{
  "best_metric": 3.510607957839966,
  "best_model_checkpoint": "./Phi-2_PT_QA_1/checkpoint-200",
  "epoch": 0.0022854530910753055,
  "eval_steps": 1,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.1427265455376528e-05,
      "grad_norm": 0.7554183006286621,
      "learning_rate": 2.5e-05,
      "loss": 7.6787,
      "step": 1
    },
    {
      "epoch": 1.1427265455376528e-05,
      "eval_loss": 7.217936038970947,
      "eval_runtime": 3184.807,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 1
    },
    {
      "epoch": 2.2854530910753055e-05,
      "grad_norm": 0.809639573097229,
      "learning_rate": 2.4874371859296484e-05,
      "loss": 7.5645,
      "step": 2
    },
    {
      "epoch": 2.2854530910753055e-05,
      "eval_loss": 7.216185569763184,
      "eval_runtime": 3195.682,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 2
    },
    {
      "epoch": 3.428179636612959e-05,
      "grad_norm": 0.5419557094573975,
      "learning_rate": 2.4748743718592964e-05,
      "loss": 5.6844,
      "step": 3
    },
    {
      "epoch": 3.428179636612959e-05,
      "eval_loss": 7.211589336395264,
      "eval_runtime": 3196.3014,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 3
    },
    {
      "epoch": 4.570906182150611e-05,
      "grad_norm": 0.6964669823646545,
      "learning_rate": 2.462311557788945e-05,
      "loss": 6.8741,
      "step": 4
    },
    {
      "epoch": 4.570906182150611e-05,
      "eval_loss": 7.204798221588135,
      "eval_runtime": 3194.9301,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 4
    },
    {
      "epoch": 5.713632727688264e-05,
      "grad_norm": 0.670452356338501,
      "learning_rate": 2.449748743718593e-05,
      "loss": 6.6605,
      "step": 5
    },
    {
      "epoch": 5.713632727688264e-05,
      "eval_loss": 7.197990894317627,
      "eval_runtime": 3194.2197,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 5
    },
    {
      "epoch": 6.856359273225918e-05,
      "grad_norm": 0.8575164675712585,
      "learning_rate": 2.4371859296482413e-05,
      "loss": 8.0797,
      "step": 6
    },
    {
      "epoch": 6.856359273225918e-05,
      "eval_loss": 7.1901774406433105,
      "eval_runtime": 3196.5055,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 6
    },
    {
      "epoch": 7.99908581876357e-05,
      "grad_norm": 0.46564584970474243,
      "learning_rate": 2.4246231155778896e-05,
      "loss": 4.6497,
      "step": 7
    },
    {
      "epoch": 7.99908581876357e-05,
      "eval_loss": 7.182745933532715,
      "eval_runtime": 3195.2045,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 7
    },
    {
      "epoch": 9.141812364301222e-05,
      "grad_norm": 0.8245170712471008,
      "learning_rate": 2.4120603015075376e-05,
      "loss": 7.1603,
      "step": 8
    },
    {
      "epoch": 9.141812364301222e-05,
      "eval_loss": 7.173906326293945,
      "eval_runtime": 3196.5564,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 8
    },
    {
      "epoch": 0.00010284538909838876,
      "grad_norm": 0.7266675233840942,
      "learning_rate": 2.3994974874371863e-05,
      "loss": 5.9704,
      "step": 9
    },
    {
      "epoch": 0.00010284538909838876,
      "eval_loss": 7.165711879730225,
      "eval_runtime": 3195.6693,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 9
    },
    {
      "epoch": 0.00011427265455376528,
      "grad_norm": 0.8064714670181274,
      "learning_rate": 2.3869346733668342e-05,
      "loss": 6.5571,
      "step": 10
    },
    {
      "epoch": 0.00011427265455376528,
      "eval_loss": 7.157066822052002,
      "eval_runtime": 3196.6621,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 10
    },
    {
      "epoch": 0.00012569992000914182,
      "grad_norm": 1.181254267692566,
      "learning_rate": 2.3743718592964825e-05,
      "loss": 8.2521,
      "step": 11
    },
    {
      "epoch": 0.00012569992000914182,
      "eval_loss": 7.1473798751831055,
      "eval_runtime": 3195.7223,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 11
    },
    {
      "epoch": 0.00013712718546451835,
      "grad_norm": 0.8531734347343445,
      "learning_rate": 2.361809045226131e-05,
      "loss": 6.5084,
      "step": 12
    },
    {
      "epoch": 0.00013712718546451835,
      "eval_loss": 7.1375651359558105,
      "eval_runtime": 3197.1154,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 12
    },
    {
      "epoch": 0.00014855445091989486,
      "grad_norm": 0.8652932643890381,
      "learning_rate": 2.3492462311557788e-05,
      "loss": 6.1662,
      "step": 13
    },
    {
      "epoch": 0.00014855445091989486,
      "eval_loss": 7.127073287963867,
      "eval_runtime": 3199.1778,
      "eval_samples_per_second": 1.719,
      "eval_steps_per_second": 1.719,
      "step": 13
    },
    {
      "epoch": 0.0001599817163752714,
      "grad_norm": 1.108816146850586,
      "learning_rate": 2.3366834170854275e-05,
      "loss": 7.9397,
      "step": 14
    },
    {
      "epoch": 0.0001599817163752714,
      "eval_loss": 7.116166114807129,
      "eval_runtime": 3199.5106,
      "eval_samples_per_second": 1.719,
      "eval_steps_per_second": 1.719,
      "step": 14
    },
    {
      "epoch": 0.00017140898183064793,
      "grad_norm": 1.090276837348938,
      "learning_rate": 2.3241206030150754e-05,
      "loss": 7.066,
      "step": 15
    },
    {
      "epoch": 0.00017140898183064793,
      "eval_loss": 7.104613780975342,
      "eval_runtime": 3197.1218,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 15
    },
    {
      "epoch": 0.00018283624728602444,
      "grad_norm": 1.5131642818450928,
      "learning_rate": 2.3115577889447238e-05,
      "loss": 8.6329,
      "step": 16
    },
    {
      "epoch": 0.00018283624728602444,
      "eval_loss": 7.092124938964844,
      "eval_runtime": 3198.3437,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 16
    },
    {
      "epoch": 0.00019426351274140098,
      "grad_norm": 1.1908084154129028,
      "learning_rate": 2.298994974874372e-05,
      "loss": 7.2422,
      "step": 17
    },
    {
      "epoch": 0.00019426351274140098,
      "eval_loss": 7.078921794891357,
      "eval_runtime": 3196.3444,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 17
    },
    {
      "epoch": 0.00020569077819677752,
      "grad_norm": 1.157004714012146,
      "learning_rate": 2.28643216080402e-05,
      "loss": 6.7972,
      "step": 18
    },
    {
      "epoch": 0.00020569077819677752,
      "eval_loss": 7.065500736236572,
      "eval_runtime": 3196.895,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 18
    },
    {
      "epoch": 0.00021711804365215405,
      "grad_norm": 1.2554625272750854,
      "learning_rate": 2.2738693467336687e-05,
      "loss": 6.9076,
      "step": 19
    },
    {
      "epoch": 0.00021711804365215405,
      "eval_loss": 7.051235675811768,
      "eval_runtime": 3197.6043,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 19
    },
    {
      "epoch": 0.00022854530910753056,
      "grad_norm": 1.1883549690246582,
      "learning_rate": 2.2613065326633167e-05,
      "loss": 6.5687,
      "step": 20
    },
    {
      "epoch": 0.00022854530910753056,
      "eval_loss": 7.036637306213379,
      "eval_runtime": 3196.0972,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 20
    },
    {
      "epoch": 0.0002399725745629071,
      "grad_norm": 1.328796148300171,
      "learning_rate": 2.248743718592965e-05,
      "loss": 6.7253,
      "step": 21
    },
    {
      "epoch": 0.0002399725745629071,
      "eval_loss": 7.021302700042725,
      "eval_runtime": 3197.9644,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 21
    },
    {
      "epoch": 0.00025139984001828363,
      "grad_norm": 1.2712606191635132,
      "learning_rate": 2.2361809045226133e-05,
      "loss": 6.1315,
      "step": 22
    },
    {
      "epoch": 0.00025139984001828363,
      "eval_loss": 7.005300521850586,
      "eval_runtime": 3198.4581,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 22
    },
    {
      "epoch": 0.00026282710547366017,
      "grad_norm": 1.5367112159729004,
      "learning_rate": 2.2236180904522613e-05,
      "loss": 7.0915,
      "step": 23
    },
    {
      "epoch": 0.00026282710547366017,
      "eval_loss": 6.98922872543335,
      "eval_runtime": 3194.9885,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 23
    },
    {
      "epoch": 0.0002742543709290367,
      "grad_norm": 1.4399423599243164,
      "learning_rate": 2.21105527638191e-05,
      "loss": 6.5237,
      "step": 24
    },
    {
      "epoch": 0.0002742543709290367,
      "eval_loss": 6.972210884094238,
      "eval_runtime": 3194.7666,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 24
    },
    {
      "epoch": 0.0002856816363844132,
      "grad_norm": 0.9845247864723206,
      "learning_rate": 2.198492462311558e-05,
      "loss": 5.0536,
      "step": 25
    },
    {
      "epoch": 0.0002856816363844132,
      "eval_loss": 6.955036163330078,
