{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.4644549763033177,
  "eval_steps": 6,
  "global_step": 130,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.018957345971563982,
      "grad_norm": 1.21875,
      "learning_rate": 2e-05,
      "loss": 2.0442,
      "step": 1
    },
    {
      "epoch": 0.018957345971563982,
      "eval_loss": 2.073391914367676,
      "eval_runtime": 6.8335,
      "eval_samples_per_second": 52.096,
      "eval_steps_per_second": 13.024,
      "step": 1
    },
    {
      "epoch": 0.037914691943127965,
      "grad_norm": 1.40625,
      "learning_rate": 4e-05,
      "loss": 2.1817,
      "step": 2
    },
    {
      "epoch": 0.05687203791469194,
      "grad_norm": 1.3125,
      "learning_rate": 6e-05,
      "loss": 2.0671,
      "step": 3
    },
    {
      "epoch": 0.07582938388625593,
      "grad_norm": 1.5,
      "learning_rate": 8e-05,
      "loss": 1.9289,
      "step": 4
    },
    {
      "epoch": 0.0947867298578199,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0001,
      "loss": 1.7705,
      "step": 5
    },
    {
      "epoch": 0.11374407582938388,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00012,
      "loss": 1.449,
      "step": 6
    },
    {
      "epoch": 0.11374407582938388,
      "eval_loss": 1.2773768901824951,
      "eval_runtime": 7.1055,
      "eval_samples_per_second": 50.102,
      "eval_steps_per_second": 12.526,
      "step": 6
    },
    {
      "epoch": 0.13270142180094788,
      "grad_norm": 1.125,
      "learning_rate": 0.00014,
      "loss": 1.3112,
      "step": 7
    },
    {
      "epoch": 0.15165876777251186,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00016,
      "loss": 1.2196,
      "step": 8
    },
    {
      "epoch": 0.17061611374407584,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00018,
      "loss": 1.1166,
      "step": 9
    },
    {
      "epoch": 0.1895734597156398,
      "grad_norm": 0.8125,
      "learning_rate": 0.0002,
      "loss": 1.0836,
      "step": 10
    },
    {
      "epoch": 0.20853080568720378,
      "grad_norm": 0.4609375,
      "learning_rate": 0.00019997685019798912,
      "loss": 0.898,
      "step": 11
    },
    {
      "epoch": 0.22748815165876776,
      "grad_norm": 0.376953125,
      "learning_rate": 0.00019990741151022301,
      "loss": 0.8548,
      "step": 12
    },
    {
      "epoch": 0.22748815165876776,
      "eval_loss": 0.9005841612815857,
      "eval_runtime": 6.5385,
      "eval_samples_per_second": 54.447,
      "eval_steps_per_second": 13.612,
      "step": 12
    },
    {
      "epoch": 0.24644549763033174,
      "grad_norm": 0.384765625,
      "learning_rate": 0.00019979171608653924,
      "loss": 0.8835,
      "step": 13
    },
    {
      "epoch": 0.26540284360189575,
      "grad_norm": 0.326171875,
      "learning_rate": 0.00019962981749346078,
      "loss": 0.8374,
      "step": 14
    },
    {
      "epoch": 0.2843601895734597,
      "grad_norm": 0.32421875,
      "learning_rate": 0.0001994217906893952,
      "loss": 0.8658,
      "step": 15
    },
    {
      "epoch": 0.3033175355450237,
      "grad_norm": 0.326171875,
      "learning_rate": 0.000199167731989929,
      "loss": 0.8232,
      "step": 16
    },
    {
      "epoch": 0.3222748815165877,
      "grad_norm": 0.234375,
      "learning_rate": 0.00019886775902323405,
      "loss": 0.7541,
      "step": 17
    },
    {
      "epoch": 0.3412322274881517,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.00019852201067560606,
      "loss": 0.8561,
      "step": 18
    },
    {
      "epoch": 0.3412322274881517,
      "eval_loss": 0.7924264669418335,
      "eval_runtime": 7.7645,
      "eval_samples_per_second": 45.85,
      "eval_steps_per_second": 11.462,
