{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.03593531643042523,
  "eval_steps": 10,
  "global_step": 90,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0003992812936713915,
      "grad_norm": 11.861096382141113,
      "learning_rate": 2.4999051176541087e-07,
      "loss": 3.5277,
      "step": 1
    },
    {
      "epoch": 0.0003992812936713915,
      "eval_loss": 3.8839635848999023,
      "eval_runtime": 287.9309,
      "eval_samples_per_second": 0.889,
      "eval_steps_per_second": 0.222,
      "step": 1
    },
    {
      "epoch": 0.000798562587342783,
      "grad_norm": 11.42729377746582,
      "learning_rate": 2.4998101670529935e-07,
      "loss": 3.4112,
      "step": 2
    },
    {
      "epoch": 0.0011978438810141745,
      "grad_norm": 9.308838844299316,
      "learning_rate": 2.499715148122976e-07,
      "loss": 3.8903,
      "step": 3
    },
    {
      "epoch": 0.001597125174685566,
      "grad_norm": 5.668716907501221,
      "learning_rate": 2.499620060790273e-07,
      "loss": 3.3354,
      "step": 4
    },
    {
      "epoch": 0.0019964064683569574,
      "grad_norm": 7.672369003295898,
      "learning_rate": 2.499524904980996e-07,
      "loss": 3.6227,
      "step": 5
    },
    {
      "epoch": 0.002395687762028349,
      "grad_norm": 4.239442825317383,
      "learning_rate": 2.4994296806211477e-07,
      "loss": 3.105,
      "step": 6
    },
    {
      "epoch": 0.0027949690556997404,
      "grad_norm": 3.5887157917022705,
      "learning_rate": 2.4993343876366256e-07,
      "loss": 2.9517,
      "step": 7
    },
    {
      "epoch": 0.003194250349371132,
      "grad_norm": 3.6663546562194824,
      "learning_rate": 2.4992390259532204e-07,
      "loss": 2.9232,
      "step": 8
    },
    {
      "epoch": 0.0035935316430425234,
      "grad_norm": 1.8050780296325684,
      "learning_rate": 2.4991435954966147e-07,
      "loss": 2.7544,
      "step": 9
    },
    {
      "epoch": 0.003992812936713915,
      "grad_norm": 1.7229609489440918,
      "learning_rate": 2.499048096192385e-07,
      "loss": 2.5745,
      "step": 10
    },
    {
      "epoch": 0.003992812936713915,
      "eval_loss": 2.8815834522247314,
      "eval_runtime": 290.9736,
      "eval_samples_per_second": 0.88,
      "eval_steps_per_second": 0.22,
      "step": 10
    },
    {
      "epoch": 0.004392094230385306,
      "grad_norm": 2.123410701751709,
      "learning_rate": 2.4989525279659993e-07,
      "loss": 3.0969,
      "step": 11
    },
    {
      "epoch": 0.004791375524056698,
      "grad_norm": 1.229800820350647,
      "learning_rate": 2.4988568907428204e-07,
      "loss": 2.8336,
      "step": 12
    },
    {
      "epoch": 0.005190656817728089,
      "grad_norm": 0.9092569351196289,
      "learning_rate": 2.4987611844481e-07,
      "loss": 2.7336,
      "step": 13
    },
    {
      "epoch": 0.005589938111399481,
      "grad_norm": 1.8590848445892334,
      "learning_rate": 2.498665409006984e-07,
      "loss": 2.8651,
      "step": 14
    },
    {
      "epoch": 0.005989219405070872,
      "grad_norm": 1.1895430088043213,
      "learning_rate": 2.4985695643445094e-07,
      "loss": 2.5926,
      "step": 15
    },
    {
      "epoch": 0.006388500698742264,
      "grad_norm": 0.8829777836799622,
      "learning_rate": 2.4984736503856043e-07,
      "loss": 2.5937,
      "step": 16
    },
    {
      "epoch": 0.006787781992413655,
      "grad_norm": 0.7874141335487366,
      "learning_rate": 2.498377667055089e-07,
      "loss": 2.7383,
      "step": 17
    },
    {
      "epoch": 0.007187063286085047,
      "grad_norm": 0.8703321814537048,