      "eval_runtime": 3197.7715,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 25
    },
    {
      "epoch": 0.0002971089018397897,
      "grad_norm": 1.1501060724258423,
      "learning_rate": 2.1859296482412062e-05,
      "loss": 5.3398,
      "step": 26
    },
    {
      "epoch": 0.0002971089018397897,
      "eval_loss": 6.937135696411133,
      "eval_runtime": 3193.8534,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 26
    },
    {
      "epoch": 0.00030853616729516626,
      "grad_norm": 0.6898478865623474,
      "learning_rate": 2.1733668341708545e-05,
      "loss": 4.2246,
      "step": 27
    },
    {
      "epoch": 0.00030853616729516626,
      "eval_loss": 6.919322490692139,
      "eval_runtime": 3194.157,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 27
    },
    {
      "epoch": 0.0003199634327505428,
      "grad_norm": 1.9928979873657227,
      "learning_rate": 2.1608040201005025e-05,
      "loss": 8.0199,
      "step": 28
    },
    {
      "epoch": 0.0003199634327505428,
      "eval_loss": 6.901232719421387,
      "eval_runtime": 3194.869,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 28
    },
    {
      "epoch": 0.00033139069820591933,
      "grad_norm": 1.6013647317886353,
      "learning_rate": 2.1482412060301508e-05,
      "loss": 6.4213,
      "step": 29
    },
    {
      "epoch": 0.00033139069820591933,
      "eval_loss": 6.882281303405762,
      "eval_runtime": 3196.4271,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 29
    },
    {
      "epoch": 0.00034281796366129587,
      "grad_norm": 0.6780699491500854,
      "learning_rate": 2.135678391959799e-05,
      "loss": 4.0142,
      "step": 30
    },
    {
      "epoch": 0.00034281796366129587,
      "eval_loss": 6.86350154876709,
      "eval_runtime": 3195.3922,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 30
    },
    {
      "epoch": 0.0003542452291166724,
      "grad_norm": 1.243430256843567,
      "learning_rate": 2.1231155778894474e-05,
      "loss": 5.1075,
      "step": 31
    },
    {
      "epoch": 0.0003542452291166724,
      "eval_loss": 6.844424724578857,
      "eval_runtime": 3194.7916,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 31
    },
    {
      "epoch": 0.0003656724945720489,
      "grad_norm": 1.993019700050354,
      "learning_rate": 2.1105527638190957e-05,
      "loss": 6.6294,
      "step": 32
    },
    {
      "epoch": 0.0003656724945720489,
      "eval_loss": 6.824941635131836,
      "eval_runtime": 3194.2906,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 32
    },
    {
      "epoch": 0.0003770997600274254,
      "grad_norm": 2.094905138015747,
      "learning_rate": 2.0979899497487437e-05,
      "loss": 7.1042,
      "step": 33
    },
    {
      "epoch": 0.0003770997600274254,
      "eval_loss": 6.804728031158447,
      "eval_runtime": 3194.5706,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 33
    },
    {
      "epoch": 0.00038852702548280196,
      "grad_norm": 1.7997703552246094,
      "learning_rate": 2.085427135678392e-05,
      "loss": 6.6241,
      "step": 34
    },
    {
      "epoch": 0.00038852702548280196,
      "eval_loss": 6.784134864807129,
      "eval_runtime": 3194.7504,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 34
    },
    {
      "epoch": 0.0003999542909381785,
      "grad_norm": 1.6641287803649902,
      "learning_rate": 2.0728643216080403e-05,
      "loss": 5.6431,
      "step": 35
    },
    {
      "epoch": 0.0003999542909381785,
      "eval_loss": 6.762852191925049,
      "eval_runtime": 3195.8642,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 35
    },
    {
      "epoch": 0.00041138155639355503,
      "grad_norm": 1.9099078178405762,
      "learning_rate": 2.0603015075376886e-05,
      "loss": 6.4051,
      "step": 36
    },
    {
      "epoch": 0.00041138155639355503,
      "eval_loss": 6.741316795349121,
      "eval_runtime": 3192.3732,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 36
    },
    {
      "epoch": 0.00042280882184893157,
      "grad_norm": 1.958937406539917,
      "learning_rate": 2.047738693467337e-05,
      "loss": 6.8499,
      "step": 37
    },
    {
      "epoch": 0.00042280882184893157,
      "eval_loss": 6.719349384307861,
      "eval_runtime": 3195.1382,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 37
    },
    {
      "epoch": 0.0004342360873043081,
      "grad_norm": 1.903361201286316,
      "learning_rate": 2.035175879396985e-05,
      "loss": 6.1026,
      "step": 38
    },
    {
      "epoch": 0.0004342360873043081,
      "eval_loss": 6.697410583496094,
      "eval_runtime": 3195.9588,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 38
    },
    {
      "epoch": 0.0004456633527596846,
      "grad_norm": 2.3142635822296143,
      "learning_rate": 2.0226130653266332e-05,
      "loss": 6.834,
      "step": 39
    },
    {
      "epoch": 0.0004456633527596846,
      "eval_loss": 6.674659252166748,
      "eval_runtime": 3194.6685,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 39
    },
    {
      "epoch": 0.0004570906182150611,
      "grad_norm": 2.355628252029419,
      "learning_rate": 2.0100502512562815e-05,
      "loss": 7.05,
      "step": 40
    },
    {
      "epoch": 0.0004570906182150611,
      "eval_loss": 6.651575088500977,
      "eval_runtime": 3192.9283,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 40
    },
    {
      "epoch": 0.00046851788367043766,
      "grad_norm": 2.478034496307373,
      "learning_rate": 1.9974874371859298e-05,
      "loss": 7.203,
      "step": 41
    },
    {
      "epoch": 0.00046851788367043766,
      "eval_loss": 6.6284260749816895,
      "eval_runtime": 3195.3222,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 41
    },
    {
      "epoch": 0.0004799451491258142,
      "grad_norm": 1.591928482055664,
      "learning_rate": 1.984924623115578e-05,
      "loss": 4.5537,
      "step": 42
    },
    {
      "epoch": 0.0004799451491258142,
      "eval_loss": 6.604694366455078,
      "eval_runtime": 3196.8557,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 42
    },
    {
      "epoch": 0.0004913724145811907,
      "grad_norm": 2.615412712097168,
      "learning_rate": 1.972361809045226e-05,
      "loss": 6.8927,
      "step": 43
    },
    {
      "epoch": 0.0004913724145811907,
      "eval_loss": 6.580879211425781,
      "eval_runtime": 3197.1663,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 43
    },
    {
      "epoch": 0.0005027996800365673,
      "grad_norm": 2.0230987071990967,
      "learning_rate": 1.9597989949748744e-05,
      "loss": 6.0638,
      "step": 44
    },
    {
      "epoch": 0.0005027996800365673,
      "eval_loss": 6.556508541107178,
      "eval_runtime": 3196.2452,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 44
    },
    {
      "epoch": 0.0005142269454919438,
      "grad_norm": 2.6316287517547607,
      "learning_rate": 1.9472361809045227e-05,
      "loss": 6.8353,
      "step": 45
    },
    {
      "epoch": 0.0005142269454919438,
      "eval_loss": 6.532052516937256,
      "eval_runtime": 3195.9887,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 45
    },
    {
      "epoch": 0.0005256542109473203,
      "grad_norm": 1.5371630191802979,
      "learning_rate": 1.934673366834171e-05,
      "loss": 4.5244,
      "step": 46
    },
    {
      "epoch": 0.0005256542109473203,
      "eval_loss": 6.507321834564209,
      "eval_runtime": 3196.0239,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 46
    },
    {
      "epoch": 0.0005370814764026968,
      "grad_norm": 1.7422285079956055,
      "learning_rate": 1.9221105527638193e-05,
      "loss": 4.9649,
      "step": 47
    },
    {
      "epoch": 0.0005370814764026968,
      "eval_loss": 6.482240200042725,
      "eval_runtime": 3195.5121,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 47
    },
    {
      "epoch": 0.0005485087418580734,
      "grad_norm": 1.1683274507522583,
      "learning_rate": 1.9095477386934673e-05,
      "loss": 3.7341,
      "step": 48
    },
    {
      "epoch": 0.0005485087418580734,
      "eval_loss": 6.456914901733398,
      "eval_runtime": 3194.9287,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 48
    },
    {
      "epoch": 0.0005599360073134499,
      "grad_norm": 3.5512888431549072,
      "learning_rate": 1.8969849246231156e-05,
      "loss": 8.3966,
      "step": 49
    },
    {
      "epoch": 0.0005599360073134499,
      "eval_loss": 6.431957244873047,
      "eval_runtime": 3193.8838,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 49
    },
    {
      "epoch": 0.0005713632727688264,
      "grad_norm": 1.8522059917449951,
      "learning_rate": 1.884422110552764e-05,
      "loss": 4.837,
      "step": 50
    },
    {
      "epoch": 0.0005713632727688264,
      "eval_loss": 6.406413555145264,
      "eval_runtime": 3195.7335,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 50
    },
    {
      "epoch": 0.000582790538224203,
      "grad_norm": 2.5635313987731934,