      "step": 18
    },
    {
      "epoch": 0.36018957345971564,
      "grad_norm": 0.267578125,
      "learning_rate": 0.00019813064702716094,
      "loss": 0.7909,
      "step": 19
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 0.3046875,
      "learning_rate": 0.0001976938492777182,
      "loss": 0.7657,
      "step": 20
    },
    {
      "epoch": 0.3981042654028436,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00019721181966290613,
      "loss": 0.7214,
      "step": 21
    },
    {
      "epoch": 0.41706161137440756,
      "grad_norm": 0.33203125,
      "learning_rate": 0.00019668478136052774,
      "loss": 0.7511,
      "step": 22
    },
    {
      "epoch": 0.43601895734597157,
      "grad_norm": 0.35546875,
      "learning_rate": 0.0001961129783872301,
      "loss": 0.7159,
      "step": 23
    },
    {
      "epoch": 0.4549763033175355,
      "grad_norm": 0.279296875,
      "learning_rate": 0.00019549667548552556,
      "loss": 0.744,
      "step": 24
    },
    {
      "epoch": 0.4549763033175355,
      "eval_loss": 0.717645525932312,
      "eval_runtime": 6.6339,
      "eval_samples_per_second": 53.664,
      "eval_steps_per_second": 13.416,
      "step": 24
    },
    {
      "epoch": 0.47393364928909953,
      "grad_norm": 0.259765625,
      "learning_rate": 0.00019483615800121716,
      "loss": 0.6753,
      "step": 25
    },
    {
      "epoch": 0.4928909952606635,
      "grad_norm": 0.26953125,
      "learning_rate": 0.00019413173175128473,
      "loss": 0.7068,
      "step": 26
    },
    {
      "epoch": 0.5118483412322274,
      "grad_norm": 0.263671875,
      "learning_rate": 0.0001933837228822925,
      "loss": 0.6545,
      "step": 27
    },
    {
      "epoch": 0.5308056872037915,
      "grad_norm": 0.29296875,
      "learning_rate": 0.000192592477719385,
      "loss": 0.7141,
      "step": 28
    },
    {
      "epoch": 0.5497630331753555,
      "grad_norm": 0.259765625,
      "learning_rate": 0.00019175836260593938,
      "loss": 0.6713,
      "step": 29
    },
    {
      "epoch": 0.5687203791469194,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0001908817637339503,
      "loss": 0.6752,
      "step": 30
    },
    {
      "epoch": 0.5687203791469194,
      "eval_loss": 0.6603197455406189,
      "eval_runtime": 7.3719,
      "eval_samples_per_second": 48.291,
      "eval_steps_per_second": 12.073,
      "step": 30
    },
    {
      "epoch": 0.5876777251184834,
      "grad_norm": 0.326171875,
      "learning_rate": 0.00018996308696522433,
      "loss": 0.6389,
      "step": 31
    },
    {
      "epoch": 0.6066350710900474,
      "grad_norm": 0.3046875,
      "learning_rate": 0.00018900275764346768,
      "loss": 0.6711,
      "step": 32
    },
    {
      "epoch": 0.6255924170616114,
      "grad_norm": 0.322265625,
      "learning_rate": 0.00018800122039735358,
      "loss": 0.6097,
      "step": 33
    },
    {
      "epoch": 0.6445497630331753,
      "grad_norm": 0.322265625,
      "learning_rate": 0.0001869589389346611,
      "loss": 0.6204,
      "step": 34
    },
    {
      "epoch": 0.6635071090047393,
      "grad_norm": 0.34765625,
      "learning_rate": 0.00018587639582758031,
      "loss": 0.6288,
      "step": 35
    },
    {
      "epoch": 0.6824644549763034,
      "grad_norm": 0.34375,
      "learning_rate": 0.00018475409228928312,
      "loss": 0.5908,
      "step": 36
    },
    {
      "epoch": 0.6824644549763034,
      "eval_loss": 0.6117041707038879,
      "eval_runtime": 7.5585,
      "eval_samples_per_second": 47.099,
      "eval_steps_per_second": 11.775,
      "step": 36
    },
    {
      "epoch": 0.7014218009478673,
      "grad_norm": 0.369140625,
      "learning_rate": 0.0001835925479418637,
      "loss": 0.5851,