      "learning_rate": 2.498281614277675e-07,
      "loss": 2.7594,
      "step": 18
    },
    {
      "epoch": 0.007586344579756438,
      "grad_norm": 1.0839616060256958,
      "learning_rate": 2.4981854919779646e-07,
      "loss": 2.5339,
      "step": 19
    },
    {
      "epoch": 0.00798562587342783,
      "grad_norm": 0.9549999237060547,
      "learning_rate": 2.4980893000804503e-07,
      "loss": 2.8402,
      "step": 20
    },
    {
      "epoch": 0.00798562587342783,
      "eval_loss": 2.659167766571045,
      "eval_runtime": 290.5049,
      "eval_samples_per_second": 0.881,
      "eval_steps_per_second": 0.22,
      "step": 20
    },
    {
      "epoch": 0.008384907167099222,
      "grad_norm": 1.1994329690933228,
      "learning_rate": 2.497993038509517e-07,
      "loss": 2.7734,
      "step": 21
    },
    {
      "epoch": 0.008784188460770613,
      "grad_norm": 1.2531729936599731,
      "learning_rate": 2.497896707189437e-07,
      "loss": 2.4779,
      "step": 22
    },
    {
      "epoch": 0.009183469754442005,
      "grad_norm": 0.8631401658058167,
      "learning_rate": 2.4978003060443764e-07,
      "loss": 2.7183,
      "step": 23
    },
    {
      "epoch": 0.009582751048113396,
      "grad_norm": 0.6393347978591919,
      "learning_rate": 2.4977038349983886e-07,
      "loss": 2.5175,
      "step": 24
    },
    {
      "epoch": 0.009982032341784788,
      "grad_norm": 0.7909420728683472,
      "learning_rate": 2.497607293975418e-07,
      "loss": 2.7034,
      "step": 25
    },
    {
      "epoch": 0.010381313635456179,
      "grad_norm": 0.8058650493621826,
      "learning_rate": 2.4975106828992983e-07,
      "loss": 2.688,
      "step": 26
    },
    {
      "epoch": 0.010780594929127571,
      "grad_norm": 0.7194178700447083,
      "learning_rate": 2.497414001693753e-07,
      "loss": 2.415,
      "step": 27
    },
    {
      "epoch": 0.011179876222798962,
      "grad_norm": 0.7787066698074341,
      "learning_rate": 2.4973172502823945e-07,
      "loss": 2.4489,
      "step": 28
    },
    {
      "epoch": 0.011579157516470354,
      "grad_norm": 0.7317251563072205,
      "learning_rate": 2.497220428588724e-07,
      "loss": 2.6261,
      "step": 29
    },
    {
      "epoch": 0.011978438810141745,
      "grad_norm": 0.643670916557312,
      "learning_rate": 2.4971235365361323e-07,
      "loss": 2.3531,
      "step": 30
    },
    {
      "epoch": 0.011978438810141745,
      "eval_loss": 2.5479750633239746,
      "eval_runtime": 288.2263,
      "eval_samples_per_second": 0.888,
      "eval_steps_per_second": 0.222,
      "step": 30
    },
    {
      "epoch": 0.012377720103813137,
      "grad_norm": 1.17843496799469,
      "learning_rate": 2.4970265740478976e-07,
      "loss": 2.6157,
      "step": 31
    },
    {
      "epoch": 0.012777001397484528,
      "grad_norm": 0.7883862257003784,
      "learning_rate": 2.496929541047188e-07,
      "loss": 2.5571,
      "step": 32
    },
    {
      "epoch": 0.01317628269115592,
      "grad_norm": 0.7334747314453125,
      "learning_rate": 2.4968324374570584e-07,
      "loss": 2.6712,
      "step": 33
    },
    {
      "epoch": 0.01357556398482731,
      "grad_norm": 0.8014112114906311,
      "learning_rate": 2.496735263200453e-07,
      "loss": 2.774,
      "step": 34
    },
    {
      "epoch": 0.013974845278498703,
      "grad_norm": 0.6167514324188232,
      "learning_rate": 2.496638018200202e-07,
      "loss": 2.6011,
      "step": 35
    },
    {
      "epoch": 0.014374126572170094,
      "grad_norm": 0.5952684879302979,
      "learning_rate": 2.4965407023790255e-07,
      "loss": 2.325,