      "learning_rate": 1.8718592964824123e-05,
      "loss": 6.2677,
      "step": 51
    },
    {
      "epoch": 0.000582790538224203,
      "eval_loss": 6.380771160125732,
      "eval_runtime": 3193.483,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 51
    },
    {
      "epoch": 0.0005942178036795794,
      "grad_norm": 3.0012006759643555,
      "learning_rate": 1.8592964824120602e-05,
      "loss": 6.3779,
      "step": 52
    },
    {
      "epoch": 0.0005942178036795794,
      "eval_loss": 6.354803085327148,
      "eval_runtime": 3191.7709,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 52
    },
    {
      "epoch": 0.000605645069134956,
      "grad_norm": 2.9274518489837646,
      "learning_rate": 1.8467336683417085e-05,
      "loss": 6.4349,
      "step": 53
    },
    {
      "epoch": 0.000605645069134956,
      "eval_loss": 6.329031467437744,
      "eval_runtime": 3194.6961,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 53
    },
    {
      "epoch": 0.0006170723345903325,
      "grad_norm": 3.2986245155334473,
      "learning_rate": 1.834170854271357e-05,
      "loss": 7.2301,
      "step": 54
    },
    {
      "epoch": 0.0006170723345903325,
      "eval_loss": 6.30263090133667,
      "eval_runtime": 3193.3906,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 54
    },
    {
      "epoch": 0.0006284996000457091,
      "grad_norm": 2.258892297744751,
      "learning_rate": 1.821608040201005e-05,
      "loss": 5.288,
      "step": 55
    },
    {
      "epoch": 0.0006284996000457091,
      "eval_loss": 6.27651834487915,
      "eval_runtime": 3193.9175,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 55
    },
    {
      "epoch": 0.0006399268655010856,
      "grad_norm": 3.4918830394744873,
      "learning_rate": 1.8090452261306535e-05,
      "loss": 7.0831,
      "step": 56
    },
    {
      "epoch": 0.0006399268655010856,
      "eval_loss": 6.249880790710449,
      "eval_runtime": 3195.1438,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 56
    },
    {
      "epoch": 0.0006513541309564621,
      "grad_norm": 3.48417592048645,
      "learning_rate": 1.7964824120603014e-05,
      "loss": 7.1039,
      "step": 57
    },
    {
      "epoch": 0.0006513541309564621,
      "eval_loss": 6.223054885864258,
      "eval_runtime": 3193.8782,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 57
    },
    {
      "epoch": 0.0006627813964118387,
      "grad_norm": 2.575576066970825,
      "learning_rate": 1.7839195979899497e-05,
      "loss": 5.38,
      "step": 58
    },
    {
      "epoch": 0.0006627813964118387,
      "eval_loss": 6.195952892303467,
      "eval_runtime": 3193.8956,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 58
    },
    {
      "epoch": 0.0006742086618672151,
      "grad_norm": 3.869877815246582,
      "learning_rate": 1.771356783919598e-05,
      "loss": 7.1632,
      "step": 59
    },
    {
      "epoch": 0.0006742086618672151,
      "eval_loss": 6.1690497398376465,
      "eval_runtime": 3195.4516,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 59
    },
    {
      "epoch": 0.0006856359273225917,
      "grad_norm": 2.590207815170288,
      "learning_rate": 1.7587939698492464e-05,
      "loss": 5.6661,
      "step": 60
    },
    {
      "epoch": 0.0006856359273225917,
      "eval_loss": 6.141936779022217,
      "eval_runtime": 3196.1077,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 60
    },
    {
      "epoch": 0.0006970631927779682,
      "grad_norm": 3.316432237625122,
      "learning_rate": 1.7462311557788947e-05,
      "loss": 6.4535,
      "step": 61
    },
    {
      "epoch": 0.0006970631927779682,
      "eval_loss": 6.114187717437744,
      "eval_runtime": 3196.6866,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 61
    },
    {
      "epoch": 0.0007084904582333448,
      "grad_norm": 3.1938068866729736,
      "learning_rate": 1.7336683417085427e-05,
      "loss": 6.0447,
      "step": 62
    },
    {
      "epoch": 0.0007084904582333448,
      "eval_loss": 6.08643913269043,
      "eval_runtime": 3195.265,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 62
    },
    {
      "epoch": 0.0007199177236887213,
      "grad_norm": 3.078531265258789,
      "learning_rate": 1.721105527638191e-05,
      "loss": 5.802,
      "step": 63
    },
    {
      "epoch": 0.0007199177236887213,
      "eval_loss": 6.058650016784668,
      "eval_runtime": 3193.9302,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 63
    },
    {
      "epoch": 0.0007313449891440978,
      "grad_norm": 3.1577515602111816,
      "learning_rate": 1.7085427135678393e-05,
      "loss": 5.8267,
      "step": 64
    },
    {
      "epoch": 0.0007313449891440978,
      "eval_loss": 6.030569553375244,
      "eval_runtime": 3193.2589,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 64
    },
    {
      "epoch": 0.0007427722545994744,
      "grad_norm": 3.361140012741089,
      "learning_rate": 1.6959798994974876e-05,
      "loss": 6.1471,
      "step": 65
    },
    {
      "epoch": 0.0007427722545994744,
      "eval_loss": 6.002427577972412,
      "eval_runtime": 3194.7411,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 65
    },
    {
      "epoch": 0.0007541995200548508,
      "grad_norm": 1.4381821155548096,
      "learning_rate": 1.683417085427136e-05,
      "loss": 3.696,
      "step": 66
    },
    {
      "epoch": 0.0007541995200548508,
      "eval_loss": 5.974757671356201,
      "eval_runtime": 3195.246,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 66
    },
    {
      "epoch": 0.0007656267855102274,
      "grad_norm": 3.1451377868652344,
      "learning_rate": 1.670854271356784e-05,
      "loss": 5.9335,
      "step": 67
    },
    {
      "epoch": 0.0007656267855102274,
      "eval_loss": 5.946895599365234,
      "eval_runtime": 3195.4763,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 67
    },
    {
      "epoch": 0.0007770540509656039,
      "grad_norm": 2.85153865814209,
      "learning_rate": 1.6582914572864322e-05,
      "loss": 5.5253,
      "step": 68
    },
    {
      "epoch": 0.0007770540509656039,
      "eval_loss": 5.918909072875977,
      "eval_runtime": 3194.2966,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 68
    },
    {
      "epoch": 0.0007884813164209805,
      "grad_norm": 0.18748074769973755,
      "learning_rate": 1.6457286432160805e-05,
      "loss": 2.4955,
      "step": 69
    },
    {
      "epoch": 0.0007884813164209805,
      "eval_loss": 5.893527507781982,
      "eval_runtime": 3192.8875,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 69
    },
    {
      "epoch": 0.000799908581876357,
      "grad_norm": 3.377441167831421,
      "learning_rate": 1.6331658291457288e-05,
      "loss": 6.0589,
      "step": 70
    },
    {
      "epoch": 0.000799908581876357,
      "eval_loss": 5.8677473068237305,
      "eval_runtime": 3192.1113,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 70
    },
    {
      "epoch": 0.0008113358473317335,
      "grad_norm": 3.5671401023864746,
      "learning_rate": 1.620603015075377e-05,
      "loss": 6.1037,
      "step": 71
    },
    {
      "epoch": 0.0008113358473317335,
      "eval_loss": 5.841763019561768,
      "eval_runtime": 3193.951,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 71
    },
    {
      "epoch": 0.0008227631127871101,
      "grad_norm": 1.8638324737548828,
      "learning_rate": 1.608040201005025e-05,
      "loss": 3.9179,
      "step": 72
    },
    {
      "epoch": 0.0008227631127871101,
      "eval_loss": 5.815106391906738,
      "eval_runtime": 3193.2648,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 72
    },
    {
      "epoch": 0.0008341903782424865,
      "grad_norm": 3.3266711235046387,
      "learning_rate": 1.5954773869346734e-05,
      "loss": 5.3457,
      "step": 73
    },
    {
      "epoch": 0.0008341903782424865,
      "eval_loss": 5.788687229156494,
      "eval_runtime": 3194.6212,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 73
    },
    {
      "epoch": 0.0008456176436978631,
      "grad_norm": 4.605172157287598,
      "learning_rate": 1.5829145728643217e-05,
      "loss": 7.0032,
      "step": 74
    },
    {
      "epoch": 0.0008456176436978631,
      "eval_loss": 5.761895656585693,
      "eval_runtime": 3191.5309,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 74
    },
    {
      "epoch": 0.0008570449091532396,
      "grad_norm": 3.4830474853515625,
      "learning_rate": 1.57035175879397e-05,
      "loss": 5.9244,
      "step": 75
    },
    {
      "epoch": 0.0008570449091532396,
      "eval_loss": 5.734696388244629,
      "eval_runtime": 3191.7239,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 75
    },
    {
      "epoch": 0.0008684721746086162,
      "grad_norm": 3.59964919090271,
      "learning_rate": 1.5577889447236183e-05,
      "loss": 5.4847,
      "step": 76
    },
    {
      "epoch": 0.0008684721746086162,
      "eval_loss": 5.707438945770264,
