{
  "best_metric": 5.664193153381348,
  "best_model_checkpoint": "output_nf_3/checkpoint-1737",
  "epoch": 200.0,
  "eval_steps": 500,
  "global_step": 1800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 9.95e-07,
      "loss": 12.1762,
      "step": 9
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.002932551319648094,
      "eval_loss": 11.818068504333496,
      "eval_runtime": 3.6297,
      "eval_samples_per_second": 0.276,
      "eval_steps_per_second": 0.276,
      "step": 9
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.9e-07,
      "loss": 11.6538,
      "step": 18
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.0009775171065493646,
      "eval_loss": 11.346792221069336,
      "eval_runtime": 3.7372,
      "eval_samples_per_second": 0.268,
      "eval_steps_per_second": 0.268,
      "step": 18
    },
    {
      "epoch": 3.0,
      "learning_rate": 9.849999999999999e-07,
      "loss": 11.2876,
      "step": 27
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.0009775171065493646,
      "eval_loss": 10.912660598754883,
      "eval_runtime": 4.2343,
      "eval_samples_per_second": 0.236,
      "eval_steps_per_second": 0.236,
      "step": 27
    },
    {
      "epoch": 4.0,
      "learning_rate": 9.8e-07,
      "loss": 10.8664,
      "step": 36
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.0009775171065493646,
      "eval_loss": 10.516217231750488,
      "eval_runtime": 3.4285,
      "eval_samples_per_second": 0.292,
      "eval_steps_per_second": 0.292,
      "step": 36
    },
    {
      "epoch": 5.0,
      "learning_rate": 9.75e-07,
      "loss": 10.6184,
      "step": 45
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.0019550342130987292,
      "eval_loss": 10.181687355041504,
      "eval_runtime": 3.9045,
      "eval_samples_per_second": 0.256,
      "eval_steps_per_second": 0.256,
      "step": 45
    },
    {
      "epoch": 6.0,
      "learning_rate": 9.7e-07,
      "loss": 10.3132,
      "step": 54
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.0009775171065493646,
      "eval_loss": 9.876521110534668,
      "eval_runtime": 3.4438,
      "eval_samples_per_second": 0.29,
      "eval_steps_per_second": 0.29,
      "step": 54
    },
    {
      "epoch": 7.0,
      "learning_rate": 9.649999999999999e-07,
      "loss": 9.9087,
      "step": 63
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.0009775171065493646,
      "eval_loss": 9.607266426086426,
      "eval_runtime": 3.7078,
      "eval_samples_per_second": 0.27,
      "eval_steps_per_second": 0.27,
      "step": 63
    },
    {
      "epoch": 8.0,
      "learning_rate": 9.6e-07,
      "loss": 9.6953,
      "step": 72
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.0,
      "eval_loss": 9.324377059936523,
      "eval_runtime": 3.5648,
      "eval_samples_per_second": 0.281,
      "eval_steps_per_second": 0.281,
      "step": 72
    },
    {
      "epoch": 9.0,
      "learning_rate": 9.55e-07,
      "loss": 9.3741,
      "step": 81
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.0,
      "eval_loss": 9.031828880310059,
      "eval_runtime": 3.42,
      "eval_samples_per_second": 0.292,
      "eval_steps_per_second": 0.292,
      "step": 81
    },
    {
      "epoch": 10.0,
      "learning_rate": 9.499999999999999e-07,
      "loss": 9.2045,
      "step": 90
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.0,
      "eval_loss": 8.746149063110352,
      "eval_runtime": 3.746,
      "eval_samples_per_second": 0.267,
      "eval_steps_per_second": 0.267,
      "step": 90
    },
    {
      "epoch": 11.0,
      "learning_rate": 9.45e-07,
      "loss": 8.9079,
      "step": 99
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.0019550342130987292,
      "eval_loss": 8.47550106048584,
      "eval_runtime": 3.3518,
      "eval_samples_per_second": 0.298,
      "eval_steps_per_second": 0.298,
      "step": 99
    },
    {
      "epoch": 12.0,
      "learning_rate": 9.399999999999999e-07,
      "loss": 8.7047,
      "step": 108
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.011730205278592375,
      "eval_loss": 8.222208976745605,
      "eval_runtime": 3.4648,
      "eval_samples_per_second": 0.289,
      "eval_steps_per_second": 0.289,
      "step": 108
    },
    {
      "epoch": 13.0,
      "learning_rate": 9.35e-07,
      "loss": 8.4622,
      "step": 117
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.03128054740957967,
      "eval_loss": 7.995096206665039,
      "eval_runtime": 3.5365,
      "eval_samples_per_second": 0.283,
      "eval_steps_per_second": 0.283,
      "step": 117
    },
