{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 100,
  "global_step": 5300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18867924528301888,
      "grad_norm": 4.437798503453146,
      "learning_rate": 7.341400415124174e-07,
      "loss": 5.3755,
      "step": 100
    },
    {
      "epoch": 0.18867924528301888,
      "eval_loss": 1.0850828886032104,
      "eval_runtime": 149.7427,
      "eval_samples_per_second": 452.857,
      "eval_steps_per_second": 0.888,
      "step": 100
    },
    {
      "epoch": 0.37735849056603776,
      "grad_norm": 1.6459101244560304,
      "learning_rate": 8.446391282690362e-07,
      "loss": 0.9681,
      "step": 200
    },
    {
      "epoch": 0.37735849056603776,
      "eval_loss": 0.9421371817588806,
      "eval_runtime": 149.748,
      "eval_samples_per_second": 452.841,
      "eval_steps_per_second": 0.888,
      "step": 200
    },
    {
      "epoch": 0.5660377358490566,
      "grad_norm": 4.060133963944513,
      "learning_rate": 9.09276950385592e-07,
      "loss": 0.9538,
      "step": 300
    },
    {
      "epoch": 0.5660377358490566,
      "eval_loss": 0.9655478596687317,
      "eval_runtime": 149.5973,
      "eval_samples_per_second": 453.297,
      "eval_steps_per_second": 0.889,
      "step": 300
    },
    {
      "epoch": 0.7547169811320755,
      "grad_norm": 41.89172794379802,
      "learning_rate": 9.551382150256551e-07,
      "loss": 0.9761,
      "step": 400
    },
    {
      "epoch": 0.7547169811320755,
      "eval_loss": 0.976280152797699,
      "eval_runtime": 149.6311,
      "eval_samples_per_second": 453.195,
      "eval_steps_per_second": 0.889,
      "step": 400
    },
    {
      "epoch": 0.9433962264150944,
      "grad_norm": 70.58972777939259,
      "learning_rate": 9.907109755120069e-07,
      "loss": 0.9786,
      "step": 500
    },
    {
      "epoch": 0.9433962264150944,
      "eval_loss": 0.9799396991729736,
      "eval_runtime": 149.6428,
      "eval_samples_per_second": 453.159,
      "eval_steps_per_second": 0.889,
      "step": 500
    },
    {
      "epoch": 1.1320754716981132,
      "grad_norm": 154.52662720587995,
      "learning_rate": 1e-06,
      "loss": 0.9851,
      "step": 600
    },
    {
      "epoch": 1.1320754716981132,
      "eval_loss": 0.9846892952919006,
      "eval_runtime": 150.8433,
      "eval_samples_per_second": 449.553,
      "eval_steps_per_second": 0.882,
      "step": 600
    },
    {
      "epoch": 1.320754716981132,
      "grad_norm": 147.98867672055184,
      "learning_rate": 1e-06,
      "loss": 0.9876,
      "step": 700
    },
    {
      "epoch": 1.320754716981132,
      "eval_loss": 0.9889557957649231,
      "eval_runtime": 149.6471,
      "eval_samples_per_second": 453.146,
      "eval_steps_per_second": 0.889,
      "step": 700
    },
    {
      "epoch": 1.509433962264151,
      "grad_norm": 93.50922205523368,
      "learning_rate": 1e-06,
      "loss": 0.9892,
      "step": 800
    },
    {
      "epoch": 1.509433962264151,
      "eval_loss": 0.9961430430412292,
      "eval_runtime": 149.7573,
      "eval_samples_per_second": 452.813,
      "eval_steps_per_second": 0.888,
      "step": 800
    },
    {
      "epoch": 1.6981132075471699,
      "grad_norm": 53.06237347231676,
      "learning_rate": 1e-06,
      "loss": 0.9983,
      "step": 900
    },
    {