      "eval_runtime": 3191.9511,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 76
    },
    {
      "epoch": 0.0008798994400639927,
      "grad_norm": 4.7916412353515625,
      "learning_rate": 1.5452261306532663e-05,
      "loss": 7.0196,
      "step": 77
    },
    {
      "epoch": 0.0008798994400639927,
      "eval_loss": 5.680092811584473,
      "eval_runtime": 3192.5862,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 77
    },
    {
      "epoch": 0.0008913267055193692,
      "grad_norm": 3.67484450340271,
      "learning_rate": 1.5326633165829146e-05,
      "loss": 5.7853,
      "step": 78
    },
    {
      "epoch": 0.0008913267055193692,
      "eval_loss": 5.652522087097168,
      "eval_runtime": 3191.0964,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 78
    },
    {
      "epoch": 0.0009027539709747458,
      "grad_norm": 3.265822410583496,
      "learning_rate": 1.5201005025125627e-05,
      "loss": 5.3229,
      "step": 79
    },
    {
      "epoch": 0.0009027539709747458,
      "eval_loss": 5.624815940856934,
      "eval_runtime": 3191.2862,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 79
    },
    {
      "epoch": 0.0009141812364301222,
      "grad_norm": 3.9072275161743164,
      "learning_rate": 1.507537688442211e-05,
      "loss": 5.6212,
      "step": 80
    },
    {
      "epoch": 0.0009141812364301222,
      "eval_loss": 5.597079753875732,
      "eval_runtime": 3190.1642,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 80
    },
    {
      "epoch": 0.0009256085018854988,
      "grad_norm": 3.1518900394439697,
      "learning_rate": 1.4949748743718595e-05,
      "loss": 5.0651,
      "step": 81
    },
    {
      "epoch": 0.0009256085018854988,
      "eval_loss": 5.56900691986084,
      "eval_runtime": 3189.2121,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 81
    },
    {
      "epoch": 0.0009370357673408753,
      "grad_norm": 2.539473295211792,
      "learning_rate": 1.4824120603015077e-05,
      "loss": 4.0285,
      "step": 82
    },
    {
      "epoch": 0.0009370357673408753,
      "eval_loss": 5.540792942047119,
      "eval_runtime": 3188.8344,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 82
    },
    {
      "epoch": 0.0009484630327962519,
      "grad_norm": 4.24508810043335,
      "learning_rate": 1.4698492462311558e-05,
      "loss": 6.0955,
      "step": 83
    },
    {
      "epoch": 0.0009484630327962519,
      "eval_loss": 5.512935638427734,
      "eval_runtime": 3191.1876,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 83
    },
    {
      "epoch": 0.0009598902982516284,
      "grad_norm": 3.5962705612182617,
      "learning_rate": 1.457286432160804e-05,
      "loss": 5.3177,
      "step": 84
    },
    {
      "epoch": 0.0009598902982516284,
      "eval_loss": 5.484694004058838,
      "eval_runtime": 3191.2979,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 84
    },
    {
      "epoch": 0.0009713175637070049,
      "grad_norm": 4.158171653747559,
      "learning_rate": 1.4447236180904523e-05,
      "loss": 5.8912,
      "step": 85
    },
    {
      "epoch": 0.0009713175637070049,
      "eval_loss": 5.456665515899658,
      "eval_runtime": 3192.0268,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 85
    },
    {
      "epoch": 0.0009827448291623814,
      "grad_norm": 4.112974643707275,
      "learning_rate": 1.4321608040201007e-05,
      "loss": 5.9361,
      "step": 86
    },
    {
      "epoch": 0.0009827448291623814,
      "eval_loss": 5.428305149078369,
      "eval_runtime": 3189.75,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 86
    },
    {
      "epoch": 0.000994172094617758,
      "grad_norm": 3.924593925476074,
      "learning_rate": 1.4195979899497489e-05,
      "loss": 5.4892,
      "step": 87
    },
    {
      "epoch": 0.000994172094617758,
      "eval_loss": 5.400221824645996,
      "eval_runtime": 3191.2904,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 87
    },
    {
      "epoch": 0.0010055993600731345,
      "grad_norm": 5.291077613830566,
      "learning_rate": 1.407035175879397e-05,
      "loss": 6.7912,
      "step": 88
    },
    {
      "epoch": 0.0010055993600731345,
      "eval_loss": 5.371959686279297,
      "eval_runtime": 3188.0583,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 88
    },
    {
      "epoch": 0.001017026625528511,
      "grad_norm": 3.935314893722534,
      "learning_rate": 1.3944723618090452e-05,
      "loss": 5.2535,
      "step": 89
    },
    {
      "epoch": 0.001017026625528511,
      "eval_loss": 5.343928337097168,
      "eval_runtime": 3186.5407,
      "eval_samples_per_second": 1.726,
      "eval_steps_per_second": 1.726,
      "step": 89
    },
    {
      "epoch": 0.0010284538909838875,
      "grad_norm": 4.988033771514893,
      "learning_rate": 1.3819095477386935e-05,
      "loss": 6.5576,
      "step": 90
    },
    {
      "epoch": 0.0010284538909838875,
      "eval_loss": 5.315735816955566,
      "eval_runtime": 3190.1379,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 90
    },
    {
      "epoch": 0.001039881156439264,
      "grad_norm": 3.571462631225586,
      "learning_rate": 1.369346733668342e-05,
      "loss": 5.0059,
      "step": 91
    },
    {
      "epoch": 0.001039881156439264,
      "eval_loss": 5.287474632263184,
      "eval_runtime": 3183.8947,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 91
    },
    {
      "epoch": 0.0010513084218946407,
      "grad_norm": 3.0511598587036133,
      "learning_rate": 1.3567839195979901e-05,
      "loss": 4.255,
      "step": 92
    },
    {
      "epoch": 0.0010513084218946407,
      "eval_loss": 5.25911283493042,
      "eval_runtime": 3188.1641,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 92
    },
    {
      "epoch": 0.0010627356873500172,
      "grad_norm": 3.8936984539031982,
      "learning_rate": 1.3442211055276382e-05,
      "loss": 5.0144,
      "step": 93
    },
    {
      "epoch": 0.0010627356873500172,
      "eval_loss": 5.231190204620361,
      "eval_runtime": 3178.9927,
      "eval_samples_per_second": 1.73,
      "eval_steps_per_second": 1.73,
      "step": 93
    },
    {
      "epoch": 0.0010741629528053936,
      "grad_norm": 3.204448699951172,
      "learning_rate": 1.3316582914572864e-05,
      "loss": 4.539,
      "step": 94
    },
    {
      "epoch": 0.0010741629528053936,
      "eval_loss": 5.203569412231445,
      "eval_runtime": 3181.9254,
      "eval_samples_per_second": 1.729,
      "eval_steps_per_second": 1.729,
      "step": 94
    },
    {
      "epoch": 0.0010855902182607701,
      "grad_norm": 5.711158752441406,
      "learning_rate": 1.3190954773869347e-05,
      "loss": 7.1367,
      "step": 95
    },
    {
      "epoch": 0.0010855902182607701,
      "eval_loss": 5.176050662994385,
      "eval_runtime": 3182.616,
      "eval_samples_per_second": 1.728,
      "eval_steps_per_second": 1.728,
      "step": 95
    },
    {
      "epoch": 0.0010970174837161468,
      "grad_norm": 4.6051435470581055,
      "learning_rate": 1.306532663316583e-05,
      "loss": 5.5839,
      "step": 96
    },
    {
      "epoch": 0.0010970174837161468,
      "eval_loss": 5.148005962371826,
      "eval_runtime": 3185.2573,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 96
    },
    {
      "epoch": 0.0011084447491715233,
      "grad_norm": 2.9833264350891113,
      "learning_rate": 1.2939698492462313e-05,
      "loss": 4.1461,
      "step": 97
    },
    {
      "epoch": 0.0011084447491715233,
      "eval_loss": 5.120667934417725,
      "eval_runtime": 3188.7644,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 97
    },
    {
      "epoch": 0.0011198720146268998,
      "grad_norm": 3.4609251022338867,
      "learning_rate": 1.2814070351758795e-05,
      "loss": 4.3763,
      "step": 98
    },
    {
      "epoch": 0.0011198720146268998,
      "eval_loss": 5.093635559082031,
      "eval_runtime": 3186.9357,
      "eval_samples_per_second": 1.726,
      "eval_steps_per_second": 1.726,
      "step": 98
    },
    {
      "epoch": 0.0011312992800822763,
      "grad_norm": 3.9000020027160645,
      "learning_rate": 1.2688442211055276e-05,
      "loss": 4.8055,
      "step": 99
    },
    {
      "epoch": 0.0011312992800822763,
      "eval_loss": 5.06650447845459,
      "eval_runtime": 3189.8792,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 99
    },
    {
      "epoch": 0.0011427265455376528,
      "grad_norm": 4.556962966918945,
      "learning_rate": 1.2562814070351759e-05,
      "loss": 5.1023,
      "step": 100
    },
    {
      "epoch": 0.0011427265455376528,
      "eval_loss": 5.03890323638916,
      "eval_runtime": 3185.537,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 100
    },
    {
      "epoch": 0.0011541538109930295,
      "grad_norm": 3.0480575561523438,
      "learning_rate": 1.2437185929648242e-05,
      "loss": 3.8963,
      "step": 101
    },
    {
      "epoch": 0.0011541538109930295,
      "eval_loss": 5.011804580688477,