      "step": 37
    },
    {
      "epoch": 0.7203791469194313,
      "grad_norm": 0.40625,
      "learning_rate": 0.00018239230057575542,
      "loss": 0.5967,
      "step": 38
    },
    {
      "epoch": 0.7393364928909952,
      "grad_norm": 0.396484375,
      "learning_rate": 0.0001811539059007361,
      "loss": 0.5626,
      "step": 39
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 0.37109375,
      "learning_rate": 0.00017987793728863651,
      "loss": 0.5691,
      "step": 40
    },
    {
      "epoch": 0.7772511848341233,
      "grad_norm": 0.421875,
      "learning_rate": 0.00017856498550787144,
      "loss": 0.5913,
      "step": 41
    },
    {
      "epoch": 0.7962085308056872,
      "grad_norm": 0.40625,
      "learning_rate": 0.00017721565844991643,
      "loss": 0.5229,
      "step": 42
    },
    {
      "epoch": 0.7962085308056872,
      "eval_loss": 0.5701669454574585,
      "eval_runtime": 7.5799,
      "eval_samples_per_second": 46.966,
      "eval_steps_per_second": 11.742,
      "step": 42
    },
    {
      "epoch": 0.8151658767772512,
      "grad_norm": 0.423828125,
      "learning_rate": 0.00017583058084785625,
      "loss": 0.5542,
      "step": 43
    },
    {
      "epoch": 0.8341232227488151,
      "grad_norm": 0.466796875,
      "learning_rate": 0.00017441039398713608,
      "loss": 0.5131,
      "step": 44
    },
    {
      "epoch": 0.8530805687203792,
      "grad_norm": 0.41015625,
      "learning_rate": 0.00017295575540864877,
      "loss": 0.5479,
      "step": 45
    },
    {
      "epoch": 0.8720379146919431,
      "grad_norm": 0.416015625,
      "learning_rate": 0.00017146733860429612,
      "loss": 0.5456,
      "step": 46
    },
    {
      "epoch": 0.8909952606635071,
      "grad_norm": 0.400390625,
      "learning_rate": 0.0001699458327051647,
      "loss": 0.5475,
      "step": 47
    },
    {
      "epoch": 0.909952606635071,
      "grad_norm": 0.419921875,
      "learning_rate": 0.00016839194216246108,
      "loss": 0.558,
      "step": 48
    },
    {
      "epoch": 0.909952606635071,
      "eval_loss": 0.528121829032898,
      "eval_runtime": 7.5358,
      "eval_samples_per_second": 47.241,
      "eval_steps_per_second": 11.81,
      "step": 48
    },
    {
      "epoch": 0.9289099526066351,
      "grad_norm": 0.443359375,
      "learning_rate": 0.00016680638642135336,
      "loss": 0.4805,
      "step": 49
    },
    {
      "epoch": 0.9478672985781991,
      "grad_norm": 0.435546875,
      "learning_rate": 0.00016518989958787126,
      "loss": 0.4929,
      "step": 50
    },
    {
      "epoch": 0.966824644549763,
      "grad_norm": 0.423828125,
      "learning_rate": 0.00016354323008901776,
      "loss": 0.5415,
      "step": 51
    },
    {
      "epoch": 0.985781990521327,
      "grad_norm": 0.453125,
      "learning_rate": 0.00016186714032625035,
      "loss": 0.5038,
      "step": 52
    },
    {
      "epoch": 1.004739336492891,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00016016240632249224,
      "loss": 0.4915,
      "step": 53
    },
    {
      "epoch": 1.0236966824644549,
      "grad_norm": 0.5234375,
      "learning_rate": 0.00015842981736283686,
      "loss": 0.4343,
      "step": 54
    },
    {
      "epoch": 1.0236966824644549,
      "eval_loss": 0.4752185344696045,
      "eval_runtime": 7.8454,
      "eval_samples_per_second": 45.377,
      "eval_steps_per_second": 11.344,
      "step": 54
    },
    {
      "epoch": 1.042654028436019,
      "grad_norm": 0.57421875,
      "learning_rate": 0.00015667017562911176,
      "loss": 0.4282,
      "step": 55
    },
    {
      "epoch": 1.061611374407583,
      "grad_norm": 0.51953125,
      "learning_rate": 0.00015488429582847192,
      "loss": 0.396,
