{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.984,
  "eval_steps": 1,
  "global_step": 124,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.016,
      "grad_norm": 28.861963272094727,
      "learning_rate": 2.5e-05,
      "loss": 1.317,
      "step": 1
    },
    {
      "epoch": 0.016,
      "eval_exact_match": 0.12244897959183673,
      "eval_f1_a": 0.536,
      "eval_f1_m": 0.5144300144300145,
      "eval_loss": 0.9545605182647705,
      "eval_runtime": 14.8765,
      "eval_samples_per_second": 16.805,
      "eval_steps_per_second": 2.151,
      "step": 1
    },
    {
      "epoch": 0.032,
      "grad_norm": 23.436485290527344,
      "learning_rate": 5e-05,
      "loss": 1.0813,
      "step": 2
    },
    {
      "epoch": 0.032,
      "eval_exact_match": 0.08163265306122448,
      "eval_f1_a": 0.5182186234817813,
      "eval_f1_m": 0.4855184498041642,
      "eval_loss": 0.879025399684906,
      "eval_runtime": 14.844,
      "eval_samples_per_second": 16.842,
      "eval_steps_per_second": 2.156,
      "step": 2
    },
    {
      "epoch": 0.048,
      "grad_norm": 23.96354866027832,
      "learning_rate": 4.959016393442623e-05,
      "loss": 1.1106,
      "step": 3
    },
    {
      "epoch": 0.048,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.24427480916030533,
      "eval_f1_m": 0.41115607748260813,
      "eval_loss": 0.7947519421577454,
      "eval_runtime": 14.9565,
      "eval_samples_per_second": 16.715,
      "eval_steps_per_second": 2.14,
      "step": 3
    },
    {
      "epoch": 0.064,
      "grad_norm": 17.556169509887695,
      "learning_rate": 4.918032786885246e-05,
      "loss": 0.8645,
      "step": 4
    },
    {
      "epoch": 0.064,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.8334511518478394,
      "eval_runtime": 14.9727,
      "eval_samples_per_second": 16.697,
      "eval_steps_per_second": 2.137,
      "step": 4
    },
    {
      "epoch": 0.08,
      "grad_norm": 26.969839096069336,
      "learning_rate": 4.8770491803278687e-05,
      "loss": 1.1064,
      "step": 5
    },
    {
      "epoch": 0.08,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.8471953272819519,
      "eval_runtime": 14.8399,
      "eval_samples_per_second": 16.846,
      "eval_steps_per_second": 2.156,
      "step": 5
    },
    {
      "epoch": 0.096,
      "grad_norm": 9.782492637634277,
      "learning_rate": 4.836065573770492e-05,
      "loss": 0.7013,
      "step": 6
    },
    {
      "epoch": 0.096,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.8111386895179749,
      "eval_runtime": 14.8404,
      "eval_samples_per_second": 16.846,
      "eval_steps_per_second": 2.156,
      "step": 6
    },
    {
      "epoch": 0.112,
      "grad_norm": 40.01118469238281,
      "learning_rate": 4.795081967213115e-05,
      "loss": 1.3504,
      "step": 7
    },
    {
      "epoch": 0.112,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7666797041893005,
      "eval_runtime": 14.8407,
      "eval_samples_per_second": 16.846,
      "eval_steps_per_second": 2.156,
      "step": 7
    },
    {
      "epoch": 0.128,
      "grad_norm": 19.116071701049805,
      "learning_rate": 4.754098360655738e-05,
      "loss": 0.9683,
      "step": 8
    },
    {
      "epoch": 0.128,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.10619469026548671,
      "eval_f1_m": 0.3700268552309369,
      "eval_loss": 0.7262851595878601,
      "eval_runtime": 14.8429,
      "eval_samples_per_second": 16.843,
      "eval_steps_per_second": 2.156,
      "step": 8
    },
    {
      "epoch": 0.144,
      "grad_norm": 6.786198616027832,
      "learning_rate": 4.713114754098361e-05,
      "loss": 0.6998,
      "step": 9
    },
    {
      "epoch": 0.144,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.12903225806451613,
      "eval_f1_m": 0.3618708955443649,
      "eval_loss": 0.6986015439033508,
      "eval_runtime": 14.9502,
      "eval_samples_per_second": 16.722,
      "eval_steps_per_second": 2.14,
      "step": 9
    },
    {
      "epoch": 0.16,
      "grad_norm": 8.353687286376953,
      "learning_rate": 4.672131147540984e-05,
      "loss": 0.622,
      "step": 10
    },
    {
      "epoch": 0.16,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.3668639053254438,
      "eval_f1_m": 0.4365885814865408,
      "eval_loss": 0.6866718530654907,
      "eval_runtime": 14.9736,
      "eval_samples_per_second": 16.696,
      "eval_steps_per_second": 2.137,
      "step": 10
    },
    {
      "epoch": 0.176,
      "grad_norm": 19.929168701171875,
      "learning_rate": 4.631147540983607e-05,
      "loss": 0.7718,
      "step": 11
    },
    {
      "epoch": 0.176,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.4060913705583757,
      "eval_f1_m": 0.4290875224548694,
      "eval_loss": 0.6896015405654907,
      "eval_runtime": 14.8531,
      "eval_samples_per_second": 16.832,
      "eval_steps_per_second": 2.154,
      "step": 11
    },
    {
      "epoch": 0.192,
      "grad_norm": 12.27945613861084,
      "learning_rate": 4.59016393442623e-05,
      "loss": 0.4989,
      "step": 12
    },
    {
      "epoch": 0.192,
      "eval_exact_match": 0.061224489795918366,
      "eval_f1_a": 0.5,
      "eval_f1_m": 0.5085295656724227,
      "eval_loss": 0.6914140582084656,
      "eval_runtime": 14.9377,
      "eval_samples_per_second": 16.736,
      "eval_steps_per_second": 2.142,
      "step": 12
    },
    {
      "epoch": 0.208,
      "grad_norm": 14.775979995727539,
      "learning_rate": 4.549180327868853e-05,
      "loss": 0.8284,
      "step": 13
    },
    {
      "epoch": 0.208,
      "eval_exact_match": 0.08163265306122448,
      "eval_f1_a": 0.5,
      "eval_f1_m": 0.5017629196200623,
      "eval_loss": 0.6864062547683716,
      "eval_runtime": 14.8403,
      "eval_samples_per_second": 16.846,
      "eval_steps_per_second": 2.156,
      "step": 13
    },
    {
      "epoch": 0.224,
      "grad_norm": 7.6953935623168945,
      "learning_rate": 4.508196721311476e-05,
      "loss": 0.7036,
      "step": 14
    },
    {
      "epoch": 0.224,
      "eval_exact_match": 0.061224489795918366,
      "eval_f1_a": 0.509433962264151,
      "eval_f1_m": 0.5210648761669169,
      "eval_loss": 0.6841718554496765,
      "eval_runtime": 14.9471,
      "eval_samples_per_second": 16.726,
      "eval_steps_per_second": 2.141,
      "step": 14
    },
    {
      "epoch": 0.24,
      "grad_norm": 9.705180168151855,
      "learning_rate": 4.467213114754098e-05,
      "loss": 0.6891,
      "step": 15
    },
    {
      "epoch": 0.24,
      "eval_exact_match": 0.061224489795918366,
      "eval_f1_a": 0.47761194029850745,
      "eval_f1_m": 0.49961716288246905,
      "eval_loss": 0.680414080619812,