      "eval_runtime": 3186.5554,
      "eval_samples_per_second": 1.726,
      "eval_steps_per_second": 1.726,
      "step": 101
    },
    {
      "epoch": 0.001165581076448406,
      "grad_norm": 4.384181499481201,
      "learning_rate": 1.2311557788944725e-05,
      "loss": 5.3644,
      "step": 102
    },
    {
      "epoch": 0.001165581076448406,
      "eval_loss": 4.984678745269775,
      "eval_runtime": 3185.3238,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 102
    },
    {
      "epoch": 0.0011770083419037824,
      "grad_norm": 3.9702224731445312,
      "learning_rate": 1.2185929648241207e-05,
      "loss": 4.6785,
      "step": 103
    },
    {
      "epoch": 0.0011770083419037824,
      "eval_loss": 4.957573890686035,
      "eval_runtime": 3183.6056,
      "eval_samples_per_second": 1.728,
      "eval_steps_per_second": 1.728,
      "step": 103
    },
    {
      "epoch": 0.001188435607359159,
      "grad_norm": 2.205033540725708,
      "learning_rate": 1.2060301507537688e-05,
      "loss": 3.2692,
      "step": 104
    },
    {
      "epoch": 0.001188435607359159,
      "eval_loss": 4.930599212646484,
      "eval_runtime": 3187.0691,
      "eval_samples_per_second": 1.726,
      "eval_steps_per_second": 1.726,
      "step": 104
    },
    {
      "epoch": 0.0011998628728145354,
      "grad_norm": 5.425383567810059,
      "learning_rate": 1.1934673366834171e-05,
      "loss": 5.8127,
      "step": 105
    },
    {
      "epoch": 0.0011998628728145354,
      "eval_loss": 4.903759002685547,
      "eval_runtime": 3191.0573,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 105
    },
    {
      "epoch": 0.001211290138269912,
      "grad_norm": 2.0843021869659424,
      "learning_rate": 1.1809045226130654e-05,
      "loss": 3.2769,
      "step": 106
    },
    {
      "epoch": 0.001211290138269912,
      "eval_loss": 4.877487659454346,
      "eval_runtime": 3192.0095,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 106
    },
    {
      "epoch": 0.0012227174037252886,
      "grad_norm": 5.174865245819092,
      "learning_rate": 1.1683417085427137e-05,
      "loss": 5.5865,
      "step": 107
    },
    {
      "epoch": 0.0012227174037252886,
      "eval_loss": 4.85078763961792,
      "eval_runtime": 3181.3647,
      "eval_samples_per_second": 1.729,
      "eval_steps_per_second": 1.729,
      "step": 107
    },
    {
      "epoch": 0.001234144669180665,
      "grad_norm": 4.530418395996094,
      "learning_rate": 1.1557788944723619e-05,
      "loss": 4.9615,
      "step": 108
    },
    {
      "epoch": 0.001234144669180665,
      "eval_loss": 4.824298858642578,
      "eval_runtime": 3188.1147,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 108
    },
    {
      "epoch": 0.0012455719346360415,
      "grad_norm": 4.7975172996521,
      "learning_rate": 1.14321608040201e-05,
      "loss": 5.4098,
      "step": 109
    },
    {
      "epoch": 0.0012455719346360415,
      "eval_loss": 4.7980499267578125,
      "eval_runtime": 3187.1944,
      "eval_samples_per_second": 1.726,
      "eval_steps_per_second": 1.726,
      "step": 109
    },
    {
      "epoch": 0.0012569992000914182,
      "grad_norm": 4.517176151275635,
      "learning_rate": 1.1306532663316583e-05,
      "loss": 5.0089,
      "step": 110
    },
    {
      "epoch": 0.0012569992000914182,
      "eval_loss": 4.771963596343994,
      "eval_runtime": 3195.7844,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 110
    },
    {
      "epoch": 0.0012684264655467947,
      "grad_norm": 4.071757793426514,
      "learning_rate": 1.1180904522613066e-05,
      "loss": 4.3814,
      "step": 111
    },
    {
      "epoch": 0.0012684264655467947,
      "eval_loss": 4.745976448059082,
      "eval_runtime": 3197.3702,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 111
    },
    {
      "epoch": 0.0012798537310021712,
      "grad_norm": 3.3903536796569824,
      "learning_rate": 1.105527638190955e-05,
      "loss": 4.0058,
      "step": 112
    },
    {
      "epoch": 0.0012798537310021712,
      "eval_loss": 4.720351696014404,
      "eval_runtime": 3192.7959,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 112
    },
    {
      "epoch": 0.0012912809964575477,
      "grad_norm": 3.1727375984191895,
      "learning_rate": 1.0929648241206031e-05,
      "loss": 3.7689,
      "step": 113
    },
    {
      "epoch": 0.0012912809964575477,
      "eval_loss": 4.694741249084473,
      "eval_runtime": 3189.1299,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 113
    },
    {
      "epoch": 0.0013027082619129242,
      "grad_norm": 2.47040057182312,
      "learning_rate": 1.0804020100502512e-05,
      "loss": 3.3884,
      "step": 114
    },
    {
      "epoch": 0.0013027082619129242,
      "eval_loss": 4.6697916984558105,
      "eval_runtime": 3190.5928,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 114
    },
    {
      "epoch": 0.0013141355273683009,
      "grad_norm": 3.910038471221924,
      "learning_rate": 1.0678391959798995e-05,
      "loss": 4.1467,
      "step": 115
    },
    {
      "epoch": 0.0013141355273683009,
      "eval_loss": 4.645045280456543,
      "eval_runtime": 3188.0361,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 115
    },
    {
      "epoch": 0.0013255627928236773,
      "grad_norm": 5.768658638000488,
      "learning_rate": 1.0552763819095479e-05,
      "loss": 5.5406,
      "step": 116
    },
    {
      "epoch": 0.0013255627928236773,
      "eval_loss": 4.620331764221191,
      "eval_runtime": 3188.8345,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 116
    },
    {
      "epoch": 0.0013369900582790538,
      "grad_norm": 4.449304580688477,
      "learning_rate": 1.042713567839196e-05,
      "loss": 4.4227,
      "step": 117
    },
    {
      "epoch": 0.0013369900582790538,
      "eval_loss": 4.595801830291748,
      "eval_runtime": 3189.9885,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 117
    },
    {
      "epoch": 0.0013484173237344303,
      "grad_norm": 5.598243236541748,
      "learning_rate": 1.0301507537688443e-05,
      "loss": 5.0256,
      "step": 118
    },
    {
      "epoch": 0.0013484173237344303,
      "eval_loss": 4.571651458740234,
      "eval_runtime": 3184.1254,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 118
    },
    {
      "epoch": 0.0013598445891898068,
      "grad_norm": 2.351072311401367,
      "learning_rate": 1.0175879396984924e-05,
      "loss": 3.3235,
      "step": 119
    },
    {
      "epoch": 0.0013598445891898068,
      "eval_loss": 4.5476908683776855,
      "eval_runtime": 3195.9101,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 119
    },
    {
      "epoch": 0.0013712718546451835,
      "grad_norm": 4.657642364501953,
      "learning_rate": 1.0050251256281408e-05,
      "loss": 4.6996,
      "step": 120
    },
    {
      "epoch": 0.0013712718546451835,
      "eval_loss": 4.524058818817139,
      "eval_runtime": 3187.7906,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 120
    },
    {
      "epoch": 0.00138269912010056,
      "grad_norm": 4.85261344909668,
      "learning_rate": 9.92462311557789e-06,
      "loss": 4.9033,
      "step": 121
    },
    {
      "epoch": 0.00138269912010056,
      "eval_loss": 4.500573635101318,
      "eval_runtime": 3191.7978,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 121
    },
    {
      "epoch": 0.0013941263855559364,
      "grad_norm": 2.878826379776001,
      "learning_rate": 9.798994974874372e-06,
      "loss": 3.7129,
      "step": 122
    },
    {
      "epoch": 0.0013941263855559364,
      "eval_loss": 4.477457046508789,
      "eval_runtime": 3188.7301,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 122
    },
    {
      "epoch": 0.001405553651011313,
      "grad_norm": 4.454727649688721,
      "learning_rate": 9.673366834170855e-06,
      "loss": 4.4705,
      "step": 123
    },
    {
      "epoch": 0.001405553651011313,
      "eval_loss": 4.454273223876953,
      "eval_runtime": 3193.5195,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 123
    },
    {
      "epoch": 0.0014169809164666896,
      "grad_norm": 5.7791547775268555,
      "learning_rate": 9.547738693467337e-06,
      "loss": 5.1999,
      "step": 124
    },
    {
      "epoch": 0.0014169809164666896,
      "eval_loss": 4.431214332580566,
      "eval_runtime": 3197.3551,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 124
    },
    {
      "epoch": 0.001428408181922066,
      "grad_norm": 4.197294235229492,
      "learning_rate": 9.42211055276382e-06,
      "loss": 4.0018,
      "step": 125
    },
    {
      "epoch": 0.001428408181922066,
      "eval_loss": 4.408187389373779,
      "eval_runtime": 3199.7906,
      "eval_samples_per_second": 1.719,
      "eval_steps_per_second": 1.719,
      "step": 125
    },
    {
      "epoch": 0.0014398354473774426,
      "grad_norm": 4.467915058135986,
      "learning_rate": 9.296482412060301e-06,
      "loss": 4.3018,
      "step": 126