      "step": 56
    },
    {
      "epoch": 1.080568720379147,
      "grad_norm": 0.53515625,
      "learning_rate": 0.00015307300481619333,
      "loss": 0.4685,
      "step": 57
    },
    {
      "epoch": 1.099526066350711,
      "grad_norm": 0.59375,
      "learning_rate": 0.0001512371412128424,
      "loss": 0.428,
      "step": 58
    },
    {
      "epoch": 1.1184834123222749,
      "grad_norm": 0.5234375,
      "learning_rate": 0.00014937755501599772,
      "loss": 0.4002,
      "step": 59
    },
    {
      "epoch": 1.1374407582938388,
      "grad_norm": 0.498046875,
      "learning_rate": 0.00014749510720670506,
      "loss": 0.4039,
      "step": 60
    },
    {
      "epoch": 1.1374407582938388,
      "eval_loss": 0.4151807725429535,
      "eval_runtime": 7.7375,
      "eval_samples_per_second": 46.01,
      "eval_steps_per_second": 11.502,
      "step": 60
    },
    {
      "epoch": 1.1563981042654028,
      "grad_norm": 0.4140625,
      "learning_rate": 0.00014559066935084588,
      "loss": 0.3903,
      "step": 61
    },
    {
      "epoch": 1.1753554502369667,
      "grad_norm": 0.2890625,
      "learning_rate": 0.0001436651231956064,
      "loss": 0.3936,
      "step": 62
    },
    {
      "epoch": 1.1943127962085307,
      "grad_norm": 0.314453125,
      "learning_rate": 0.00014171936026123168,
      "loss": 0.3759,
      "step": 63
    },
    {
      "epoch": 1.2132701421800949,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0001397542814282556,
      "loss": 0.3949,
      "step": 64
    },
    {
      "epoch": 1.2322274881516588,
      "grad_norm": 2.8125,
      "learning_rate": 0.0001377707965203965,
      "loss": 0.3979,
      "step": 65
    },
    {
      "epoch": 1.2511848341232228,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0001357698238833126,
      "loss": 0.3744,
      "step": 66
    },
    {
      "epoch": 1.2511848341232228,
      "eval_loss": 0.422529935836792,
      "eval_runtime": 7.4489,
      "eval_samples_per_second": 47.792,
      "eval_steps_per_second": 11.948,
      "step": 66
    },
    {
      "epoch": 1.2701421800947867,
      "grad_norm": 2.421875,
      "learning_rate": 0.00013375228995941133,
      "loss": 0.3851,
      "step": 67
    },
    {
      "epoch": 1.2890995260663507,
      "grad_norm": 0.515625,
      "learning_rate": 0.00013171912885891063,
      "loss": 0.3524,
      "step": 68
    },
    {
      "epoch": 1.3080568720379147,
      "grad_norm": 0.421875,
      "learning_rate": 0.00012967128192734902,
      "loss": 0.3688,
      "step": 69
    },
    {
      "epoch": 1.3270142180094786,
      "grad_norm": 0.2421875,
      "learning_rate": 0.00012760969730974694,
      "loss": 0.3338,
      "step": 70
    },
    {
      "epoch": 1.3459715639810428,
      "grad_norm": 0.212890625,
      "learning_rate": 0.0001255353295116187,
      "loss": 0.3206,
      "step": 71
    },
    {
      "epoch": 1.3649289099526065,
      "grad_norm": 0.2109375,
      "learning_rate": 0.00012344913895704097,
      "loss": 0.3313,
      "step": 72
    },
    {
      "epoch": 1.3649289099526065,
      "eval_loss": 0.3852485120296478,
      "eval_runtime": 7.69,
      "eval_samples_per_second": 46.294,
      "eval_steps_per_second": 11.573,
      "step": 72
    },
    {
      "epoch": 1.3838862559241707,
      "grad_norm": 0.23046875,
      "learning_rate": 0.00012135209154397962,
      "loss": 0.3514,
      "step": 73
    },
    {
      "epoch": 1.4028436018957346,
      "grad_norm": 0.22265625,
      "learning_rate": 0.000119245158197083,
      "loss": 0.3608,
      "step": 74
    },
    {
      "epoch": 1.4218009478672986,
      "grad_norm": 0.2060546875,
      "learning_rate": 0.00011712931441814776,
      "loss": 0.3361,
      "step": 75