      "eval_runtime": 14.9451,
      "eval_samples_per_second": 16.728,
      "eval_steps_per_second": 2.141,
      "step": 15
    },
    {
      "epoch": 0.256,
      "grad_norm": 7.837550640106201,
      "learning_rate": 4.426229508196721e-05,
      "loss": 0.6342,
      "step": 16
    },
    {
      "epoch": 0.256,
      "eval_exact_match": 0.061224489795918366,
      "eval_f1_a": 0.36477987421383645,
      "eval_f1_m": 0.4679963760395931,
      "eval_loss": 0.6784765720367432,
      "eval_runtime": 14.9458,
      "eval_samples_per_second": 16.727,
      "eval_steps_per_second": 2.141,
      "step": 16
    },
    {
      "epoch": 0.272,
      "grad_norm": 15.83466625213623,
      "learning_rate": 4.3852459016393444e-05,
      "loss": 0.6824,
      "step": 17
    },
    {
      "epoch": 0.272,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.25675675675675674,
      "eval_f1_m": 0.3939566102831409,
      "eval_loss": 0.6770390868186951,
      "eval_runtime": 14.847,
      "eval_samples_per_second": 16.838,
      "eval_steps_per_second": 2.155,
      "step": 17
    },
    {
      "epoch": 0.288,
      "grad_norm": 6.236104965209961,
      "learning_rate": 4.3442622950819674e-05,
      "loss": 0.6458,
      "step": 18
    },
    {
      "epoch": 0.288,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.21768707482993196,
      "eval_f1_m": 0.3807364137796312,
      "eval_loss": 0.6826757788658142,
      "eval_runtime": 14.9372,
      "eval_samples_per_second": 16.737,
      "eval_steps_per_second": 2.142,
      "step": 18
    },
    {
      "epoch": 0.304,
      "grad_norm": 16.222063064575195,
      "learning_rate": 4.3032786885245904e-05,
      "loss": 0.7531,
      "step": 19
    },
    {
      "epoch": 0.304,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.2,
      "eval_f1_m": 0.38457067475674933,
      "eval_loss": 0.6964296698570251,
      "eval_runtime": 14.8472,
      "eval_samples_per_second": 16.838,
      "eval_steps_per_second": 2.155,
      "step": 19
    },
    {
      "epoch": 0.32,
      "grad_norm": 6.506382942199707,
      "learning_rate": 4.262295081967213e-05,
      "loss": 0.6011,
      "step": 20
    },
    {
      "epoch": 0.32,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.19858156028368795,
      "eval_f1_m": 0.3815094502669533,
      "eval_loss": 0.7106679677963257,
      "eval_runtime": 14.9465,
      "eval_samples_per_second": 16.726,
      "eval_steps_per_second": 2.141,
      "step": 20
    },
    {
      "epoch": 0.336,
      "grad_norm": 13.0455961227417,
      "learning_rate": 4.2213114754098365e-05,
      "loss": 0.7092,
      "step": 21
    },
    {
      "epoch": 0.336,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.1971830985915493,
      "eval_f1_m": 0.3764074094506268,
      "eval_loss": 0.7177265882492065,
      "eval_runtime": 14.8434,
      "eval_samples_per_second": 16.842,
      "eval_steps_per_second": 2.156,
      "step": 21
    },
    {
      "epoch": 0.352,
      "grad_norm": 11.621403694152832,
      "learning_rate": 4.1803278688524595e-05,
      "loss": 0.7849,
      "step": 22
    },
    {
      "epoch": 0.352,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.1971830985915493,
      "eval_f1_m": 0.3764074094506268,
      "eval_loss": 0.7185742259025574,
      "eval_runtime": 14.8398,
      "eval_samples_per_second": 16.847,
      "eval_steps_per_second": 2.156,
      "step": 22
    },
    {
      "epoch": 0.368,
      "grad_norm": 10.119505882263184,
      "learning_rate": 4.1393442622950826e-05,
      "loss": 0.7273,
      "step": 23
    },
    {
      "epoch": 0.368,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.19858156028368795,
      "eval_f1_m": 0.3815094502669533,
      "eval_loss": 0.723312497138977,
      "eval_runtime": 14.944,
      "eval_samples_per_second": 16.729,
      "eval_steps_per_second": 2.141,
      "step": 23
    },
    {
      "epoch": 0.384,
      "grad_norm": 12.396435737609863,
      "learning_rate": 4.098360655737705e-05,
      "loss": 0.7308,
      "step": 24
    },
    {
      "epoch": 0.384,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.18705035971223022,
      "eval_f1_m": 0.37541939519730644,
      "eval_loss": 0.7317734360694885,
      "eval_runtime": 14.9534,
      "eval_samples_per_second": 16.719,
      "eval_steps_per_second": 2.14,
      "step": 24
    },
    {
      "epoch": 0.4,
      "grad_norm": 10.40471076965332,
      "learning_rate": 4.057377049180328e-05,
      "loss": 0.6329,
      "step": 25
    },
    {
      "epoch": 0.4,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.16176470588235292,
      "eval_f1_m": 0.3606963631273355,
      "eval_loss": 0.7368066310882568,
      "eval_runtime": 14.8468,
      "eval_samples_per_second": 16.839,
      "eval_steps_per_second": 2.155,
      "step": 25
    },
    {
      "epoch": 0.416,
      "grad_norm": 10.4788236618042,
      "learning_rate": 4.016393442622951e-05,
      "loss": 0.6854,
      "step": 26
    },
    {
      "epoch": 0.416,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.1925925925925926,
      "eval_f1_m": 0.3871894278857065,
      "eval_loss": 0.722824215888977,
      "eval_runtime": 14.8755,
      "eval_samples_per_second": 16.806,
      "eval_steps_per_second": 2.151,
      "step": 26
    },
    {
      "epoch": 0.432,
      "grad_norm": 11.982220649719238,
      "learning_rate": 3.975409836065574e-05,
      "loss": 0.6819,
      "step": 27
    },
    {
      "epoch": 0.432,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.15037593984962408,
      "eval_f1_m": 0.36286086529183775,
      "eval_loss": 0.7107167840003967,
      "eval_runtime": 14.8778,
      "eval_samples_per_second": 16.804,
      "eval_steps_per_second": 2.151,
      "step": 27
    },
    {
      "epoch": 0.448,
      "grad_norm": 15.360139846801758,
      "learning_rate": 3.934426229508197e-05,
      "loss": 0.6321,
      "step": 28
    },
    {
      "epoch": 0.448,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.05607476635514018,
      "eval_f1_m": 0.3736604398669225,
      "eval_loss": 0.6969433426856995,
      "eval_runtime": 14.8437,
      "eval_samples_per_second": 16.842,
      "eval_steps_per_second": 2.156,
      "step": 28
    },
    {
      "epoch": 0.464,
      "grad_norm": 10.889041900634766,
      "learning_rate": 3.89344262295082e-05,
      "loss": 0.6502,
      "step": 29
    },
    {
      "epoch": 0.464,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.6961054801940918,
      "eval_runtime": 14.8492,
      "eval_samples_per_second": 16.836,
      "eval_steps_per_second": 2.155,
      "step": 29
    },
    {
      "epoch": 0.48,
      "grad_norm": 20.48206329345703,