    },
    {
      "epoch": 0.0014398354473774426,
      "eval_loss": 4.386054515838623,
      "eval_runtime": 3193.7722,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 126
    },
    {
      "epoch": 0.001451262712832819,
      "grad_norm": 5.470479488372803,
      "learning_rate": 9.170854271356784e-06,
      "loss": 4.8658,
      "step": 127
    },
    {
      "epoch": 0.001451262712832819,
      "eval_loss": 4.363833427429199,
      "eval_runtime": 3195.1881,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 127
    },
    {
      "epoch": 0.0014626899782881956,
      "grad_norm": 4.576855182647705,
      "learning_rate": 9.045226130653267e-06,
      "loss": 4.5801,
      "step": 128
    },
    {
      "epoch": 0.0014626899782881956,
      "eval_loss": 4.341481685638428,
      "eval_runtime": 3196.2749,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 128
    },
    {
      "epoch": 0.0014741172437435722,
      "grad_norm": 3.8720321655273438,
      "learning_rate": 8.919597989949749e-06,
      "loss": 3.9047,
      "step": 129
    },
    {
      "epoch": 0.0014741172437435722,
      "eval_loss": 4.319530010223389,
      "eval_runtime": 3192.3427,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 129
    },
    {
      "epoch": 0.0014855445091989487,
      "grad_norm": 5.404055595397949,
      "learning_rate": 8.793969849246232e-06,
      "loss": 4.5479,
      "step": 130
    },
    {
      "epoch": 0.0014855445091989487,
      "eval_loss": 4.298058032989502,
      "eval_runtime": 3192.0239,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 130
    },
    {
      "epoch": 0.0014969717746543252,
      "grad_norm": 5.212480068206787,
      "learning_rate": 8.668341708542713e-06,
      "loss": 4.6837,
      "step": 131
    },
    {
      "epoch": 0.0014969717746543252,
      "eval_loss": 4.276992321014404,
      "eval_runtime": 3190.8662,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 131
    },
    {
      "epoch": 0.0015083990401097017,
      "grad_norm": 5.532873153686523,
      "learning_rate": 8.542713567839196e-06,
      "loss": 4.5492,
      "step": 132
    },
    {
      "epoch": 0.0015083990401097017,
      "eval_loss": 4.256262302398682,
      "eval_runtime": 3195.153,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 132
    },
    {
      "epoch": 0.0015198263055650782,
      "grad_norm": 5.229096412658691,
      "learning_rate": 8.41708542713568e-06,
      "loss": 4.6645,
      "step": 133
    },
    {
      "epoch": 0.0015198263055650782,
      "eval_loss": 4.235621452331543,
      "eval_runtime": 3193.1225,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 133
    },
    {
      "epoch": 0.0015312535710204549,
      "grad_norm": 6.112283229827881,
      "learning_rate": 8.291457286432161e-06,
      "loss": 5.1063,
      "step": 134
    },
    {
      "epoch": 0.0015312535710204549,
      "eval_loss": 4.215306282043457,
      "eval_runtime": 3190.5362,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 134
    },
    {
      "epoch": 0.0015426808364758314,
      "grad_norm": 5.098965644836426,
      "learning_rate": 8.165829145728644e-06,
      "loss": 4.3275,
      "step": 135
    },
    {
      "epoch": 0.0015426808364758314,
      "eval_loss": 4.195265769958496,
      "eval_runtime": 3197.2965,
      "eval_samples_per_second": 1.72,
      "eval_steps_per_second": 1.72,
      "step": 135
    },
    {
      "epoch": 0.0015541081019312078,
      "grad_norm": 4.572518825531006,
      "learning_rate": 8.040201005025125e-06,
      "loss": 4.1709,
      "step": 136
    },
    {
      "epoch": 0.0015541081019312078,
      "eval_loss": 4.174986362457275,
      "eval_runtime": 3195.4093,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 136
    },
    {
      "epoch": 0.0015655353673865843,
      "grad_norm": 4.504931926727295,
      "learning_rate": 7.914572864321608e-06,
      "loss": 4.1334,
      "step": 137
    },
    {
      "epoch": 0.0015655353673865843,
      "eval_loss": 4.155473232269287,
      "eval_runtime": 3196.5363,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 137
    },
    {
      "epoch": 0.001576962632841961,
      "grad_norm": 4.817331790924072,
      "learning_rate": 7.788944723618092e-06,
      "loss": 4.0312,
      "step": 138
    },
    {
      "epoch": 0.001576962632841961,
      "eval_loss": 4.136098861694336,
      "eval_runtime": 3192.698,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 138
    },
    {
      "epoch": 0.0015883898982973375,
      "grad_norm": 5.8819146156311035,
      "learning_rate": 7.663316582914573e-06,
      "loss": 4.8895,
      "step": 139
    },
    {
      "epoch": 0.0015883898982973375,
      "eval_loss": 4.116922378540039,
      "eval_runtime": 3194.9621,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 139
    },
    {
      "epoch": 0.001599817163752714,
      "grad_norm": 4.634926795959473,
      "learning_rate": 7.537688442211055e-06,
      "loss": 4.1536,
      "step": 140
    },
    {
      "epoch": 0.001599817163752714,
      "eval_loss": 4.097559452056885,
      "eval_runtime": 3190.2406,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 140
    },
    {
      "epoch": 0.0016112444292080905,
      "grad_norm": 4.497166156768799,
      "learning_rate": 7.412060301507538e-06,
      "loss": 4.1384,
      "step": 141
    },
    {
      "epoch": 0.0016112444292080905,
      "eval_loss": 4.079128742218018,
      "eval_runtime": 3192.3172,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 141
    },
    {
      "epoch": 0.001622671694663467,
      "grad_norm": 5.713669300079346,
      "learning_rate": 7.28643216080402e-06,
      "loss": 4.7611,
      "step": 142
    },
    {
      "epoch": 0.001622671694663467,
      "eval_loss": 4.0609354972839355,
      "eval_runtime": 3191.9262,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 142
    },
    {
      "epoch": 0.0016340989601188436,
      "grad_norm": 4.149837017059326,
      "learning_rate": 7.160804020100504e-06,
      "loss": 3.8286,
      "step": 143
    },
    {
      "epoch": 0.0016340989601188436,
      "eval_loss": 4.042760372161865,
      "eval_runtime": 3188.8273,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 143
    },
    {
      "epoch": 0.0016455262255742201,
      "grad_norm": 6.028262138366699,
      "learning_rate": 7.035175879396985e-06,
      "loss": 4.6085,
      "step": 144
    },
    {
      "epoch": 0.0016455262255742201,
      "eval_loss": 4.024686336517334,
      "eval_runtime": 3191.547,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 144
    },
    {
      "epoch": 0.0016569534910295966,
      "grad_norm": 5.114198207855225,
      "learning_rate": 6.909547738693467e-06,
      "loss": 4.2915,
      "step": 145
    },
    {
      "epoch": 0.0016569534910295966,
      "eval_loss": 4.007183074951172,
      "eval_runtime": 3193.0349,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 145
    },
    {
      "epoch": 0.001668380756484973,
      "grad_norm": 5.462469100952148,
      "learning_rate": 6.7839195979899505e-06,
      "loss": 4.1471,
      "step": 146
    },
    {
      "epoch": 0.001668380756484973,
      "eval_loss": 3.9897186756134033,
      "eval_runtime": 3190.8896,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 146
    },
    {
      "epoch": 0.0016798080219403496,
      "grad_norm": 0.8918362855911255,
      "learning_rate": 6.658291457286432e-06,
      "loss": 2.468,
      "step": 147
    },
    {
      "epoch": 0.0016798080219403496,
      "eval_loss": 3.9733145236968994,
      "eval_runtime": 3186.1248,
      "eval_samples_per_second": 1.726,
      "eval_steps_per_second": 1.726,
      "step": 147
    },
    {
      "epoch": 0.0016912352873957263,
      "grad_norm": 4.393204689025879,
      "learning_rate": 6.532663316582915e-06,
      "loss": 3.9712,
      "step": 148
    },
    {
      "epoch": 0.0016912352873957263,
      "eval_loss": 3.956583023071289,
      "eval_runtime": 3187.8674,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 148
    },
    {
      "epoch": 0.0017026625528511028,
      "grad_norm": 6.782311916351318,
      "learning_rate": 6.407035175879397e-06,
      "loss": 5.212,
      "step": 149
    },
    {
      "epoch": 0.0017026625528511028,
      "eval_loss": 3.9407119750976562,
      "eval_runtime": 3189.4481,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 149
    },
    {
      "epoch": 0.0017140898183064792,
      "grad_norm": 3.583444356918335,
      "learning_rate": 6.2814070351758795e-06,
      "loss": 3.5865,
      "step": 150
    },
    {
      "epoch": 0.0017140898183064792,
      "eval_loss": 3.9249038696289062,
      "eval_runtime": 3193.1787,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 150
    },
    {
      "epoch": 0.0017255170837618557,
      "grad_norm": 4.707357406616211,
      "learning_rate": 6.155778894472363e-06,
      "loss": 3.508,
      "step": 151
    },
    {
      "epoch": 0.0017255170837618557,
      "eval_loss": 3.9093074798583984,