      "epoch": 1.6981132075471699,
      "eval_loss": 1.0119000673294067,
      "eval_runtime": 149.7765,
      "eval_samples_per_second": 452.755,
      "eval_steps_per_second": 0.888,
      "step": 900
    },
    {
      "epoch": 1.8867924528301887,
      "grad_norm": 65.28104430915417,
      "learning_rate": 1e-06,
      "loss": 1.0228,
      "step": 1000
    },
    {
      "epoch": 1.8867924528301887,
      "eval_loss": 1.027756690979004,
      "eval_runtime": 149.4845,
      "eval_samples_per_second": 453.639,
      "eval_steps_per_second": 0.89,
      "step": 1000
    },
    {
      "epoch": 2.0754716981132075,
      "grad_norm": 84.50132037152585,
      "learning_rate": 1e-06,
      "loss": 1.0433,
      "step": 1100
    },
    {
      "epoch": 2.0754716981132075,
      "eval_loss": 1.059239149093628,
      "eval_runtime": 149.7269,
      "eval_samples_per_second": 452.905,
      "eval_steps_per_second": 0.888,
      "step": 1100
    },
    {
      "epoch": 2.2641509433962264,
      "grad_norm": 29.00236133301261,
      "learning_rate": 1e-06,
      "loss": 1.0644,
      "step": 1200
    },
    {
      "epoch": 2.2641509433962264,
      "eval_loss": 1.0800738334655762,
      "eval_runtime": 149.9162,
      "eval_samples_per_second": 452.333,
      "eval_steps_per_second": 0.887,
      "step": 1200
    },
    {
      "epoch": 2.452830188679245,
      "grad_norm": 411.7064146634104,
      "learning_rate": 1e-06,
      "loss": 1.0917,
      "step": 1300
    },
    {
      "epoch": 2.452830188679245,
      "eval_loss": 1.1134124994277954,
      "eval_runtime": 149.7734,
      "eval_samples_per_second": 452.764,
      "eval_steps_per_second": 0.888,
      "step": 1300
    },
    {
      "epoch": 2.641509433962264,
      "grad_norm": 229.88146571929195,
      "learning_rate": 1e-06,
      "loss": 1.1147,
      "step": 1400
    },
    {
      "epoch": 2.641509433962264,
      "eval_loss": 1.1160598993301392,
      "eval_runtime": 149.7257,
      "eval_samples_per_second": 452.908,
      "eval_steps_per_second": 0.888,
      "step": 1400
    },
    {
      "epoch": 2.830188679245283,
      "grad_norm": 7.414451982821549,
      "learning_rate": 1e-06,
      "loss": 1.1581,
      "step": 1500
    },
    {
      "epoch": 2.830188679245283,
      "eval_loss": 1.2751212120056152,
      "eval_runtime": 149.6311,
      "eval_samples_per_second": 453.195,
      "eval_steps_per_second": 0.889,
      "step": 1500
    },
    {
      "epoch": 3.018867924528302,
      "grad_norm": 51.75196724068806,
      "learning_rate": 1e-06,
      "loss": 1.4543,
      "step": 1600
    },
    {
      "epoch": 3.018867924528302,
      "eval_loss": 1.5799795389175415,
      "eval_runtime": 149.5252,
      "eval_samples_per_second": 453.516,
      "eval_steps_per_second": 0.889,
      "step": 1600
    },
    {
      "epoch": 3.207547169811321,
      "grad_norm": 8.197678623516119,
      "learning_rate": 1e-06,
      "loss": 1.6332,
      "step": 1700
    },
    {
      "epoch": 3.207547169811321,
      "eval_loss": 1.7048678398132324,
      "eval_runtime": 149.7414,
      "eval_samples_per_second": 452.861,
      "eval_steps_per_second": 0.888,
      "step": 1700
    },
    {
      "epoch": 3.3962264150943398,
      "grad_norm": 9.609718378646408,
      "learning_rate": 1e-06,
      "loss": 1.7915,
      "step": 1800
    },
    {