      "learning_rate": 3.8524590163934424e-05,
      "loss": 0.6842,
      "step": 30
    },
    {
      "epoch": 0.48,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7016035318374634,
      "eval_runtime": 14.8427,
      "eval_samples_per_second": 16.843,
      "eval_steps_per_second": 2.156,
      "step": 30
    },
    {
      "epoch": 0.496,
      "grad_norm": 10.129874229431152,
      "learning_rate": 3.8114754098360655e-05,
      "loss": 0.6571,
      "step": 31
    },
    {
      "epoch": 0.496,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7069492340087891,
      "eval_runtime": 14.9592,
      "eval_samples_per_second": 16.712,
      "eval_steps_per_second": 2.139,
      "step": 31
    },
    {
      "epoch": 0.512,
      "grad_norm": 11.055068969726562,
      "learning_rate": 3.7704918032786885e-05,
      "loss": 0.8142,
      "step": 32
    },
    {
      "epoch": 0.512,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7069980502128601,
      "eval_runtime": 14.9338,
      "eval_samples_per_second": 16.741,
      "eval_steps_per_second": 2.143,
      "step": 32
    },
    {
      "epoch": 0.528,
      "grad_norm": 7.073890209197998,
      "learning_rate": 3.729508196721312e-05,
      "loss": 0.7139,
      "step": 33
    },
    {
      "epoch": 0.528,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.6995410323143005,
      "eval_runtime": 14.938,
      "eval_samples_per_second": 16.736,
      "eval_steps_per_second": 2.142,
      "step": 33
    },
    {
      "epoch": 0.544,
      "grad_norm": 9.653722763061523,
      "learning_rate": 3.6885245901639346e-05,
      "loss": 0.739,
      "step": 34
    },
    {
      "epoch": 0.544,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7059023380279541,
      "eval_runtime": 14.9369,
      "eval_samples_per_second": 16.737,
      "eval_steps_per_second": 2.142,
      "step": 34
    },
    {
      "epoch": 0.56,
      "grad_norm": 14.42663288116455,
      "learning_rate": 3.6475409836065576e-05,
      "loss": 0.754,
      "step": 35
    },
    {
      "epoch": 0.56,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.711187481880188,
      "eval_runtime": 14.9372,
      "eval_samples_per_second": 16.737,
      "eval_steps_per_second": 2.142,
      "step": 35
    },
    {
      "epoch": 0.576,
      "grad_norm": 15.347558975219727,
      "learning_rate": 3.6065573770491806e-05,
      "loss": 0.6121,
      "step": 36
    },
    {
      "epoch": 0.576,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7238671779632568,
      "eval_runtime": 14.8763,
      "eval_samples_per_second": 16.805,
      "eval_steps_per_second": 2.151,
      "step": 36
    },
    {
      "epoch": 0.592,
      "grad_norm": 14.132453918457031,
      "learning_rate": 3.5655737704918037e-05,
      "loss": 0.6459,
      "step": 37
    },
    {
      "epoch": 0.592,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7489452958106995,
      "eval_runtime": 14.8277,
      "eval_samples_per_second": 16.86,
      "eval_steps_per_second": 2.158,
      "step": 37
    },
    {
      "epoch": 0.608,
      "grad_norm": 9.270454406738281,
      "learning_rate": 3.524590163934427e-05,
      "loss": 0.8344,
      "step": 38
    },
    {
      "epoch": 0.608,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7667334079742432,
      "eval_runtime": 14.8307,
      "eval_samples_per_second": 16.857,
      "eval_steps_per_second": 2.158,
      "step": 38
    },
    {
      "epoch": 0.624,
      "grad_norm": 19.537500381469727,
      "learning_rate": 3.483606557377049e-05,
      "loss": 0.8093,
      "step": 39
    },
    {
      "epoch": 0.624,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7680966854095459,
      "eval_runtime": 14.819,
      "eval_samples_per_second": 16.87,
      "eval_steps_per_second": 2.159,
      "step": 39
    },
    {
      "epoch": 0.64,
      "grad_norm": 7.517643451690674,
      "learning_rate": 3.442622950819672e-05,
      "loss": 0.479,
      "step": 40
    },
    {
      "epoch": 0.64,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7835771441459656,
      "eval_runtime": 14.8238,
      "eval_samples_per_second": 16.865,
      "eval_steps_per_second": 2.159,
      "step": 40
    },
    {
      "epoch": 0.656,
      "grad_norm": 17.782522201538086,
      "learning_rate": 3.401639344262295e-05,
      "loss": 0.7549,
      "step": 41
    },
    {
      "epoch": 0.656,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7776455283164978,
      "eval_runtime": 14.9327,
      "eval_samples_per_second": 16.742,
      "eval_steps_per_second": 2.143,
      "step": 41
    },
    {
      "epoch": 0.672,
      "grad_norm": 21.214466094970703,
      "learning_rate": 3.360655737704918e-05,
      "loss": 0.8666,
      "step": 42
    },
    {
      "epoch": 0.672,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7636923789978027,
      "eval_runtime": 14.9379,
      "eval_samples_per_second": 16.736,
      "eval_steps_per_second": 2.142,
      "step": 42
    },
    {
      "epoch": 0.688,
      "grad_norm": 11.387210845947266,
      "learning_rate": 3.319672131147541e-05,
      "loss": 0.9136,
      "step": 43
    },
    {
      "epoch": 0.688,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7419248223304749,
      "eval_runtime": 14.9343,
      "eval_samples_per_second": 16.74,
      "eval_steps_per_second": 2.143,
      "step": 43
    },
    {
      "epoch": 0.704,
      "grad_norm": 20.449684143066406,
      "learning_rate": 3.2786885245901635e-05,
      "loss": 0.8711,
      "step": 44
    },
    {
      "epoch": 0.704,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7206465005874634,
      "eval_runtime": 14.835,
      "eval_samples_per_second": 16.852,
      "eval_steps_per_second": 2.157,
      "step": 44
    },
    {
      "epoch": 0.72,
      "grad_norm": 11.209440231323242,
      "learning_rate": 3.237704918032787e-05,
      "loss": 0.6057,
      "step": 45
    },
    {
      "epoch": 0.72,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7194277048110962,
      "eval_runtime": 14.8379,
      "eval_samples_per_second": 16.849,
      "eval_steps_per_second": 2.157,
      "step": 45
    },
    {
      "epoch": 0.736,
      "grad_norm": 9.76227855682373,
      "learning_rate": 3.19672131147541e-05,
      "loss": 0.717,
      "step": 46
    },
    {
      "epoch": 0.736,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7115039229393005,
      "eval_runtime": 14.9392,
      "eval_samples_per_second": 16.734,
      "eval_steps_per_second": 2.142,
      "step": 46
    },
    {
      "epoch": 0.752,
      "grad_norm": 14.18212890625,
      "learning_rate": 3.155737704918033e-05,
      "loss": 0.9202,
      "step": 47
    },
    {