    },
    {
      "epoch": 1.4407582938388626,
      "grad_norm": 0.1884765625,
      "learning_rate": 0.00011500553983446527,
      "loss": 0.3095,
      "step": 76
    },
    {
      "epoch": 1.4597156398104265,
      "grad_norm": 0.2060546875,
      "learning_rate": 0.0001128748177452581,
      "loss": 0.318,
      "step": 77
    },
    {
      "epoch": 1.4786729857819905,
      "grad_norm": 0.2265625,
      "learning_rate": 0.00011073813466641632,
      "loss": 0.374,
      "step": 78
    },
    {
      "epoch": 1.4786729857819905,
      "eval_loss": 0.3739808201789856,
      "eval_runtime": 7.5558,
      "eval_samples_per_second": 47.116,
      "eval_steps_per_second": 11.779,
      "step": 78
    },
    {
      "epoch": 1.4976303317535544,
      "grad_norm": 0.216796875,
      "learning_rate": 0.00010859647987374467,
      "loss": 0.3174,
      "step": 79
    },
    {
      "epoch": 1.5165876777251186,
      "grad_norm": 0.251953125,
      "learning_rate": 0.00010645084494493165,
      "loss": 0.3281,
      "step": 80
    },
    {
      "epoch": 1.5355450236966823,
      "grad_norm": 0.2236328125,
      "learning_rate": 0.00010430222330045304,
      "loss": 0.3235,
      "step": 81
    },
    {
      "epoch": 1.5545023696682465,
      "grad_norm": 0.2236328125,
      "learning_rate": 0.00010215160974362223,
      "loss": 0.3064,
      "step": 82
    },
    {
      "epoch": 1.5734597156398105,
      "grad_norm": 0.25390625,
      "learning_rate": 0.0001,
      "loss": 0.3278,
      "step": 83
    },
    {
      "epoch": 1.5924170616113744,
      "grad_norm": 0.2373046875,
      "learning_rate": 9.784839025637778e-05,
      "loss": 0.3246,
      "step": 84
    },
    {
      "epoch": 1.5924170616113744,
      "eval_loss": 0.3657410442829132,
      "eval_runtime": 7.41,
      "eval_samples_per_second": 48.043,
      "eval_steps_per_second": 12.011,
      "step": 84
    },
    {
      "epoch": 1.6113744075829384,
      "grad_norm": 0.23046875,
      "learning_rate": 9.569777669954694e-05,
      "loss": 0.3257,
      "step": 85
    },
    {
      "epoch": 1.6303317535545023,
      "grad_norm": 0.19921875,
      "learning_rate": 9.354915505506839e-05,
      "loss": 0.3247,
      "step": 86
    },
    {
      "epoch": 1.6492890995260665,
      "grad_norm": 0.234375,
      "learning_rate": 9.140352012625537e-05,
      "loss": 0.3372,
      "step": 87
    },
    {
      "epoch": 1.6682464454976302,
      "grad_norm": 0.2216796875,
      "learning_rate": 8.92618653335837e-05,
      "loss": 0.2964,
      "step": 88
    },
    {
      "epoch": 1.6872037914691944,
      "grad_norm": 0.2255859375,
      "learning_rate": 8.712518225474191e-05,
      "loss": 0.3604,
      "step": 89
    },
    {
      "epoch": 1.7061611374407581,
      "grad_norm": 0.2197265625,
      "learning_rate": 8.499446016553474e-05,
      "loss": 0.3392,
      "step": 90
    },
    {
      "epoch": 1.7061611374407581,
      "eval_loss": 0.3591402769088745,
      "eval_runtime": 7.6868,
      "eval_samples_per_second": 46.313,
      "eval_steps_per_second": 11.578,
      "step": 90
    },
    {
      "epoch": 1.7251184834123223,
      "grad_norm": 0.2197265625,
      "learning_rate": 8.287068558185225e-05,
      "loss": 0.3367,
      "step": 91
    },
    {
      "epoch": 1.7440758293838863,
      "grad_norm": 0.2138671875,
      "learning_rate": 8.075484180291701e-05,
      "loss": 0.2959,
      "step": 92
    },
    {
      "epoch": 1.7630331753554502,
      "grad_norm": 0.2119140625,
      "learning_rate": 7.864790845602039e-05,
      "loss": 0.3239,
      "step": 93
    },
    {
      "epoch": 1.7819905213270142,
      "grad_norm": 0.2060546875,
      "learning_rate": 7.655086104295904e-05,
      "loss": 0.2984,