      "eval_runtime": 3187.8945,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 151
    },
    {
      "epoch": 0.0017369443492172324,
      "grad_norm": 4.769847393035889,
      "learning_rate": 6.030150753768844e-06,
      "loss": 3.8634,
      "step": 152
    },
    {
      "epoch": 0.0017369443492172324,
      "eval_loss": 3.894160270690918,
      "eval_runtime": 3191.7342,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 152
    },
    {
      "epoch": 0.001748371614672609,
      "grad_norm": 2.5702831745147705,
      "learning_rate": 5.904522613065327e-06,
      "loss": 2.8717,
      "step": 153
    },
    {
      "epoch": 0.001748371614672609,
      "eval_loss": 3.879079818725586,
      "eval_runtime": 3192.0831,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 153
    },
    {
      "epoch": 0.0017597988801279854,
      "grad_norm": 6.417558670043945,
      "learning_rate": 5.778894472361809e-06,
      "loss": 4.4295,
      "step": 154
    },
    {
      "epoch": 0.0017597988801279854,
      "eval_loss": 3.8640522956848145,
      "eval_runtime": 3194.0863,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 154
    },
    {
      "epoch": 0.0017712261455833619,
      "grad_norm": 4.981814861297607,
      "learning_rate": 5.653266331658292e-06,
      "loss": 3.6863,
      "step": 155
    },
    {
      "epoch": 0.0017712261455833619,
      "eval_loss": 3.849724769592285,
      "eval_runtime": 3192.4569,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 155
    },
    {
      "epoch": 0.0017826534110387383,
      "grad_norm": 3.5112829208374023,
      "learning_rate": 5.527638190954775e-06,
      "loss": 3.5074,
      "step": 156
    },
    {
      "epoch": 0.0017826534110387383,
      "eval_loss": 3.835315704345703,
      "eval_runtime": 3195.7661,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 156
    },
    {
      "epoch": 0.001794080676494115,
      "grad_norm": 4.1393513679504395,
      "learning_rate": 5.402010050251256e-06,
      "loss": 3.6329,
      "step": 157
    },
    {
      "epoch": 0.001794080676494115,
      "eval_loss": 3.820890188217163,
      "eval_runtime": 3191.7062,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 157
    },
    {
      "epoch": 0.0018055079419494915,
      "grad_norm": 5.644866943359375,
      "learning_rate": 5.276381909547739e-06,
      "loss": 3.8676,
      "step": 158
    },
    {
      "epoch": 0.0018055079419494915,
      "eval_loss": 3.8074545860290527,
      "eval_runtime": 3189.2491,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 158
    },
    {
      "epoch": 0.001816935207404868,
      "grad_norm": 5.790183067321777,
      "learning_rate": 5.1507537688442215e-06,
      "loss": 4.4173,
      "step": 159
    },
    {
      "epoch": 0.001816935207404868,
      "eval_loss": 3.7939019203186035,
      "eval_runtime": 3184.9276,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 159
    },
    {
      "epoch": 0.0018283624728602445,
      "grad_norm": 4.434901237487793,
      "learning_rate": 5.025125628140704e-06,
      "loss": 3.561,
      "step": 160
    },
    {
      "epoch": 0.0018283624728602445,
      "eval_loss": 3.7808175086975098,
      "eval_runtime": 3191.4708,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 160
    },
    {
      "epoch": 0.001839789738315621,
      "grad_norm": 5.6928606033325195,
      "learning_rate": 4.899497487437186e-06,
      "loss": 4.0468,
      "step": 161
    },
    {
      "epoch": 0.001839789738315621,
      "eval_loss": 3.7684166431427,
      "eval_runtime": 3189.152,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 161
    },
    {
      "epoch": 0.0018512170037709977,
      "grad_norm": 5.777411937713623,
      "learning_rate": 4.773869346733668e-06,
      "loss": 4.046,
      "step": 162
    },
    {
      "epoch": 0.0018512170037709977,
      "eval_loss": 3.755950927734375,
      "eval_runtime": 3194.3133,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 162
    },
    {
      "epoch": 0.0018626442692263742,
      "grad_norm": 4.234452247619629,
      "learning_rate": 4.6482412060301506e-06,
      "loss": 3.4605,
      "step": 163
    },
    {
      "epoch": 0.0018626442692263742,
      "eval_loss": 3.743618965148926,
      "eval_runtime": 3190.5794,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 163
    },
    {
      "epoch": 0.0018740715346817506,
      "grad_norm": 4.58046817779541,
      "learning_rate": 4.522613065326634e-06,
      "loss": 3.8547,
      "step": 164
    },
    {
      "epoch": 0.0018740715346817506,
      "eval_loss": 3.731459856033325,
      "eval_runtime": 3195.1963,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 164
    },
    {
      "epoch": 0.0018854988001371271,
      "grad_norm": 4.460115909576416,
      "learning_rate": 4.396984924623116e-06,
      "loss": 3.4863,
      "step": 165
    },
    {
      "epoch": 0.0018854988001371271,
      "eval_loss": 3.7194907665252686,
      "eval_runtime": 3192.9919,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 165
    },
    {
      "epoch": 0.0018969260655925038,
      "grad_norm": 3.752925395965576,
      "learning_rate": 4.271356783919598e-06,
      "loss": 3.1784,
      "step": 166
    },
    {
      "epoch": 0.0018969260655925038,
      "eval_loss": 3.708251476287842,
      "eval_runtime": 3191.38,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 166
    },
    {
      "epoch": 0.0019083533310478803,
      "grad_norm": 5.099639415740967,
      "learning_rate": 4.1457286432160804e-06,
      "loss": 3.9205,
      "step": 167
    },
    {
      "epoch": 0.0019083533310478803,
      "eval_loss": 3.6969456672668457,
      "eval_runtime": 3184.4302,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 167
    },
    {
      "epoch": 0.0019197805965032568,
      "grad_norm": 2.3207623958587646,
      "learning_rate": 4.020100502512563e-06,
      "loss": 2.687,
      "step": 168
    },
    {
      "epoch": 0.0019197805965032568,
      "eval_loss": 3.686185359954834,
      "eval_runtime": 3190.1367,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 168
    },
    {
      "epoch": 0.0019312078619586333,
      "grad_norm": 5.285946846008301,
      "learning_rate": 3.894472361809046e-06,
      "loss": 3.916,
      "step": 169
    },
    {
      "epoch": 0.0019312078619586333,
      "eval_loss": 3.675804615020752,
      "eval_runtime": 3187.8804,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 169
    },
    {
      "epoch": 0.0019426351274140097,
      "grad_norm": 4.83530855178833,
      "learning_rate": 3.7688442211055276e-06,
      "loss": 3.7806,
      "step": 170
    },
    {
      "epoch": 0.0019426351274140097,
      "eval_loss": 3.6655375957489014,
      "eval_runtime": 3183.8288,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 170
    },
    {
      "epoch": 0.0019540623928693862,
      "grad_norm": 5.306794166564941,
      "learning_rate": 3.64321608040201e-06,
      "loss": 3.6463,
      "step": 171
    },
    {
      "epoch": 0.0019540623928693862,
      "eval_loss": 3.6555092334747314,
      "eval_runtime": 3189.4936,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 171
    },
    {
      "epoch": 0.0019654896583247627,
      "grad_norm": 7.400163173675537,
      "learning_rate": 3.5175879396984926e-06,
      "loss": 4.8809,
      "step": 172
    },
    {
      "epoch": 0.0019654896583247627,
      "eval_loss": 3.645655632019043,
      "eval_runtime": 3189.5993,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 172
    },
    {
      "epoch": 0.0019769169237801396,
      "grad_norm": 5.281580448150635,
      "learning_rate": 3.3919597989949752e-06,
      "loss": 3.7168,
      "step": 173
    },
    {
      "epoch": 0.0019769169237801396,
      "eval_loss": 3.6365396976470947,
      "eval_runtime": 3191.3086,
      "eval_samples_per_second": 1.723,
      "eval_steps_per_second": 1.723,
      "step": 173
    },
    {
      "epoch": 0.001988344189235516,
      "grad_norm": 5.500361919403076,
      "learning_rate": 3.2663316582914575e-06,
      "loss": 3.5465,
      "step": 174
    },
    {
      "epoch": 0.001988344189235516,
      "eval_loss": 3.627671957015991,
      "eval_runtime": 3191.166,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 174
    },
    {
      "epoch": 0.0019997714546908926,
      "grad_norm": 5.550091743469238,
      "learning_rate": 3.1407035175879398e-06,
      "loss": 3.7022,
      "step": 175
    },
    {
      "epoch": 0.0019997714546908926,
      "eval_loss": 3.6191136837005615,
      "eval_runtime": 3190.3196,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 175
    },
    {
      "epoch": 0.002011198720146269,
      "grad_norm": 4.048425197601318,
      "learning_rate": 3.015075376884422e-06,
      "loss": 3.5034,
      "step": 176
    },
    {
      "epoch": 0.002011198720146269,
      "eval_loss": 3.6107609272003174,
      "eval_runtime": 3196.3566,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 176
    },