      "epoch": 0.752,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7000234127044678,
      "eval_runtime": 14.8408,
      "eval_samples_per_second": 16.845,
      "eval_steps_per_second": 2.156,
      "step": 47
    },
    {
      "epoch": 0.768,
      "grad_norm": 15.248029708862305,
      "learning_rate": 3.114754098360656e-05,
      "loss": 0.6564,
      "step": 48
    },
    {
      "epoch": 0.768,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7032949328422546,
      "eval_runtime": 14.8352,
      "eval_samples_per_second": 16.852,
      "eval_steps_per_second": 2.157,
      "step": 48
    },
    {
      "epoch": 0.784,
      "grad_norm": 10.2866849899292,
      "learning_rate": 3.073770491803279e-05,
      "loss": 0.6748,
      "step": 49
    },
    {
      "epoch": 0.784,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7020527124404907,
      "eval_runtime": 14.9486,
      "eval_samples_per_second": 16.724,
      "eval_steps_per_second": 2.141,
      "step": 49
    },
    {
      "epoch": 0.8,
      "grad_norm": 20.127962112426758,
      "learning_rate": 3.0327868852459017e-05,
      "loss": 0.6995,
      "step": 50
    },
    {
      "epoch": 0.8,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.6938906311988831,
      "eval_runtime": 14.8235,
      "eval_samples_per_second": 16.865,
      "eval_steps_per_second": 2.159,
      "step": 50
    },
    {
      "epoch": 0.816,
      "grad_norm": 7.355772018432617,
      "learning_rate": 2.9918032786885248e-05,
      "loss": 0.661,
      "step": 51
    },
    {
      "epoch": 0.816,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.6910097599029541,
      "eval_runtime": 14.9278,
      "eval_samples_per_second": 16.747,
      "eval_steps_per_second": 2.144,
      "step": 51
    },
    {
      "epoch": 0.832,
      "grad_norm": 7.9901123046875,
      "learning_rate": 2.9508196721311478e-05,
      "loss": 0.5164,
      "step": 52
    },
    {
      "epoch": 0.832,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.6915390491485596,
      "eval_runtime": 14.821,
      "eval_samples_per_second": 16.868,
      "eval_steps_per_second": 2.159,
      "step": 52
    },
    {
      "epoch": 0.848,
      "grad_norm": 17.773469924926758,
      "learning_rate": 2.9098360655737705e-05,
      "loss": 0.5481,
      "step": 53
    },
    {
      "epoch": 0.848,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7027539014816284,
      "eval_runtime": 14.9548,
      "eval_samples_per_second": 16.717,
      "eval_steps_per_second": 2.14,
      "step": 53
    },
    {
      "epoch": 0.864,
      "grad_norm": 6.020360469818115,
      "learning_rate": 2.8688524590163935e-05,
      "loss": 0.637,
      "step": 54
    },
    {
      "epoch": 0.864,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7196269631385803,
      "eval_runtime": 14.8795,
      "eval_samples_per_second": 16.802,
      "eval_steps_per_second": 2.151,
      "step": 54
    },
    {
      "epoch": 0.88,
      "grad_norm": 7.110622882843018,
      "learning_rate": 2.8278688524590162e-05,
      "loss": 0.6921,
      "step": 55
    },
    {
      "epoch": 0.88,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7422675490379333,
      "eval_runtime": 14.929,
      "eval_samples_per_second": 16.746,
      "eval_steps_per_second": 2.143,
      "step": 55
    },
    {
      "epoch": 0.896,
      "grad_norm": 8.011894226074219,
      "learning_rate": 2.7868852459016392e-05,
      "loss": 0.5475,
      "step": 56
    },
    {
      "epoch": 0.896,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7658622860908508,
      "eval_runtime": 14.8404,
      "eval_samples_per_second": 16.846,
      "eval_steps_per_second": 2.156,
      "step": 56
    },
    {
      "epoch": 0.912,
      "grad_norm": 11.89902114868164,
      "learning_rate": 2.7459016393442626e-05,
      "loss": 0.7004,
      "step": 57
    },
    {
      "epoch": 0.912,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.795605480670929,
      "eval_runtime": 14.9551,
      "eval_samples_per_second": 16.717,
      "eval_steps_per_second": 2.14,
      "step": 57
    },
    {
      "epoch": 0.928,
      "grad_norm": 17.989233016967773,
      "learning_rate": 2.7049180327868856e-05,
      "loss": 0.7371,
      "step": 58
    },
    {
      "epoch": 0.928,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.8144462704658508,
      "eval_runtime": 14.9635,
      "eval_samples_per_second": 16.707,
      "eval_steps_per_second": 2.139,
      "step": 58
    },
    {
      "epoch": 0.944,
      "grad_norm": 20.869640350341797,
      "learning_rate": 2.6639344262295087e-05,
      "loss": 0.7959,
      "step": 59
    },
    {
      "epoch": 0.944,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.8224091529846191,
      "eval_runtime": 14.8308,
      "eval_samples_per_second": 16.857,
      "eval_steps_per_second": 2.158,
      "step": 59
    },
    {
      "epoch": 0.96,
      "grad_norm": 18.80642318725586,
      "learning_rate": 2.6229508196721314e-05,
      "loss": 0.9499,
      "step": 60
    },
    {
      "epoch": 0.96,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.8185468912124634,
      "eval_runtime": 14.9357,
      "eval_samples_per_second": 16.738,
      "eval_steps_per_second": 2.143,
      "step": 60
    },
    {
      "epoch": 0.976,
      "grad_norm": 13.714020729064941,
      "learning_rate": 2.5819672131147544e-05,
      "loss": 0.686,
      "step": 61
    },
    {
      "epoch": 0.976,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.804934561252594,
      "eval_runtime": 14.9362,
      "eval_samples_per_second": 16.738,
      "eval_steps_per_second": 2.142,
      "step": 61
    },
    {
      "epoch": 0.992,
      "grad_norm": 8.152863502502441,
      "learning_rate": 2.540983606557377e-05,
      "loss": 0.586,
      "step": 62
    },
    {
      "epoch": 0.992,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7904736399650574,
      "eval_runtime": 14.8214,
      "eval_samples_per_second": 16.867,
      "eval_steps_per_second": 2.159,
      "step": 62
    },
    {
      "epoch": 1.008,
      "grad_norm": 15.25986099243164,
      "learning_rate": 2.5e-05,
      "loss": 0.8011,
      "step": 63
    },
    {
      "epoch": 1.008,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7669472694396973,
      "eval_runtime": 14.9396,
      "eval_samples_per_second": 16.734,
      "eval_steps_per_second": 2.142,
      "step": 63
    },
    {
      "epoch": 1.024,
      "grad_norm": 9.822798728942871,
      "learning_rate": 2.459016393442623e-05,
      "loss": 0.671,
      "step": 64
    },
    {
      "epoch": 1.024,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7386738061904907,
      "eval_runtime": 14.9319,