      "step": 94
    },
    {
      "epoch": 1.8009478672985781,
      "grad_norm": 0.236328125,
      "learning_rate": 7.446467048838131e-05,
      "loss": 0.3346,
      "step": 95
    },
    {
      "epoch": 1.8199052132701423,
      "grad_norm": 0.255859375,
      "learning_rate": 7.239030269025311e-05,
      "loss": 0.3309,
      "step": 96
    },
    {
      "epoch": 1.8199052132701423,
      "eval_loss": 0.3504635691642761,
      "eval_runtime": 8.1061,
      "eval_samples_per_second": 43.918,
      "eval_steps_per_second": 10.979,
      "step": 96
    },
    {
      "epoch": 1.838862559241706,
      "grad_norm": 0.2333984375,
      "learning_rate": 7.032871807265096e-05,
      "loss": 0.3313,
      "step": 97
    },
    {
      "epoch": 1.8578199052132702,
      "grad_norm": 0.267578125,
      "learning_rate": 6.82808711410894e-05,
      "loss": 0.32,
      "step": 98
    },
    {
      "epoch": 1.876777251184834,
      "grad_norm": 0.267578125,
      "learning_rate": 6.624771004058868e-05,
      "loss": 0.4074,
      "step": 99
    },
    {
      "epoch": 1.8957345971563981,
      "grad_norm": 0.2041015625,
      "learning_rate": 6.423017611668745e-05,
      "loss": 0.2994,
      "step": 100
    },
    {
      "epoch": 1.914691943127962,
      "grad_norm": 0.2216796875,
      "learning_rate": 6.22292034796035e-05,
      "loss": 0.3063,
      "step": 101
    },
    {
      "epoch": 1.933649289099526,
      "grad_norm": 0.2138671875,
      "learning_rate": 6.024571857174443e-05,
      "loss": 0.3621,
      "step": 102
    },
    {
      "epoch": 1.933649289099526,
      "eval_loss": 0.343742311000824,
      "eval_runtime": 7.3443,
      "eval_samples_per_second": 48.473,
      "eval_steps_per_second": 12.118,
      "step": 102
    },
    {
      "epoch": 1.95260663507109,
      "grad_norm": 0.265625,
      "learning_rate": 5.828063973876834e-05,
      "loss": 0.302,
      "step": 103
    },
    {
      "epoch": 1.971563981042654,
      "grad_norm": 0.232421875,
      "learning_rate": 5.633487680439361e-05,
      "loss": 0.3248,
      "step": 104
    },
    {
      "epoch": 1.9905213270142181,
      "grad_norm": 0.201171875,
      "learning_rate": 5.440933064915414e-05,
      "loss": 0.2853,
      "step": 105
    },
    {
      "epoch": 2.009478672985782,
      "grad_norm": 0.2109375,
      "learning_rate": 5.2504892793295e-05,
      "loss": 0.2788,
      "step": 106
    },
    {
      "epoch": 2.028436018957346,
      "grad_norm": 0.2177734375,
      "learning_rate": 5.062244498400228e-05,
      "loss": 0.2791,
      "step": 107
    },
    {
      "epoch": 2.0473933649289098,
      "grad_norm": 0.208984375,
      "learning_rate": 4.876285878715764e-05,
      "loss": 0.2819,
      "step": 108
    },
    {
      "epoch": 2.0473933649289098,
      "eval_loss": 0.3415816128253937,
      "eval_runtime": 7.5308,
      "eval_samples_per_second": 47.273,
      "eval_steps_per_second": 11.818,
      "step": 108
    },
    {
      "epoch": 2.066350710900474,
      "grad_norm": 0.224609375,
      "learning_rate": 4.6926995183806644e-05,
      "loss": 0.2547,
      "step": 109
    },
    {
      "epoch": 2.085308056872038,
      "grad_norm": 0.22265625,
      "learning_rate": 4.5115704171528105e-05,
      "loss": 0.2911,
      "step": 110
    },
    {
      "epoch": 2.104265402843602,
      "grad_norm": 0.220703125,
      "learning_rate": 4.332982437088825e-05,
      "loss": 0.2756,
      "step": 111
    },
    {
      "epoch": 2.123222748815166,
      "grad_norm": 0.267578125,
      "learning_rate": 4.1570182637163155e-05,
      "loss": 0.2825,
      "step": 112
    },
    {
      "epoch": 2.1421800947867298,
      "grad_norm": 0.26171875,
      "learning_rate": 3.9837593677507726e-05,