      "epoch": 3.3962264150943398,
      "eval_loss": 1.8891370296478271,
      "eval_runtime": 149.8519,
      "eval_samples_per_second": 452.527,
      "eval_steps_per_second": 0.888,
      "step": 1800
    },
    {
      "epoch": 3.5849056603773586,
      "grad_norm": 87.68727169428583,
      "learning_rate": 1e-06,
      "loss": 1.9812,
      "step": 1900
    },
    {
      "epoch": 3.5849056603773586,
      "eval_loss": 2.057527780532837,
      "eval_runtime": 149.7699,
      "eval_samples_per_second": 452.774,
      "eval_steps_per_second": 0.888,
      "step": 1900
    },
    {
      "epoch": 3.7735849056603774,
      "grad_norm": 12.995456781993322,
      "learning_rate": 1e-06,
      "loss": 2.0982,
      "step": 2000
    },
    {
      "epoch": 3.7735849056603774,
      "eval_loss": 2.1332101821899414,
      "eval_runtime": 149.6771,
      "eval_samples_per_second": 453.055,
      "eval_steps_per_second": 0.889,
      "step": 2000
    },
    {
      "epoch": 3.9622641509433962,
      "grad_norm": 13.605685201407708,
      "learning_rate": 1e-06,
      "loss": 2.1784,
      "step": 2100
    },
    {
      "epoch": 3.9622641509433962,
      "eval_loss": 2.2318856716156006,
      "eval_runtime": 149.8102,
      "eval_samples_per_second": 452.653,
      "eval_steps_per_second": 0.888,
      "step": 2100
    },
    {
      "epoch": 4.150943396226415,
      "grad_norm": 12.732750088048045,
      "learning_rate": 1e-06,
      "loss": 2.2661,
      "step": 2200
    },
    {
      "epoch": 4.150943396226415,
      "eval_loss": 2.3092291355133057,
      "eval_runtime": 149.6232,
      "eval_samples_per_second": 453.219,
      "eval_steps_per_second": 0.889,
      "step": 2200
    },
    {
      "epoch": 4.339622641509434,
      "grad_norm": 13.451602224902139,
      "learning_rate": 1e-06,
      "loss": 2.3259,
      "step": 2300
    },
    {
      "epoch": 4.339622641509434,
      "eval_loss": 2.3665931224823,
      "eval_runtime": 149.8265,
      "eval_samples_per_second": 452.604,
      "eval_steps_per_second": 0.888,
      "step": 2300
    },
    {
      "epoch": 4.528301886792453,
      "grad_norm": 13.331046925646502,
      "learning_rate": 1e-06,
      "loss": 2.3908,
      "step": 2400
    },
    {
      "epoch": 4.528301886792453,
      "eval_loss": 2.4305644035339355,
      "eval_runtime": 149.9235,
      "eval_samples_per_second": 452.311,
      "eval_steps_per_second": 0.887,
      "step": 2400
    },
    {
      "epoch": 4.716981132075472,
      "grad_norm": 17.21909294548451,
      "learning_rate": 1e-06,
      "loss": 2.4521,
      "step": 2500
    },
    {
      "epoch": 4.716981132075472,
      "eval_loss": 2.545186758041382,
      "eval_runtime": 149.7293,
      "eval_samples_per_second": 452.897,
      "eval_steps_per_second": 0.888,
      "step": 2500
    },
    {
      "epoch": 4.90566037735849,
      "grad_norm": 16.704993975768705,
      "learning_rate": 1e-06,
      "loss": 2.5541,
      "step": 2600
    },
    {
      "epoch": 4.90566037735849,
      "eval_loss": 2.6449944972991943,
      "eval_runtime": 149.6085,
      "eval_samples_per_second": 453.263,
      "eval_steps_per_second": 0.889,
      "step": 2600
    },
    {
      "epoch": 5.09433962264151,
      "grad_norm": 17.23044971023092,
      "learning_rate": 1e-06,
      "loss": 2.7022,
      "step": 2700
    },