      "eval_samples_per_second": 16.743,
      "eval_steps_per_second": 2.143,
      "step": 64
    },
    {
      "epoch": 1.04,
      "grad_norm": 16.625377655029297,
      "learning_rate": 2.418032786885246e-05,
      "loss": 0.804,
      "step": 65
    },
    {
      "epoch": 1.04,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7147324085235596,
      "eval_runtime": 14.8352,
      "eval_samples_per_second": 16.852,
      "eval_steps_per_second": 2.157,
      "step": 65
    },
    {
      "epoch": 1.056,
      "grad_norm": 18.697471618652344,
      "learning_rate": 2.377049180327869e-05,
      "loss": 0.7867,
      "step": 66
    },
    {
      "epoch": 1.056,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.6943925619125366,
      "eval_runtime": 14.9325,
      "eval_samples_per_second": 16.742,
      "eval_steps_per_second": 2.143,
      "step": 66
    },
    {
      "epoch": 1.072,
      "grad_norm": 9.6091947555542,
      "learning_rate": 2.336065573770492e-05,
      "loss": 0.6648,
      "step": 67
    },
    {
      "epoch": 1.072,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.07407407407407407,
      "eval_f1_m": 0.3703287375556284,
      "eval_loss": 0.6778144240379333,
      "eval_runtime": 14.9373,
      "eval_samples_per_second": 16.737,
      "eval_steps_per_second": 2.142,
      "step": 67
    },
    {
      "epoch": 1.088,
      "grad_norm": 14.563115119934082,
      "learning_rate": 2.295081967213115e-05,
      "loss": 0.614,
      "step": 68
    },
    {
      "epoch": 1.088,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.2835820895522388,
      "eval_f1_m": 0.4365348370450412,
      "eval_loss": 0.671093761920929,
      "eval_runtime": 14.8252,
      "eval_samples_per_second": 16.863,
      "eval_steps_per_second": 2.158,
      "step": 68
    },
    {
      "epoch": 1.104,
      "grad_norm": 11.200250625610352,
      "learning_rate": 2.254098360655738e-05,
      "loss": 0.5507,
      "step": 69
    },
    {
      "epoch": 1.104,
      "eval_exact_match": 0.061224489795918366,
      "eval_f1_a": 0.36601307189542487,
      "eval_f1_m": 0.4596024157248647,
      "eval_loss": 0.667160153388977,
      "eval_runtime": 14.8406,
      "eval_samples_per_second": 16.846,
      "eval_steps_per_second": 2.156,
      "step": 69
    },
    {
      "epoch": 1.12,
      "grad_norm": 8.560547828674316,
      "learning_rate": 2.2131147540983607e-05,
      "loss": 0.5892,
      "step": 70
    },
    {
      "epoch": 1.12,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.4022988505747126,
      "eval_f1_m": 0.45429236296583236,
      "eval_loss": 0.6715586185455322,
      "eval_runtime": 14.9417,
      "eval_samples_per_second": 16.732,
      "eval_steps_per_second": 2.142,
      "step": 70
    },
    {
      "epoch": 1.1360000000000001,
      "grad_norm": 8.961625099182129,
      "learning_rate": 2.1721311475409837e-05,
      "loss": 0.7664,
      "step": 71
    },
    {
      "epoch": 1.1360000000000001,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.3742690058479532,
      "eval_f1_m": 0.43382439555908947,
      "eval_loss": 0.6713827848434448,
      "eval_runtime": 14.8405,
      "eval_samples_per_second": 16.846,
      "eval_steps_per_second": 2.156,
      "step": 71
    },
    {
      "epoch": 1.152,
      "grad_norm": 7.123610973358154,
      "learning_rate": 2.1311475409836064e-05,
      "loss": 0.7411,
      "step": 72
    },
    {
      "epoch": 1.152,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.3977272727272727,
      "eval_f1_m": 0.44959213122478425,
      "eval_loss": 0.6721289157867432,
      "eval_runtime": 14.9381,
      "eval_samples_per_second": 16.736,
      "eval_steps_per_second": 2.142,
      "step": 72
    },
    {
      "epoch": 1.168,
      "grad_norm": 16.729991912841797,
      "learning_rate": 2.0901639344262298e-05,
      "loss": 0.8682,
      "step": 73
    },
    {
      "epoch": 1.168,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.393063583815029,
      "eval_f1_m": 0.45827664399092977,
      "eval_loss": 0.6705155968666077,
      "eval_runtime": 14.9797,
      "eval_samples_per_second": 16.689,
      "eval_steps_per_second": 2.136,
      "step": 73
    },
    {
      "epoch": 1.184,
      "grad_norm": 6.746473789215088,
      "learning_rate": 2.0491803278688525e-05,
      "loss": 0.6803,
      "step": 74
    },
    {
      "epoch": 1.184,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.3508771929824561,
      "eval_f1_m": 0.4224165856818919,
      "eval_loss": 0.6745976805686951,
      "eval_runtime": 14.9511,
      "eval_samples_per_second": 16.721,
      "eval_steps_per_second": 2.14,
      "step": 74
    },
    {
      "epoch": 1.2,
      "grad_norm": 13.287238121032715,
      "learning_rate": 2.0081967213114755e-05,
      "loss": 0.6991,
      "step": 75
    },
    {
      "epoch": 1.2,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.41111111111111115,
      "eval_f1_m": 0.4558379262460896,
      "eval_loss": 0.6722421646118164,
      "eval_runtime": 15.4359,
      "eval_samples_per_second": 16.196,
      "eval_steps_per_second": 2.073,
      "step": 75
    },
    {
      "epoch": 1.216,
      "grad_norm": 20.5496883392334,
      "learning_rate": 1.9672131147540985e-05,
      "loss": 0.7513,
      "step": 76
    },
    {
      "epoch": 1.216,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.3953488372093023,
      "eval_f1_m": 0.45770680566598937,
      "eval_loss": 0.6721171736717224,
      "eval_runtime": 15.0052,
      "eval_samples_per_second": 16.661,
      "eval_steps_per_second": 2.133,
      "step": 76
    },
    {
      "epoch": 1.232,
      "grad_norm": 5.193902015686035,
      "learning_rate": 1.9262295081967212e-05,
      "loss": 0.6274,
      "step": 77
    },
    {
      "epoch": 1.232,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.3932584269662921,
      "eval_f1_m": 0.44960334450130357,
      "eval_loss": 0.6729999780654907,
      "eval_runtime": 14.876,
      "eval_samples_per_second": 16.806,
      "eval_steps_per_second": 2.151,
      "step": 77
    },
    {
      "epoch": 1.248,
      "grad_norm": 13.002022743225098,
      "learning_rate": 1.8852459016393442e-05,
      "loss": 0.6462,
      "step": 78
    },
    {
      "epoch": 1.248,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.384180790960452,
      "eval_f1_m": 0.44293518273110116,
      "eval_loss": 0.6707578301429749,
      "eval_runtime": 14.9856,
      "eval_samples_per_second": 16.683,
      "eval_steps_per_second": 2.135,
      "step": 78
    },
    {
      "epoch": 1.264,
      "grad_norm": 12.032450675964355,
      "learning_rate": 1.8442622950819673e-05,
      "loss": 0.5559,
      "step": 79