      "step": 36
    },
    {
      "epoch": 0.014773407865841486,
      "grad_norm": 0.56275874376297,
      "learning_rate": 2.4964433156595294e-07,
      "loss": 2.4661,
      "step": 37
    },
    {
      "epoch": 0.015172689159512876,
      "grad_norm": 0.6017836332321167,
      "learning_rate": 2.496345857964207e-07,
      "loss": 2.4421,
      "step": 38
    },
    {
      "epoch": 0.015571970453184269,
      "grad_norm": 0.5634634494781494,
      "learning_rate": 2.4962483292154404e-07,
      "loss": 2.7153,
      "step": 39
    },
    {
      "epoch": 0.01597125174685566,
      "grad_norm": 0.6215035915374756,
      "learning_rate": 2.496150729335494e-07,
      "loss": 2.325,
      "step": 40
    },
    {
      "epoch": 0.01597125174685566,
      "eval_loss": 2.4711108207702637,
      "eval_runtime": 289.8366,
      "eval_samples_per_second": 0.883,
      "eval_steps_per_second": 0.221,
      "step": 40
    },
    {
      "epoch": 0.01637053304052705,
      "grad_norm": 0.5762554407119751,
      "learning_rate": 2.4960530582465247e-07,
      "loss": 2.5452,
      "step": 41
    },
    {
      "epoch": 0.016769814334198444,
      "grad_norm": 0.5260322690010071,
      "learning_rate": 2.49595531587057e-07,
      "loss": 2.5862,
      "step": 42
    },
    {
      "epoch": 0.017169095627869835,
      "grad_norm": 0.5681716799736023,
      "learning_rate": 2.495857502129558e-07,
      "loss": 2.7202,
      "step": 43
    },
    {
      "epoch": 0.017568376921541225,
      "grad_norm": 0.6019343733787537,
      "learning_rate": 2.4957596169453007e-07,
      "loss": 2.5932,
      "step": 44
    },
    {
      "epoch": 0.017967658215212616,
      "grad_norm": 0.5380938053131104,
      "learning_rate": 2.495661660239497e-07,
      "loss": 2.5838,
      "step": 45
    },
    {
      "epoch": 0.01836693950888401,
      "grad_norm": 0.5226567983627319,
      "learning_rate": 2.4955636319337287e-07,
      "loss": 2.6499,
      "step": 46
    },
    {
      "epoch": 0.0187662208025554,
      "grad_norm": 0.5076204538345337,
      "learning_rate": 2.495465531949466e-07,
      "loss": 2.4434,
      "step": 47
    },
    {
      "epoch": 0.01916550209622679,
      "grad_norm": 0.8190023303031921,
      "learning_rate": 2.495367360208062e-07,
      "loss": 2.4931,
      "step": 48
    },
    {
      "epoch": 0.019564783389898182,
      "grad_norm": 0.6802453994750977,
      "learning_rate": 2.495269116630757e-07,
      "loss": 2.5744,
      "step": 49
    },
    {
      "epoch": 0.019964064683569576,
      "grad_norm": 0.4862814247608185,
      "learning_rate": 2.495170801138674e-07,
      "loss": 2.2362,
      "step": 50
    },
    {
      "epoch": 0.019964064683569576,
      "eval_loss": 2.414149522781372,
      "eval_runtime": 289.0292,
      "eval_samples_per_second": 0.886,
      "eval_steps_per_second": 0.221,
      "step": 50
    },
    {
      "epoch": 0.020363345977240967,
      "grad_norm": 0.5404508113861084,
      "learning_rate": 2.495072413652821e-07,
      "loss": 2.75,
      "step": 51
    },
    {
      "epoch": 0.020762627270912357,
      "grad_norm": 0.4755939543247223,
      "learning_rate": 2.4949739540940906e-07,
      "loss": 2.2318,
      "step": 52
    },
    {
      "epoch": 0.021161908564583748,
      "grad_norm": 0.7830485701560974,
      "learning_rate": 2.494875422383259e-07,
      "loss": 2.6674,
      "step": 53
    },
    {
      "epoch": 0.021561189858255142,
      "grad_norm": 0.5113694667816162,
      "learning_rate": 2.494776818440987e-07,
      "loss": 2.5452,
      "step": 54
    },
    {
      "epoch": 0.021960471151926533,