    {
      "epoch": 5.09433962264151,
      "eval_loss": 2.7793149948120117,
      "eval_runtime": 149.7512,
      "eval_samples_per_second": 452.831,
      "eval_steps_per_second": 0.888,
      "step": 2700
    },
    {
      "epoch": 5.283018867924528,
      "grad_norm": 17.47208816215434,
      "learning_rate": 1e-06,
      "loss": 2.8302,
      "step": 2800
    },
    {
      "epoch": 5.283018867924528,
      "eval_loss": 2.888209581375122,
      "eval_runtime": 149.8872,
      "eval_samples_per_second": 452.42,
      "eval_steps_per_second": 0.887,
      "step": 2800
    },
    {
      "epoch": 5.471698113207547,
      "grad_norm": 17.705909664870283,
      "learning_rate": 1e-06,
      "loss": 2.9344,
      "step": 2900
    },
    {
      "epoch": 5.471698113207547,
      "eval_loss": 2.9951322078704834,
      "eval_runtime": 149.9777,
      "eval_samples_per_second": 452.147,
      "eval_steps_per_second": 0.887,
      "step": 2900
    },
    {
      "epoch": 5.660377358490566,
      "grad_norm": 19.101859660265248,
      "learning_rate": 1e-06,
      "loss": 3.0392,
      "step": 3000
    },
    {
      "epoch": 5.660377358490566,
      "eval_loss": 3.0882513523101807,
      "eval_runtime": 149.9153,
      "eval_samples_per_second": 452.335,
      "eval_steps_per_second": 0.887,
      "step": 3000
    },
    {
      "epoch": 5.849056603773585,
      "grad_norm": 19.07389675964658,
      "learning_rate": 1e-06,
      "loss": 3.0997,
      "step": 3100
    },
    {
      "epoch": 5.849056603773585,
      "eval_loss": 3.1218082904815674,
      "eval_runtime": 149.7905,
      "eval_samples_per_second": 452.712,
      "eval_steps_per_second": 0.888,
      "step": 3100
    },
    {
      "epoch": 6.037735849056604,
      "grad_norm": 18.92461175759707,
      "learning_rate": 1e-06,
      "loss": 3.1321,
      "step": 3200
    },
    {
      "epoch": 6.037735849056604,
      "eval_loss": 3.153890609741211,
      "eval_runtime": 149.8617,
      "eval_samples_per_second": 452.497,
      "eval_steps_per_second": 0.887,
      "step": 3200
    },
    {
      "epoch": 6.226415094339623,
      "grad_norm": 20.348062318970786,
      "learning_rate": 1e-06,
      "loss": 3.1567,
      "step": 3300
    },
    {
      "epoch": 6.226415094339623,
      "eval_loss": 3.180058240890503,
      "eval_runtime": 149.6522,
      "eval_samples_per_second": 453.131,
      "eval_steps_per_second": 0.889,
      "step": 3300
    },
    {
      "epoch": 6.415094339622642,
      "grad_norm": 19.380403417398043,
      "learning_rate": 1e-06,
      "loss": 3.1885,
      "step": 3400
    },
    {
      "epoch": 6.415094339622642,
      "eval_loss": 3.2085678577423096,
      "eval_runtime": 149.9193,
      "eval_samples_per_second": 452.323,
      "eval_steps_per_second": 0.887,
      "step": 3400
    },
    {
      "epoch": 6.60377358490566,
      "grad_norm": 19.172157540450666,
      "learning_rate": 1e-06,
      "loss": 3.2066,
      "step": 3500
    },
    {
      "epoch": 6.60377358490566,
      "eval_loss": 3.2430124282836914,
      "eval_runtime": 149.5804,
      "eval_samples_per_second": 453.348,
      "eval_steps_per_second": 0.889,
      "step": 3500
    },
    {
      "epoch": 6.7924528301886795,
      "grad_norm": 20.834808216022473,
      "learning_rate": 1e-06,
      "loss": 3.2529,
      "step": 3600
    },