    },
    {
      "epoch": 1.264,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.40437158469945356,
      "eval_f1_m": 0.4476750007362253,
      "eval_loss": 0.674972653388977,
      "eval_runtime": 14.9758,
      "eval_samples_per_second": 16.694,
      "eval_steps_per_second": 2.137,
      "step": 79
    },
    {
      "epoch": 1.28,
      "grad_norm": 22.6571044921875,
      "learning_rate": 1.8032786885245903e-05,
      "loss": 0.8383,
      "step": 80
    },
    {
      "epoch": 1.28,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.37569060773480667,
      "eval_f1_m": 0.4270836872877689,
      "eval_loss": 0.6759648323059082,
      "eval_runtime": 14.8726,
      "eval_samples_per_second": 16.809,
      "eval_steps_per_second": 2.152,
      "step": 80
    },
    {
      "epoch": 1.296,
      "grad_norm": 5.78735876083374,
      "learning_rate": 1.7622950819672133e-05,
      "loss": 0.4898,
      "step": 81
    },
    {
      "epoch": 1.296,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.40437158469945356,
      "eval_f1_m": 0.45088199782077343,
      "eval_loss": 0.6767851710319519,
      "eval_runtime": 14.9835,
      "eval_samples_per_second": 16.685,
      "eval_steps_per_second": 2.136,
      "step": 81
    },
    {
      "epoch": 1.312,
      "grad_norm": 9.217117309570312,
      "learning_rate": 1.721311475409836e-05,
      "loss": 0.7316,
      "step": 82
    },
    {
      "epoch": 1.312,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.4148936170212766,
      "eval_f1_m": 0.4551692072100236,
      "eval_loss": 0.6751718521118164,
      "eval_runtime": 14.7473,
      "eval_samples_per_second": 16.952,
      "eval_steps_per_second": 2.17,
      "step": 82
    },
    {
      "epoch": 1.328,
      "grad_norm": 20.60079574584961,
      "learning_rate": 1.680327868852459e-05,
      "loss": 0.7904,
      "step": 83
    },
    {
      "epoch": 1.328,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.40449438202247184,
      "eval_f1_m": 0.4541353770945608,
      "eval_loss": 0.6738359332084656,
      "eval_runtime": 14.9829,
      "eval_samples_per_second": 16.686,
      "eval_steps_per_second": 2.136,
      "step": 83
    },
    {
      "epoch": 1.3439999999999999,
      "grad_norm": 7.957705497741699,
      "learning_rate": 1.6393442622950818e-05,
      "loss": 0.7006,
      "step": 84
    },
    {
      "epoch": 1.3439999999999999,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.38636363636363635,
      "eval_f1_m": 0.4470978001590247,
      "eval_loss": 0.6728906035423279,
      "eval_runtime": 14.9863,
      "eval_samples_per_second": 16.682,
      "eval_steps_per_second": 2.135,
      "step": 84
    },
    {
      "epoch": 1.3599999999999999,
      "grad_norm": 12.297500610351562,
      "learning_rate": 1.598360655737705e-05,
      "loss": 0.7209,
      "step": 85
    },
    {
      "epoch": 1.3599999999999999,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.4318181818181818,
      "eval_f1_m": 0.48761612537122734,
      "eval_loss": 0.6701562404632568,
      "eval_runtime": 14.8996,
      "eval_samples_per_second": 16.779,
      "eval_steps_per_second": 2.148,
      "step": 85
    },
    {
      "epoch": 1.376,
      "grad_norm": 11.166065216064453,
      "learning_rate": 1.557377049180328e-05,
      "loss": 0.6469,
      "step": 86
    },
    {
      "epoch": 1.376,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.4022988505747126,
      "eval_f1_m": 0.46504402626851604,
      "eval_loss": 0.6728047132492065,
      "eval_runtime": 14.9978,
      "eval_samples_per_second": 16.669,
      "eval_steps_per_second": 2.134,
      "step": 86
    },
    {
      "epoch": 1.392,
      "grad_norm": 5.032862186431885,
      "learning_rate": 1.5163934426229509e-05,
      "loss": 0.7119,
      "step": 87
    },
    {
      "epoch": 1.392,
      "eval_exact_match": 0.061224489795918366,
      "eval_f1_a": 0.3950617283950618,
      "eval_f1_m": 0.4722391554024206,
      "eval_loss": 0.6710312366485596,
      "eval_runtime": 14.8702,
      "eval_samples_per_second": 16.812,
      "eval_steps_per_second": 2.152,
      "step": 87
    },
    {
      "epoch": 1.408,
      "grad_norm": 23.827577590942383,
      "learning_rate": 1.4754098360655739e-05,
      "loss": 0.7244,
      "step": 88
    },
    {
      "epoch": 1.408,
      "eval_exact_match": 0.08163265306122448,
      "eval_f1_a": 0.3766233766233766,
      "eval_f1_m": 0.476664258777104,
      "eval_loss": 0.6692422032356262,
      "eval_runtime": 14.873,
      "eval_samples_per_second": 16.809,
      "eval_steps_per_second": 2.152,
      "step": 88
    },
    {
      "epoch": 1.424,
      "grad_norm": 17.192716598510742,
      "learning_rate": 1.4344262295081968e-05,
      "loss": 0.7095,
      "step": 89
    },
    {
      "epoch": 1.424,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.3088235294117647,
      "eval_f1_m": 0.439154213133805,
      "eval_loss": 0.6696093678474426,
      "eval_runtime": 14.9667,
      "eval_samples_per_second": 16.704,
      "eval_steps_per_second": 2.138,
      "step": 89
    },
    {
      "epoch": 1.44,
      "grad_norm": 11.38125228881836,
      "learning_rate": 1.3934426229508196e-05,
      "loss": 0.7006,
      "step": 90
    },
    {
      "epoch": 1.44,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.32352941176470584,
      "eval_f1_m": 0.44573533042920804,
      "eval_loss": 0.6697929501533508,
      "eval_runtime": 14.9754,
      "eval_samples_per_second": 16.694,
      "eval_steps_per_second": 2.137,
      "step": 90
    },
    {
      "epoch": 1.456,
      "grad_norm": 9.435693740844727,
      "learning_rate": 1.3524590163934428e-05,
      "loss": 0.6404,
      "step": 91
    },
    {
      "epoch": 1.456,
      "eval_exact_match": 0.04081632653061224,
      "eval_f1_a": 0.2741935483870968,
      "eval_f1_m": 0.4377146662860949,
      "eval_loss": 0.6705585718154907,
      "eval_runtime": 14.9011,
      "eval_samples_per_second": 16.777,
      "eval_steps_per_second": 2.147,
      "step": 91
    },
    {
      "epoch": 1.472,
      "grad_norm": 10.128896713256836,
      "learning_rate": 1.3114754098360657e-05,
      "loss": 0.7173,
      "step": 92
    },
    {
      "epoch": 1.472,
      "eval_exact_match": 0.02040816326530612,
      "eval_f1_a": 0.17391304347826086,
      "eval_f1_m": 0.40892491861879615,
      "eval_loss": 0.6743906140327454,
      "eval_runtime": 14.8853,
      "eval_samples_per_second": 16.795,
      "eval_steps_per_second": 2.15,
      "step": 92
    },
    {
      "epoch": 1.488,
      "grad_norm": 17.08680534362793,
      "learning_rate": 1.2704918032786885e-05,
      "loss": 0.704,
      "step": 93
    },
    {