      "grad_norm": 0.597758412361145,
      "learning_rate": 2.494678142187818e-07,
      "loss": 2.116,
      "step": 55
    },
    {
      "epoch": 0.022359752445597923,
      "grad_norm": 0.5865931510925293,
      "learning_rate": 2.49457939354418e-07,
      "loss": 2.3269,
      "step": 56
    },
    {
      "epoch": 0.022759033739269314,
      "grad_norm": 0.6509243845939636,
      "learning_rate": 2.4944805724303826e-07,
      "loss": 2.4119,
      "step": 57
    },
    {
      "epoch": 0.023158315032940708,
      "grad_norm": 0.6174994111061096,
      "learning_rate": 2.4943816787666206e-07,
      "loss": 2.3669,
      "step": 58
    },
    {
      "epoch": 0.0235575963266121,
      "grad_norm": 0.5243375301361084,
      "learning_rate": 2.494282712472969e-07,
      "loss": 2.4973,
      "step": 59
    },
    {
      "epoch": 0.02395687762028349,
      "grad_norm": 0.5543302893638611,
      "learning_rate": 2.494183673469388e-07,
      "loss": 2.471,
      "step": 60
    },
    {
      "epoch": 0.02395687762028349,
      "eval_loss": 2.3702573776245117,
      "eval_runtime": 289.9081,
      "eval_samples_per_second": 0.883,
      "eval_steps_per_second": 0.221,
      "step": 60
    },
    {
      "epoch": 0.02435615891395488,
      "grad_norm": 0.5695878863334656,
      "learning_rate": 2.494084561675717e-07,
      "loss": 2.5227,
      "step": 61
    },
    {
      "epoch": 0.024755440207626274,
      "grad_norm": 0.48718804121017456,
      "learning_rate": 2.493985377011682e-07,
      "loss": 2.4296,
      "step": 62
    },
    {
      "epoch": 0.025154721501297665,
      "grad_norm": 0.5036266446113586,
      "learning_rate": 2.4938861193968864e-07,
      "loss": 2.4357,
      "step": 63
    },
    {
      "epoch": 0.025554002794969055,
      "grad_norm": 0.47645705938339233,
      "learning_rate": 2.4937867887508175e-07,
      "loss": 2.1584,
      "step": 64
    },
    {
      "epoch": 0.025953284088640446,
      "grad_norm": 0.4842926263809204,
      "learning_rate": 2.493687384992844e-07,
      "loss": 2.3637,
      "step": 65
    },
    {
      "epoch": 0.02635256538231184,
      "grad_norm": 0.6649777889251709,
      "learning_rate": 2.493587908042216e-07,
      "loss": 2.3034,
      "step": 66
    },
    {
      "epoch": 0.02675184667598323,
      "grad_norm": 0.49681320786476135,
      "learning_rate": 2.493488357818062e-07,
      "loss": 2.4476,
      "step": 67
    },
    {
      "epoch": 0.02715112796965462,
      "grad_norm": 0.514624297618866,
      "learning_rate": 2.4933887342393975e-07,
      "loss": 2.457,
      "step": 68
    },
    {
      "epoch": 0.02755040926332601,
      "grad_norm": 0.662382185459137,
      "learning_rate": 2.493289037225111e-07,
      "loss": 2.646,
      "step": 69
    },
    {
      "epoch": 0.027949690556997406,
      "grad_norm": 0.4702717661857605,
      "learning_rate": 2.493189266693978e-07,
      "loss": 2.3054,
      "step": 70
    },
    {
      "epoch": 0.027949690556997406,
      "eval_loss": 2.331839084625244,
      "eval_runtime": 290.0915,
      "eval_samples_per_second": 0.882,
      "eval_steps_per_second": 0.221,
      "step": 70
    },
    {
      "epoch": 0.028348971850668796,
      "grad_norm": 0.5980565547943115,
      "learning_rate": 2.493089422564649e-07,
      "loss": 2.1874,
      "step": 71
    },
    {
      "epoch": 0.028748253144340187,
      "grad_norm": 0.5449456572532654,
      "learning_rate": 2.4929895047556574e-07,
      "loss": 2.3876,
      "step": 72
    },
    {
      "epoch": 0.029147534438011578,
      "grad_norm": 0.4739711582660675,
      "learning_rate": 2.492889513185416e-07,
