0.2407,
      "step": 206
    },
    {
      "epoch": 37.81818181818182,
      "grad_norm": 0.3445771634578705,
      "learning_rate": 9.072266099240286e-06,
      "loss": 0.3422,
      "step": 208
    },
    {
      "epoch": 38.0,
      "eval_loss": 0.4443068504333496,
      "eval_runtime": 21.0039,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 209
    },
    {
      "epoch": 38.18181818181818,
      "grad_norm": 0.42600157856941223,
      "learning_rate": 9.045084971874738e-06,
      "loss": 0.2673,
      "step": 210
    },
    {
      "epoch": 38.54545454545455,
      "grad_norm": 0.4967671036720276,
      "learning_rate": 9.017553353643479e-06,
      "loss": 0.3177,
      "step": 212
    },
    {
      "epoch": 38.90909090909091,
      "grad_norm": 0.39350730180740356,
      "learning_rate": 8.989673630054044e-06,
      "loss": 0.3489,
      "step": 214
    },
    {
      "epoch": 38.90909090909091,
      "eval_loss": 0.44272685050964355,
      "eval_runtime": 21.0007,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 214
    },
    {
      "epoch": 39.27272727272727,
      "grad_norm": 0.27645283937454224,
      "learning_rate": 8.961448216775955e-06,
      "loss": 0.161,
      "step": 216
    },
    {
      "epoch": 39.63636363636363,
      "grad_norm": 0.615763783454895,
      "learning_rate": 8.932879559431392e-06,
      "loss": 0.2562,
      "step": 218
    },
    {
      "epoch": 40.0,
      "grad_norm": 0.491910845041275,
      "learning_rate": 8.903970133383297e-06,
      "loss": 0.3737,
      "step": 220
    },
    {
      "epoch": 40.0,
      "eval_loss": 0.443479984998703,
      "eval_runtime": 21.0049,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 220
    },
    {
      "epoch": 40.36363636363637,
      "grad_norm": 0.47909313440322876,
      "learning_rate": 8.874722443520898e-06,
      "loss": 0.2508,
      "step": 222
    },
    {
      "epoch": 40.72727272727273,
      "grad_norm": 0.5516977906227112,
      "learning_rate": 8.845139024042664e-06,
      "loss": 0.2804,
      "step": 224
    },
    {
      "epoch": 40.90909090909091,
      "eval_loss": 0.443651407957077,
      "eval_runtime": 21.007,
      "eval_samples_per_second": 1.142,
      "eval_steps_per_second": 1.142,
      "step": 225
    },
    {
      "epoch": 41.09090909090909,
      "grad_norm": 0.5438214540481567,
      "learning_rate": 8.815222438236726e-06,
      "loss": 0.3221,
      "step": 226
    },
    {
      "epoch": 41.45454545454545,
      "grad_norm": 0.49092897772789,
      "learning_rate": 8.784975278258783e-06,
      "loss": 0.2601,
      "step": 228
    },
    {
      "epoch": 41.81818181818182,
      "grad_norm": 0.5809181332588196,
      "learning_rate": 8.754400164907496e-06,
      "loss": 0.3212,
      "step": 230
    },
    {
      "epoch": 42.0,
      "eval_loss": 0.44689202308654785,
      "eval_runtime": 21.001,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 231
    },
    {
      "epoch": 42.18181818181818,
      "grad_norm": 0.5147470831871033,
      "learning_rate": 8.723499747397415e-06,
      "loss": 0.1686,
      "step": 232
    },
    {
      "epoch": 42.54545454545455,
      "grad_norm": 0.7372314929962158,
      "learning_rate": 8.692276703129421e-06,
      "loss": 0.2733,
      "step": 234
    },
    {
      "epoch": 42.90909090909091,
      "grad_norm": 0.4814988374710083,
      "learning_rate": 8.660733737458751e-06,
      "loss": 0.2322,
      "step": 236
    },
    {
      "epoch": 42.90909090909091,
      "eval_loss": 0.4480229318141937,
      "eval_runtime": 21.0124,
      "eval_samples_per_second": 1.142,
      "eval_steps_per_second": 1.142,
      "step": 236
    },
    {
      "epoch": 43.27272727272727,
      "grad_norm": 0.7270358800888062,
      "learning_rate": 8.628873583460593e-06,
      "loss": 0.2312,
      "step": 238
    },
    {
      "epoch": 43.63636363636363,
      "grad_norm": 0.6802980303764343,
      "learning_rate": 8.596699001693257e-06,
      "loss": 0.2551,
      "step": 240
    },
    {
      "epoch": 44.0,
      "grad_norm": 0.6301150321960449,
      "learning_rate": 8.564212779959003e-06,
      "loss": 0.2569,
      "step": 242
    },
    {
      "epoch": 44.0,
      "eval_loss": 0.4507286548614502,
      "eval_runtime": 21.0008,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 242
    },
    {
      "epoch": 44.36363636363637,
      "grad_norm": 0.6666437983512878,
      "learning_rate": 8.531417733062476e-06,
      "loss": 0.2299,
      "step": 244
    },
    {
      "epoch": 44.72727272727273,
      "grad_norm": 0.6867101788520813,
      "learning_rate": 8.498316702566828e-06,
      "loss": 0.2501,
      "step": 246
    },
    {
      "epoch": 44.90909090909091,
      "eval_loss": 0.4543609619140625,
      "eval_runtime": 21.005,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 247
    },
    {
      "epoch": 45.09090909090909,
      "grad_norm": 0.571156919002533,
      "learning_rate": 8.464912556547486e-06,
      "loss": 0.2035,
      "step": 248
    },
    {
      "epoch": 45.45454545454545,
      "grad_norm": 0.761677086353302,
      "learning_rate": 8.43120818934367e-06,
      "loss": 0.1717,
      "step": 250
    },
    {
      "epoch": 45.81818181818182,
      "grad_norm": 0.6928288340568542,
      "learning_rate": 8.397206521307584e-06,
      "loss": 0.2247,
      "step": 252
    },
    {
      "epoch": 46.0,
      "eval_loss": 0.4641251266002655,
      "eval_runtime": 20.9989,
      "eval_samples_per_second": 1.143,
      "eval_steps_per_second": 1.143,
      "step": 253
    },
    {
      "epoch": 46.0,
      "step": 253,
      "total_flos": 3.15960483767124e+17,
      "train_loss": 0.7967259936888699,
      "train_runtime": 11647.481,
      "train_samples_per_second": 1.133,
      "train_steps_per_second": 0.064
    }
  ],
  "logging_steps": 2,
  "max_steps": 750,
  "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": 3.15960483767124e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}