      "epoch": 1.488,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.09174311926605504,
      "eval_f1_m": 0.380698866293104,
      "eval_loss": 0.6720312237739563,
      "eval_runtime": 14.9786,
      "eval_samples_per_second": 16.691,
      "eval_steps_per_second": 2.136,
      "step": 93
    },
    {
      "epoch": 1.504,
      "grad_norm": 7.702959060668945,
      "learning_rate": 1.2295081967213116e-05,
      "loss": 0.5519,
      "step": 94
    },
    {
      "epoch": 1.504,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.09345794392523366,
      "eval_f1_m": 0.38281036024133264,
      "eval_loss": 0.6760703325271606,
      "eval_runtime": 14.8739,
      "eval_samples_per_second": 16.808,
      "eval_steps_per_second": 2.151,
      "step": 94
    },
    {
      "epoch": 1.52,
      "grad_norm": 14.851384162902832,
      "learning_rate": 1.1885245901639344e-05,
      "loss": 0.5096,
      "step": 95
    },
    {
      "epoch": 1.52,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.07619047619047618,
      "eval_f1_m": 0.3760076391528973,
      "eval_loss": 0.6754101514816284,
      "eval_runtime": 14.7313,
      "eval_samples_per_second": 16.971,
      "eval_steps_per_second": 2.172,
      "step": 95
    },
    {
      "epoch": 1.536,
      "grad_norm": 12.864474296569824,
      "learning_rate": 1.1475409836065575e-05,
      "loss": 0.5991,
      "step": 96
    },
    {
      "epoch": 1.536,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.038834951456310676,
      "eval_f1_m": 0.36619522934048754,
      "eval_loss": 0.6819843649864197,
      "eval_runtime": 14.9845,
      "eval_samples_per_second": 16.684,
      "eval_steps_per_second": 2.136,
      "step": 96
    },
    {
      "epoch": 1.552,
      "grad_norm": 7.286766529083252,
      "learning_rate": 1.1065573770491803e-05,
      "loss": 0.5792,
      "step": 97
    },
    {
      "epoch": 1.552,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.6806367039680481,
      "eval_runtime": 14.9973,
      "eval_samples_per_second": 16.67,
      "eval_steps_per_second": 2.134,
      "step": 97
    },
    {
      "epoch": 1.568,
      "grad_norm": 17.556123733520508,
      "learning_rate": 1.0655737704918032e-05,
      "loss": 0.6297,
      "step": 98
    },
    {
      "epoch": 1.568,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.6815586090087891,
      "eval_runtime": 15.128,
      "eval_samples_per_second": 16.526,
      "eval_steps_per_second": 2.115,
      "step": 98
    },
    {
      "epoch": 1.584,
      "grad_norm": 9.33618450164795,
      "learning_rate": 1.0245901639344262e-05,
      "loss": 0.6484,
      "step": 99
    },
    {
      "epoch": 1.584,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0196078431372549,
      "eval_f1_m": 0.35848547877359416,
      "eval_loss": 0.6851093769073486,
      "eval_runtime": 14.9937,
      "eval_samples_per_second": 16.674,
      "eval_steps_per_second": 2.134,
      "step": 99
    },
    {
      "epoch": 1.6,
      "grad_norm": 10.44613265991211,
      "learning_rate": 9.836065573770493e-06,
      "loss": 0.6437,
      "step": 100
    },
    {
      "epoch": 1.6,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.6876327991485596,
      "eval_runtime": 14.9142,
      "eval_samples_per_second": 16.763,
      "eval_steps_per_second": 2.146,
      "step": 100
    },
    {
      "epoch": 1.616,
      "grad_norm": 9.857686042785645,
      "learning_rate": 9.426229508196721e-06,
      "loss": 0.595,
      "step": 101
    },
    {
      "epoch": 1.616,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.6902187466621399,
      "eval_runtime": 14.9287,
      "eval_samples_per_second": 16.746,
      "eval_steps_per_second": 2.144,
      "step": 101
    },
    {
      "epoch": 1.6320000000000001,
      "grad_norm": 17.187076568603516,
      "learning_rate": 9.016393442622952e-06,
      "loss": 0.6899,
      "step": 102
    },
    {
      "epoch": 1.6320000000000001,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.692144513130188,
      "eval_runtime": 14.9838,
      "eval_samples_per_second": 16.685,
      "eval_steps_per_second": 2.136,
      "step": 102
    },
    {
      "epoch": 1.6480000000000001,
      "grad_norm": 11.167596817016602,
      "learning_rate": 8.60655737704918e-06,
      "loss": 0.6288,
      "step": 103
    },
    {
      "epoch": 1.6480000000000001,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.6949140429496765,
      "eval_runtime": 14.9759,
      "eval_samples_per_second": 16.693,
      "eval_steps_per_second": 2.137,
      "step": 103
    },
    {
      "epoch": 1.6640000000000001,
      "grad_norm": 12.586868286132812,
      "learning_rate": 8.196721311475409e-06,
      "loss": 0.552,
      "step": 104
    },
    {
      "epoch": 1.6640000000000001,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.6984257698059082,
      "eval_runtime": 14.8729,
      "eval_samples_per_second": 16.809,
      "eval_steps_per_second": 2.152,
      "step": 104
    },
    {
      "epoch": 1.6800000000000002,
      "grad_norm": 9.765538215637207,
      "learning_rate": 7.78688524590164e-06,
      "loss": 0.6785,
      "step": 105
    },
    {
      "epoch": 1.6800000000000002,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.704703152179718,
      "eval_runtime": 14.9801,
      "eval_samples_per_second": 16.689,
      "eval_steps_per_second": 2.136,
      "step": 105
    },
    {
      "epoch": 1.696,
      "grad_norm": 6.867098808288574,
      "learning_rate": 7.3770491803278695e-06,
      "loss": 0.6315,
      "step": 106
    },
    {
      "epoch": 1.696,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7083241939544678,
      "eval_runtime": 14.727,
      "eval_samples_per_second": 16.976,
      "eval_steps_per_second": 2.173,
      "step": 106
    },
    {
      "epoch": 1.712,
      "grad_norm": 8.60350513458252,
      "learning_rate": 6.967213114754098e-06,
      "loss": 0.5195,
      "step": 107
    },
    {
      "epoch": 1.712,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7091211080551147,
      "eval_runtime": 14.874,
      "eval_samples_per_second": 16.808,
      "eval_steps_per_second": 2.151,
      "step": 107
    },
    {
      "epoch": 1.728,
      "grad_norm": 8.559430122375488,
      "learning_rate": 6.557377049180328e-06,
      "loss": 0.7978,
      "step": 108
    },
    {
      "epoch": 1.728,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7129042744636536,
      "eval_runtime": 14.9768,
      "eval_samples_per_second": 16.693,
      "eval_steps_per_second": 2.137,
      "step": 108
    },
    {
      "epoch": 1.744,
      "grad_norm": 5.806375026702881,
      "learning_rate": 6.147540983606558e-06,