      "loss": 2.3522,
      "step": 73
    },
    {
      "epoch": 0.02954681573168297,
      "grad_norm": 0.5783733129501343,
      "learning_rate": 2.492789447772216e-07,
      "loss": 2.4153,
      "step": 74
    },
    {
      "epoch": 0.029946097025354362,
      "grad_norm": 0.429364949464798,
      "learning_rate": 2.4926893084342294e-07,
      "loss": 2.3046,
      "step": 75
    },
    {
      "epoch": 0.030345378319025753,
      "grad_norm": 0.48500293493270874,
      "learning_rate": 2.4925890950895056e-07,
      "loss": 2.2062,
      "step": 76
    },
    {
      "epoch": 0.030744659612697144,
      "grad_norm": 0.4081466495990753,
      "learning_rate": 2.492488807655974e-07,
      "loss": 2.2804,
      "step": 77
    },
    {
      "epoch": 0.031143940906368538,
      "grad_norm": 0.5915509462356567,
      "learning_rate": 2.4923884460514423e-07,
      "loss": 2.304,
      "step": 78
    },
    {
      "epoch": 0.031543222200039925,
      "grad_norm": 0.5798116326332092,
      "learning_rate": 2.4922880101935963e-07,
      "loss": 2.3379,
      "step": 79
    },
    {
      "epoch": 0.03194250349371132,
      "grad_norm": 0.49052485823631287,
      "learning_rate": 2.4921874999999997e-07,
      "loss": 2.3184,
      "step": 80
    },
    {
      "epoch": 0.03194250349371132,
      "eval_loss": 2.2982735633850098,
      "eval_runtime": 290.0312,
      "eval_samples_per_second": 0.883,
      "eval_steps_per_second": 0.221,
      "step": 80
    },
    {
      "epoch": 0.03234178478738271,
      "grad_norm": 0.5482540726661682,
      "learning_rate": 2.492086915388095e-07,
      "loss": 2.2594,
      "step": 81
    },
    {
      "epoch": 0.0327410660810541,
      "grad_norm": 0.44985291361808777,
      "learning_rate": 2.4919862562752036e-07,
      "loss": 2.3427,
      "step": 82
    },
    {
      "epoch": 0.033140347374725494,
      "grad_norm": 0.6427104473114014,
      "learning_rate": 2.4918855225785203e-07,
      "loss": 2.128,
      "step": 83
    },
    {
      "epoch": 0.03353962866839689,
      "grad_norm": 0.5451573133468628,
      "learning_rate": 2.491784714215121e-07,
      "loss": 2.3032,
      "step": 84
    },
    {
      "epoch": 0.033938909962068275,
      "grad_norm": 0.5538087487220764,
      "learning_rate": 2.4916838311019566e-07,
      "loss": 2.161,
      "step": 85
    },
    {
      "epoch": 0.03433819125573967,
      "grad_norm": 0.6961225271224976,
      "learning_rate": 2.4915828731558555e-07,
      "loss": 2.3407,
      "step": 86
    },
    {
      "epoch": 0.03473747254941106,
      "grad_norm": 0.5596035718917847,
      "learning_rate": 2.4914818402935235e-07,
      "loss": 2.4305,
      "step": 87
    },
    {
      "epoch": 0.03513675384308245,
      "grad_norm": 0.5701755881309509,
      "learning_rate": 2.4913807324315406e-07,
      "loss": 2.2315,
      "step": 88
    },
    {
      "epoch": 0.035536035136753845,
      "grad_norm": 0.5389382839202881,
      "learning_rate": 2.491279549486365e-07,
      "loss": 2.3222,
      "step": 89
    },
    {
      "epoch": 0.03593531643042523,
      "grad_norm": 0.4433860778808594,
      "learning_rate": 2.491178291374329e-07,
      "loss": 2.0449,
      "step": 90
    },
    {
      "epoch": 0.03593531643042523,
      "eval_loss": 2.2697765827178955,
      "eval_runtime": 290.8528,
      "eval_samples_per_second": 0.88,
      "eval_steps_per_second": 0.22,
      "step": 90
    }
  ],
  "logging_steps": 1,
  "max_steps": 2504,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.902605301599437e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}