    {
      "epoch": 6.7924528301886795,
      "eval_loss": 3.2701263427734375,
      "eval_runtime": 149.8287,
      "eval_samples_per_second": 452.597,
      "eval_steps_per_second": 0.888,
      "step": 3600
    },
    {
      "epoch": 6.981132075471698,
      "grad_norm": 20.71072973537811,
      "learning_rate": 1e-06,
      "loss": 3.2788,
      "step": 3700
    },
    {
      "epoch": 6.981132075471698,
      "eval_loss": 3.3030686378479004,
      "eval_runtime": 149.8206,
      "eval_samples_per_second": 452.621,
      "eval_steps_per_second": 0.888,
      "step": 3700
    },
    {
      "epoch": 7.169811320754717,
      "grad_norm": 20.550695492004113,
      "learning_rate": 1e-06,
      "loss": 3.3094,
      "step": 3800
    },
    {
      "epoch": 7.169811320754717,
      "eval_loss": 3.3228628635406494,
      "eval_runtime": 149.8243,
      "eval_samples_per_second": 452.61,
      "eval_steps_per_second": 0.888,
      "step": 3800
    },
    {
      "epoch": 7.3584905660377355,
      "grad_norm": 22.750882645137363,
      "learning_rate": 1e-06,
      "loss": 3.3288,
      "step": 3900
    },
    {
      "epoch": 7.3584905660377355,
      "eval_loss": 3.355827808380127,
      "eval_runtime": 149.6529,
      "eval_samples_per_second": 453.129,
      "eval_steps_per_second": 0.889,
      "step": 3900
    },
    {
      "epoch": 7.547169811320755,
      "grad_norm": 21.426519401026805,
      "learning_rate": 1e-06,
      "loss": 3.3727,
      "step": 4000
    },
    {
      "epoch": 7.547169811320755,
      "eval_loss": 3.3865537643432617,
      "eval_runtime": 150.2322,
      "eval_samples_per_second": 451.381,
      "eval_steps_per_second": 0.885,
      "step": 4000
    },
    {
      "epoch": 7.735849056603773,
      "grad_norm": 21.858010980941717,
      "learning_rate": 1e-06,
      "loss": 3.3938,
      "step": 4100
    },
    {
      "epoch": 7.735849056603773,
      "eval_loss": 3.4126923084259033,
      "eval_runtime": 149.7102,
      "eval_samples_per_second": 452.955,
      "eval_steps_per_second": 0.888,
      "step": 4100
    },
    {
      "epoch": 7.9245283018867925,
      "grad_norm": 21.32876029273619,
      "learning_rate": 1e-06,
      "loss": 3.4334,
      "step": 4200
    },
    {
      "epoch": 7.9245283018867925,
      "eval_loss": 3.4582273960113525,
      "eval_runtime": 149.6771,
      "eval_samples_per_second": 453.055,
      "eval_steps_per_second": 0.889,
      "step": 4200
    },
    {
      "epoch": 8.11320754716981,
      "grad_norm": 23.89601752008716,
      "learning_rate": 1e-06,
      "loss": 3.4769,
      "step": 4300
    },
    {
      "epoch": 8.11320754716981,
      "eval_loss": 3.505073308944702,
      "eval_runtime": 149.8114,
      "eval_samples_per_second": 452.649,
      "eval_steps_per_second": 0.888,
      "step": 4300
    },
    {
      "epoch": 8.30188679245283,
      "grad_norm": 22.427822055394444,
      "learning_rate": 1e-06,
      "loss": 3.5019,
      "step": 4400
    },
    {
      "epoch": 8.30188679245283,
      "eval_loss": 3.5005218982696533,
      "eval_runtime": 149.7083,
      "eval_samples_per_second": 452.961,
      "eval_steps_per_second": 0.888,
      "step": 4400
    },
    {
      "epoch": 8.49056603773585,
      "grad_norm": 20.122035975196678,
      "learning_rate": 1e-06,
      "loss": 3.4841,
      "step": 4500