      "loss": 0.5901,
      "step": 109
    },
    {
      "epoch": 1.744,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.715499997138977,
      "eval_runtime": 14.8739,
      "eval_samples_per_second": 16.808,
      "eval_steps_per_second": 2.151,
      "step": 109
    },
    {
      "epoch": 1.76,
      "grad_norm": 8.308993339538574,
      "learning_rate": 5.737704918032787e-06,
      "loss": 0.7361,
      "step": 110
    },
    {
      "epoch": 1.76,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7156738042831421,
      "eval_runtime": 14.9831,
      "eval_samples_per_second": 16.685,
      "eval_steps_per_second": 2.136,
      "step": 110
    },
    {
      "epoch": 1.776,
      "grad_norm": 5.917534351348877,
      "learning_rate": 5.327868852459016e-06,
      "loss": 0.6439,
      "step": 111
    },
    {
      "epoch": 1.776,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7138105630874634,
      "eval_runtime": 14.9884,
      "eval_samples_per_second": 16.68,
      "eval_steps_per_second": 2.135,
      "step": 111
    },
    {
      "epoch": 1.792,
      "grad_norm": 12.60551929473877,
      "learning_rate": 4.918032786885246e-06,
      "loss": 0.5871,
      "step": 112
    },
    {
      "epoch": 1.792,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.714160144329071,
      "eval_runtime": 14.9755,
      "eval_samples_per_second": 16.694,
      "eval_steps_per_second": 2.137,
      "step": 112
    },
    {
      "epoch": 1.808,
      "grad_norm": 5.75510311126709,
      "learning_rate": 4.508196721311476e-06,
      "loss": 0.5484,
      "step": 113
    },
    {
      "epoch": 1.808,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7168398499488831,
      "eval_runtime": 14.8968,
      "eval_samples_per_second": 16.782,
      "eval_steps_per_second": 2.148,
      "step": 113
    },
    {
      "epoch": 1.8239999999999998,
      "grad_norm": 17.395551681518555,
      "learning_rate": 4.098360655737704e-06,
      "loss": 0.7153,
      "step": 114
    },
    {
      "epoch": 1.8239999999999998,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.712472677230835,
      "eval_runtime": 14.9892,
      "eval_samples_per_second": 16.679,
      "eval_steps_per_second": 2.135,
      "step": 114
    },
    {
      "epoch": 1.8399999999999999,
      "grad_norm": 13.188994407653809,
      "learning_rate": 3.6885245901639347e-06,
      "loss": 0.6075,
      "step": 115
    },
    {
      "epoch": 1.8399999999999999,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7133281230926514,
      "eval_runtime": 14.9832,
      "eval_samples_per_second": 16.685,
      "eval_steps_per_second": 2.136,
      "step": 115
    },
    {
      "epoch": 1.8559999999999999,
      "grad_norm": 7.989605903625488,
      "learning_rate": 3.278688524590164e-06,
      "loss": 0.6436,
      "step": 116
    },
    {
      "epoch": 1.8559999999999999,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.712271511554718,
      "eval_runtime": 14.8835,
      "eval_samples_per_second": 16.797,
      "eval_steps_per_second": 2.15,
      "step": 116
    },
    {
      "epoch": 1.8719999999999999,
      "grad_norm": 10.555407524108887,
      "learning_rate": 2.8688524590163937e-06,
      "loss": 0.6506,
      "step": 117
    },
    {
      "epoch": 1.8719999999999999,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7101816534996033,
      "eval_runtime": 14.8733,
      "eval_samples_per_second": 16.809,
      "eval_steps_per_second": 2.152,
      "step": 117
    },
    {
      "epoch": 1.888,
      "grad_norm": 6.743683815002441,
      "learning_rate": 2.459016393442623e-06,
      "loss": 0.6718,
      "step": 118
    },
    {
      "epoch": 1.888,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7138828039169312,
      "eval_runtime": 14.9743,
      "eval_samples_per_second": 16.695,
      "eval_steps_per_second": 2.137,
      "step": 118
    },
    {
      "epoch": 1.904,
      "grad_norm": 5.256937026977539,
      "learning_rate": 2.049180327868852e-06,
      "loss": 0.6927,
      "step": 119
    },
    {
      "epoch": 1.904,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7110937237739563,
      "eval_runtime": 14.8825,
      "eval_samples_per_second": 16.798,
      "eval_steps_per_second": 2.15,
      "step": 119
    },
    {
      "epoch": 1.92,
      "grad_norm": 15.087100982666016,
      "learning_rate": 1.639344262295082e-06,
      "loss": 0.7999,
      "step": 120
    },
    {
      "epoch": 1.92,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7134785056114197,
      "eval_runtime": 15.0787,
      "eval_samples_per_second": 16.58,
      "eval_steps_per_second": 2.122,
      "step": 120
    },
    {
      "epoch": 1.936,
      "grad_norm": 10.094852447509766,
      "learning_rate": 1.2295081967213116e-06,
      "loss": 0.5977,
      "step": 121
    },
    {
      "epoch": 1.936,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7145839929580688,
      "eval_runtime": 14.8868,
      "eval_samples_per_second": 16.793,
      "eval_steps_per_second": 2.15,
      "step": 121
    },
    {
      "epoch": 1.952,
      "grad_norm": 11.906929016113281,
      "learning_rate": 8.19672131147541e-07,
      "loss": 0.6042,
      "step": 122
    },
    {
      "epoch": 1.952,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7139609456062317,
      "eval_runtime": 14.9736,
      "eval_samples_per_second": 16.696,
      "eval_steps_per_second": 2.137,
      "step": 122
    },
    {
      "epoch": 1.968,
      "grad_norm": 9.525917053222656,
      "learning_rate": 4.098360655737705e-07,
      "loss": 0.6814,
      "step": 123
    },
    {
      "epoch": 1.968,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7123749852180481,
      "eval_runtime": 14.9772,
      "eval_samples_per_second": 16.692,
      "eval_steps_per_second": 2.137,
      "step": 123
    },
    {
      "epoch": 1.984,
      "grad_norm": 11.106483459472656,
      "learning_rate": 0.0,
      "loss": 0.5725,
      "step": 124
    },
    {
      "epoch": 1.984,
      "eval_exact_match": 0.0,
      "eval_f1_a": 0.0,
      "eval_f1_m": 0.35077572820670067,
      "eval_loss": 0.7136836051940918,
      "eval_runtime": 15.0058,
      "eval_samples_per_second": 16.66,
      "eval_steps_per_second": 2.133,
      "step": 124
    },
    {
      "epoch": 1.984,
      "step": 124,
      "total_flos": 3.579246986605363e+16,
      "train_loss": 0.7063398053569179,
      "train_runtime": 2392.3346,
      "train_samples_per_second": 0.836,
      "train_steps_per_second": 0.052
    }
  ],
  "logging_steps": 1,
  "max_steps": 124,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 3.579246986605363e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}