    {
      "epoch": 14.0,
      "learning_rate": 9.3e-07,
      "loss": 8.2649,
      "step": 126
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.06549364613880743,
      "eval_loss": 7.805217266082764,
      "eval_runtime": 3.456,
      "eval_samples_per_second": 0.289,
      "eval_steps_per_second": 0.289,
      "step": 126
    },
    {
      "epoch": 15.0,
      "learning_rate": 9.25e-07,
      "loss": 8.043,
      "step": 135
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.10654936461388075,
      "eval_loss": 7.649549961090088,
      "eval_runtime": 3.7554,
      "eval_samples_per_second": 0.266,
      "eval_steps_per_second": 0.266,
      "step": 135
    },
    {
      "epoch": 16.0,
      "learning_rate": 9.2e-07,
      "loss": 7.9092,
      "step": 144
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.1378299120234604,
      "eval_loss": 7.516005516052246,
      "eval_runtime": 4.0508,
      "eval_samples_per_second": 0.247,
      "eval_steps_per_second": 0.247,
      "step": 144
    },
    {
      "epoch": 17.0,
      "learning_rate": 9.15e-07,
      "loss": 7.7103,
      "step": 153
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.16911045943304007,
      "eval_loss": 7.388144016265869,
      "eval_runtime": 3.4529,
      "eval_samples_per_second": 0.29,
      "eval_steps_per_second": 0.29,
      "step": 153
    },
    {
      "epoch": 18.0,
      "learning_rate": 9.1e-07,
      "loss": 7.5701,
      "step": 162
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.18181818181818182,
      "eval_loss": 7.269326686859131,
      "eval_runtime": 3.467,
      "eval_samples_per_second": 0.288,
      "eval_steps_per_second": 0.288,
      "step": 162
    },
    {
      "epoch": 19.0,
      "learning_rate": 9.05e-07,
      "loss": 7.4483,
      "step": 171
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.19941348973607037,
      "eval_loss": 7.154385089874268,
      "eval_runtime": 4.1166,
      "eval_samples_per_second": 0.243,
      "eval_steps_per_second": 0.243,
      "step": 171
    },
    {
      "epoch": 20.0,
      "learning_rate": 9e-07,
      "loss": 7.314,
      "step": 180
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.21603128054740958,
      "eval_loss": 7.046892166137695,
      "eval_runtime": 5.2069,
      "eval_samples_per_second": 0.192,
      "eval_steps_per_second": 0.192,
      "step": 180
    },
    {
      "epoch": 21.0,
      "learning_rate": 8.95e-07,
      "loss": 7.2101,
      "step": 189
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.23264907135874877,
      "eval_loss": 6.9440107345581055,
      "eval_runtime": 3.9239,
      "eval_samples_per_second": 0.255,
      "eval_steps_per_second": 0.255,
      "step": 189
    },
    {
      "epoch": 22.0,
      "learning_rate": 8.9e-07,
      "loss": 7.1026,
      "step": 198
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.23655913978494625,
      "eval_loss": 6.854238033294678,
      "eval_runtime": 3.4311,
      "eval_samples_per_second": 0.291,
      "eval_steps_per_second": 0.291,
      "step": 198
    },
    {
      "epoch": 23.0,
      "learning_rate": 8.85e-07,
      "loss": 6.9954,
      "step": 207
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.2482893450635386,
      "eval_loss": 6.768421173095703,
      "eval_runtime": 4.5026,
      "eval_samples_per_second": 0.222,
      "eval_steps_per_second": 0.222,
      "step": 207
    },
    {
      "epoch": 24.0,
      "learning_rate": 8.799999999999999e-07,
      "loss": 6.9206,
      "step": 216
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.2512218963831867,
      "eval_loss": 6.685108184814453,
      "eval_runtime": 9.9405,
      "eval_samples_per_second": 0.101,
      "eval_steps_per_second": 0.101,
      "step": 216
    },
    {
      "epoch": 25.0,
      "learning_rate": 8.75e-07,
      "loss": 6.8588,
      "step": 225
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.2561094819159335,
      "eval_loss": 6.621737957000732,
      "eval_runtime": 3.7996,
      "eval_samples_per_second": 0.263,
      "eval_steps_per_second": 0.263,
      "step": 225
    },
    {
      "epoch": 26.0,
      "learning_rate": 8.699999999999999e-07,
      "loss": 6.7975,
      "step": 234
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.2590420332355816,
      "eval_loss": 6.562064170837402,
      "eval_runtime": 4.7111,
      "eval_samples_per_second": 0.212,
      "eval_steps_per_second": 0.212,
      "step": 234
    },
    {
      "epoch": 27.0,
      "learning_rate": 8.65e-07,
      "loss": 6.7355,
      "step": 243