    },
    {
      "epoch": 8.49056603773585,
      "eval_loss": 3.4907851219177246,
      "eval_runtime": 149.5966,
      "eval_samples_per_second": 453.299,
      "eval_steps_per_second": 0.889,
      "step": 4500
    },
    {
      "epoch": 8.679245283018869,
      "grad_norm": 7.060014135330357,
      "learning_rate": 1e-06,
      "loss": 3.4805,
      "step": 4600
    },
    {
      "epoch": 8.679245283018869,
      "eval_loss": 3.2945797443389893,
      "eval_runtime": 149.8934,
      "eval_samples_per_second": 452.402,
      "eval_steps_per_second": 0.887,
      "step": 4600
    },
    {
      "epoch": 8.867924528301886,
      "grad_norm": 17.493433128940666,
      "learning_rate": 1e-06,
      "loss": 3.332,
      "step": 4700
    },
    {
      "epoch": 8.867924528301886,
      "eval_loss": 3.3459222316741943,
      "eval_runtime": 149.8771,
      "eval_samples_per_second": 452.451,
      "eval_steps_per_second": 0.887,
      "step": 4700
    },
    {
      "epoch": 9.056603773584905,
      "grad_norm": 24.996170360974173,
      "learning_rate": 1e-06,
      "loss": 3.3797,
      "step": 4800
    },
    {
      "epoch": 9.056603773584905,
      "eval_loss": 3.4055655002593994,
      "eval_runtime": 149.7316,
      "eval_samples_per_second": 452.89,
      "eval_steps_per_second": 0.888,
      "step": 4800
    },
    {
      "epoch": 9.245283018867925,
      "grad_norm": 215.6769864867367,
      "learning_rate": 1e-06,
      "loss": 3.4551,
      "step": 4900
    },
    {
      "epoch": 9.245283018867925,
      "eval_loss": 3.5277042388916016,
      "eval_runtime": 149.7551,
      "eval_samples_per_second": 452.819,
      "eval_steps_per_second": 0.888,
      "step": 4900
    },
    {
      "epoch": 9.433962264150944,
      "grad_norm": 329.1368446406449,
      "learning_rate": 1e-06,
      "loss": 3.4127,
      "step": 5000
    },
    {
      "epoch": 9.433962264150944,
      "eval_loss": 3.399594783782959,
      "eval_runtime": 150.0027,
      "eval_samples_per_second": 452.072,
      "eval_steps_per_second": 0.887,
      "step": 5000
    },
    {
      "epoch": 9.622641509433961,
      "grad_norm": 189.47715891368014,
      "learning_rate": 1e-06,
      "loss": 3.5646,
      "step": 5100
    },
    {
      "epoch": 9.622641509433961,
      "eval_loss": 3.6223909854888916,
      "eval_runtime": 149.6317,
      "eval_samples_per_second": 453.193,
      "eval_steps_per_second": 0.889,
      "step": 5100
    },
    {
      "epoch": 9.81132075471698,
      "grad_norm": 154.57704389721005,
      "learning_rate": 1e-06,
      "loss": 3.6037,
      "step": 5200
    },
    {
      "epoch": 9.81132075471698,
      "eval_loss": 3.698052406311035,
      "eval_runtime": 149.5975,
      "eval_samples_per_second": 453.296,
      "eval_steps_per_second": 0.889,
      "step": 5200
    },
    {
      "epoch": 10.0,
      "grad_norm": 201.5213039296342,
      "learning_rate": 1e-06,
      "loss": 3.698,
      "step": 5300
    },
    {
      "epoch": 10.0,
      "eval_loss": 3.721698522567749,
      "eval_runtime": 149.7798,
      "eval_samples_per_second": 452.745,
      "eval_steps_per_second": 0.888,
      "step": 5300
    }
  ],
  "logging_steps": 100,
  "max_steps": 5300,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1664568262656000.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}