      "loss": 0.2358,
      "step": 113
    },
    {
      "epoch": 2.161137440758294,
      "grad_norm": 0.25390625,
      "learning_rate": 3.813285967374969e-05,
      "loss": 0.2672,
      "step": 114
    },
    {
      "epoch": 2.161137440758294,
      "eval_loss": 0.3413563072681427,
      "eval_runtime": 7.5937,
      "eval_samples_per_second": 46.881,
      "eval_steps_per_second": 11.72,
      "step": 114
    },
    {
      "epoch": 2.1800947867298577,
      "grad_norm": 0.2255859375,
      "learning_rate": 3.645676991098227e-05,
      "loss": 0.2753,
      "step": 115
    },
    {
      "epoch": 2.199052132701422,
      "grad_norm": 0.23046875,
      "learning_rate": 3.4810100412128747e-05,
      "loss": 0.2588,
      "step": 116
    },
    {
      "epoch": 2.2180094786729856,
      "grad_norm": 0.2578125,
      "learning_rate": 3.319361357864663e-05,
      "loss": 0.2964,
      "step": 117
    },
    {
      "epoch": 2.2369668246445498,
      "grad_norm": 0.2734375,
      "learning_rate": 3.160805783753897e-05,
      "loss": 0.2438,
      "step": 118
    },
    {
      "epoch": 2.2559241706161135,
      "grad_norm": 0.2451171875,
      "learning_rate": 3.005416729483531e-05,
      "loss": 0.262,
      "step": 119
    },
    {
      "epoch": 2.2748815165876777,
      "grad_norm": 0.2109375,
      "learning_rate": 2.853266139570391e-05,
      "loss": 0.2284,
      "step": 120
    },
    {
      "epoch": 2.2748815165876777,
      "eval_loss": 0.33752232789993286,
      "eval_runtime": 8.0472,
      "eval_samples_per_second": 44.239,
      "eval_steps_per_second": 11.06,
      "step": 120
    },
    {
      "epoch": 2.293838862559242,
      "grad_norm": 0.197265625,
      "learning_rate": 2.7044244591351232e-05,
      "loss": 0.2287,
      "step": 121
    },
    {
      "epoch": 2.3127962085308056,
      "grad_norm": 0.2333984375,
      "learning_rate": 2.5589606012863963e-05,
      "loss": 0.247,
      "step": 122
    },
    {
      "epoch": 2.3317535545023698,
      "grad_norm": 0.240234375,
      "learning_rate": 2.4169419152143768e-05,
      "loss": 0.2797,
      "step": 123
    },
    {
      "epoch": 2.3507109004739335,
      "grad_norm": 0.2333984375,
      "learning_rate": 2.2784341550083576e-05,
      "loss": 0.2577,
      "step": 124
    },
    {
      "epoch": 2.3696682464454977,
      "grad_norm": 0.23828125,
      "learning_rate": 2.1435014492128547e-05,
      "loss": 0.2792,
      "step": 125
    },
    {
      "epoch": 2.3886255924170614,
      "grad_norm": 0.2412109375,
      "learning_rate": 2.0122062711363532e-05,
      "loss": 0.2836,
      "step": 126
    },
    {
      "epoch": 2.3886255924170614,
      "eval_loss": 0.3353247344493866,
      "eval_runtime": 6.6458,
      "eval_samples_per_second": 53.567,
      "eval_steps_per_second": 13.392,
      "step": 126
    },
    {
      "epoch": 2.4075829383886256,
      "grad_norm": 0.2333984375,
      "learning_rate": 1.8846094099263912e-05,
      "loss": 0.2694,
      "step": 127
    },
    {
      "epoch": 2.4265402843601898,
      "grad_norm": 0.216796875,
      "learning_rate": 1.7607699424244585e-05,
      "loss": 0.2666,
      "step": 128
    },
    {
      "epoch": 2.4454976303317535,
      "grad_norm": 0.232421875,
      "learning_rate": 1.6407452058136296e-05,
      "loss": 0.2747,
      "step": 129
    },
    {
      "epoch": 2.4644549763033177,
      "grad_norm": 0.2470703125,
      "learning_rate": 1.5245907710716911e-05,
      "loss": 0.3057,
      "step": 130
    }
  ],
  "logging_steps": 1,
  "max_steps": 156,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 26,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.922942328719278e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}