    {
      "epoch": 0.0020226259856016456,
      "grad_norm": 4.696182727813721,
      "learning_rate": 2.8894472361809047e-06,
      "loss": 3.5201,
      "step": 177
    },
    {
      "epoch": 0.0020226259856016456,
      "eval_loss": 3.602489948272705,
      "eval_runtime": 3191.1045,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 177
    },
    {
      "epoch": 0.002034053251057022,
      "grad_norm": 4.840107440948486,
      "learning_rate": 2.7638190954773874e-06,
      "loss": 3.8883,
      "step": 178
    },
    {
      "epoch": 0.002034053251057022,
      "eval_loss": 3.5948402881622314,
      "eval_runtime": 3181.8951,
      "eval_samples_per_second": 1.729,
      "eval_steps_per_second": 1.729,
      "step": 178
    },
    {
      "epoch": 0.0020454805165123985,
      "grad_norm": 3.1070148944854736,
      "learning_rate": 2.6381909547738696e-06,
      "loss": 2.8413,
      "step": 179
    },
    {
      "epoch": 0.0020454805165123985,
      "eval_loss": 3.587570905685425,
      "eval_runtime": 3187.9084,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 179
    },
    {
      "epoch": 0.002056907781967775,
      "grad_norm": 4.61260986328125,
      "learning_rate": 2.512562814070352e-06,
      "loss": 3.704,
      "step": 180
    },
    {
      "epoch": 0.002056907781967775,
      "eval_loss": 3.5807385444641113,
      "eval_runtime": 3193.1025,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 180
    },
    {
      "epoch": 0.0020683350474231515,
      "grad_norm": 5.774682998657227,
      "learning_rate": 2.386934673366834e-06,
      "loss": 4.1228,
      "step": 181
    },
    {
      "epoch": 0.0020683350474231515,
      "eval_loss": 3.5740838050842285,
      "eval_runtime": 3188.6673,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 181
    },
    {
      "epoch": 0.002079762312878528,
      "grad_norm": 6.585142612457275,
      "learning_rate": 2.261306532663317e-06,
      "loss": 4.1375,
      "step": 182
    },
    {
      "epoch": 0.002079762312878528,
      "eval_loss": 3.5675177574157715,
      "eval_runtime": 3187.9902,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 182
    },
    {
      "epoch": 0.002091189578333905,
      "grad_norm": 4.422517776489258,
      "learning_rate": 2.135678391959799e-06,
      "loss": 3.2458,
      "step": 183
    },
    {
      "epoch": 0.002091189578333905,
      "eval_loss": 3.5616683959960938,
      "eval_runtime": 3189.6985,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 183
    },
    {
      "epoch": 0.0021026168437892814,
      "grad_norm": 5.40252685546875,
      "learning_rate": 2.0100502512562813e-06,
      "loss": 3.7518,
      "step": 184
    },
    {
      "epoch": 0.0021026168437892814,
      "eval_loss": 3.556403398513794,
      "eval_runtime": 3188.5075,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 184
    },
    {
      "epoch": 0.002114044109244658,
      "grad_norm": 3.357342004776001,
      "learning_rate": 1.8844221105527638e-06,
      "loss": 2.8388,
      "step": 185
    },
    {
      "epoch": 0.002114044109244658,
      "eval_loss": 3.551210880279541,
      "eval_runtime": 3198.6279,
      "eval_samples_per_second": 1.719,
      "eval_steps_per_second": 1.719,
      "step": 185
    },
    {
      "epoch": 0.0021254713747000343,
      "grad_norm": 5.60300350189209,
      "learning_rate": 1.7587939698492463e-06,
      "loss": 3.6586,
      "step": 186
    },
    {
      "epoch": 0.0021254713747000343,
      "eval_loss": 3.5462591648101807,
      "eval_runtime": 3194.9645,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 186
    },
    {
      "epoch": 0.002136898640155411,
      "grad_norm": 4.910702228546143,
      "learning_rate": 1.6331658291457288e-06,
      "loss": 3.6408,
      "step": 187
    },
    {
      "epoch": 0.002136898640155411,
      "eval_loss": 3.5415446758270264,
      "eval_runtime": 3185.4074,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 187
    },
    {
      "epoch": 0.0021483259056107873,
      "grad_norm": 5.6200737953186035,
      "learning_rate": 1.507537688442211e-06,
      "loss": 3.5742,
      "step": 188
    },
    {
      "epoch": 0.0021483259056107873,
      "eval_loss": 3.537212371826172,
      "eval_runtime": 3182.681,
      "eval_samples_per_second": 1.728,
      "eval_steps_per_second": 1.728,
      "step": 188
    },
    {
      "epoch": 0.0021597531710661638,
      "grad_norm": 5.1306986808776855,
      "learning_rate": 1.3819095477386937e-06,
      "loss": 3.5693,
      "step": 189
    },
    {
      "epoch": 0.0021597531710661638,
      "eval_loss": 3.5330007076263428,
      "eval_runtime": 3185.4993,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 189
    },
    {
      "epoch": 0.0021711804365215403,
      "grad_norm": 5.107001781463623,
      "learning_rate": 1.256281407035176e-06,
      "loss": 3.6025,
      "step": 190
    },
    {
      "epoch": 0.0021711804365215403,
      "eval_loss": 3.529104471206665,
      "eval_runtime": 3180.9702,
      "eval_samples_per_second": 1.729,
      "eval_steps_per_second": 1.729,
      "step": 190
    },
    {
      "epoch": 0.0021826077019769167,
      "grad_norm": 5.954830646514893,
      "learning_rate": 1.1306532663316584e-06,
      "loss": 3.5938,
      "step": 191
    },
    {
      "epoch": 0.0021826077019769167,
      "eval_loss": 3.525881052017212,
      "eval_runtime": 3184.2834,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 191
    },
    {
      "epoch": 0.0021940349674322937,
      "grad_norm": 4.140493392944336,
      "learning_rate": 1.0050251256281407e-06,
      "loss": 3.3742,
      "step": 192
    },
    {
      "epoch": 0.0021940349674322937,
      "eval_loss": 3.5228757858276367,
      "eval_runtime": 3193.6231,
      "eval_samples_per_second": 1.722,
      "eval_steps_per_second": 1.722,
      "step": 192
    },
    {
      "epoch": 0.00220546223288767,
      "grad_norm": 6.1048665046691895,
      "learning_rate": 8.793969849246231e-07,
      "loss": 3.9661,
      "step": 193
    },
    {
      "epoch": 0.00220546223288767,
      "eval_loss": 3.519984722137451,
      "eval_runtime": 3190.529,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 193
    },
    {
      "epoch": 0.0022168894983430466,
      "grad_norm": 4.738156318664551,
      "learning_rate": 7.537688442211055e-07,
      "loss": 3.1686,
      "step": 194
    },
    {
      "epoch": 0.0022168894983430466,
      "eval_loss": 3.517745018005371,
      "eval_runtime": 3189.2139,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 194
    },
    {
      "epoch": 0.002228316763798423,
      "grad_norm": 6.101266860961914,
      "learning_rate": 6.28140703517588e-07,
      "loss": 3.5854,
      "step": 195
    },
    {
      "epoch": 0.002228316763798423,
      "eval_loss": 3.515723705291748,
      "eval_runtime": 3190.1764,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 195
    },
    {
      "epoch": 0.0022397440292537996,
      "grad_norm": 3.0352277755737305,
      "learning_rate": 5.025125628140703e-07,
      "loss": 2.9543,
      "step": 196
    },
    {
      "epoch": 0.0022397440292537996,
      "eval_loss": 3.5141751766204834,
      "eval_runtime": 3188.4018,
      "eval_samples_per_second": 1.725,
      "eval_steps_per_second": 1.725,
      "step": 196
    },
    {
      "epoch": 0.002251171294709176,
      "grad_norm": 3.8082351684570312,
      "learning_rate": 3.7688442211055275e-07,
      "loss": 3.3165,
      "step": 197
    },
    {
      "epoch": 0.002251171294709176,
      "eval_loss": 3.512690305709839,
      "eval_runtime": 3190.2684,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 197
    },
    {
      "epoch": 0.0022625985601645525,
      "grad_norm": 3.919834613800049,
      "learning_rate": 2.5125628140703517e-07,
      "loss": 2.769,
      "step": 198
    },
    {
      "epoch": 0.0022625985601645525,
      "eval_loss": 3.5116732120513916,
      "eval_runtime": 3190.4709,
      "eval_samples_per_second": 1.724,
      "eval_steps_per_second": 1.724,
      "step": 198
    },
    {
      "epoch": 0.002274025825619929,
      "grad_norm": 5.8951334953308105,
      "learning_rate": 1.2562814070351758e-07,
      "loss": 3.999,
      "step": 199
    },
    {
      "epoch": 0.002274025825619929,
      "eval_loss": 3.511035442352295,
      "eval_runtime": 3199.1703,
      "eval_samples_per_second": 1.719,
      "eval_steps_per_second": 1.719,
      "step": 199
    },
    {
      "epoch": 0.0022854530910753055,
      "grad_norm": 5.609264850616455,
      "learning_rate": 0.0,
      "loss": 3.8117,
      "step": 200
    },
    {
      "epoch": 0.0022854530910753055,
      "eval_loss": 3.510607957839966,
      "eval_runtime": 3196.7225,
      "eval_samples_per_second": 1.721,
      "eval_steps_per_second": 1.721,
      "step": 200
    }
  ],
  "logging_steps": 1,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1,
  "total_flos": 2564058316800000.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}