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.26099706744868034,
      "eval_loss": 6.5111308097839355,
      "eval_runtime": 3.4887,
      "eval_samples_per_second": 0.287,
      "eval_steps_per_second": 0.287,
      "step": 243
    },
    {
      "epoch": 28.0,
      "learning_rate": 8.599999999999999e-07,
      "loss": 6.6928,
      "step": 252
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.26392961876832843,
      "eval_loss": 6.462972640991211,
      "eval_runtime": 10.4927,
      "eval_samples_per_second": 0.095,
      "eval_steps_per_second": 0.095,
      "step": 252
    },
    {
      "epoch": 29.0,
      "learning_rate": 8.55e-07,
      "loss": 6.6483,
      "step": 261
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.2697947214076246,
      "eval_loss": 6.413713455200195,
      "eval_runtime": 9.1985,
      "eval_samples_per_second": 0.109,
      "eval_steps_per_second": 0.109,
      "step": 261
    },
    {
      "epoch": 30.0,
      "learning_rate": 8.499999999999999e-07,
      "loss": 6.6169,
      "step": 270
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.27174975562072334,
      "eval_loss": 6.367518424987793,
      "eval_runtime": 4.5985,
      "eval_samples_per_second": 0.217,
      "eval_steps_per_second": 0.217,
      "step": 270
    },
    {
      "epoch": 31.0,
      "learning_rate": 8.45e-07,
      "loss": 6.5498,
      "step": 279
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.27468230694037143,
      "eval_loss": 6.329223155975342,
      "eval_runtime": 4.0947,
      "eval_samples_per_second": 0.244,
      "eval_steps_per_second": 0.244,
      "step": 279
    },
    {
      "epoch": 32.0,
      "learning_rate": 8.399999999999999e-07,
      "loss": 6.5385,
      "step": 288
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.27663734115347016,
      "eval_loss": 6.296669960021973,
      "eval_runtime": 3.6593,
      "eval_samples_per_second": 0.273,
      "eval_steps_per_second": 0.273,
      "step": 288
    },
    {
      "epoch": 33.0,
      "learning_rate": 8.349999999999999e-07,
      "loss": 6.5111,
      "step": 297
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.2785923753665689,
      "eval_loss": 6.2592363357543945,
      "eval_runtime": 4.3919,
      "eval_samples_per_second": 0.228,
      "eval_steps_per_second": 0.228,
      "step": 297
    },
    {
      "epoch": 34.0,
      "learning_rate": 8.299999999999999e-07,
      "loss": 6.4748,
      "step": 306
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.28152492668621704,
      "eval_loss": 6.231565475463867,
      "eval_runtime": 4.3688,
      "eval_samples_per_second": 0.229,
      "eval_steps_per_second": 0.229,
      "step": 306
    },
    {
      "epoch": 35.0,
      "learning_rate": 8.249999999999999e-07,
      "loss": 6.4575,
      "step": 315
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.28347996089931576,
      "eval_loss": 6.209580421447754,
      "eval_runtime": 3.5499,
      "eval_samples_per_second": 0.282,
      "eval_steps_per_second": 0.282,
      "step": 315
    },
    {
      "epoch": 36.0,
      "learning_rate": 8.199999999999999e-07,
      "loss": 6.4251,
      "step": 324
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.2844574780058651,
      "eval_loss": 6.184682369232178,
      "eval_runtime": 3.6926,
      "eval_samples_per_second": 0.271,
      "eval_steps_per_second": 0.271,
      "step": 324
    },
    {
      "epoch": 37.0,
      "learning_rate": 8.149999999999999e-07,
      "loss": 6.4096,
      "step": 333
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.2854349951124145,
      "eval_loss": 6.161267280578613,
      "eval_runtime": 3.9736,
      "eval_samples_per_second": 0.252,
      "eval_steps_per_second": 0.252,
      "step": 333
    },
    {
      "epoch": 38.0,
      "learning_rate": 8.1e-07,
      "loss": 6.3741,
      "step": 342
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.2854349951124145,
      "eval_loss": 6.140196800231934,
      "eval_runtime": 3.9847,
      "eval_samples_per_second": 0.251,
      "eval_steps_per_second": 0.251,
      "step": 342
    },
    {
      "epoch": 39.0,
      "learning_rate": 8.05e-07,
      "loss": 6.3645,
      "step": 351
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.2873900293255132,
      "eval_loss": 6.128504753112793,
      "eval_runtime": 3.6581,
      "eval_samples_per_second": 0.273,
      "eval_steps_per_second": 0.273,
      "step": 351
    },
    {
      "epoch": 40.0,
      "learning_rate": 8e-07,
      "loss": 6.3511,
      "step": 360
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.2873900293255132,
      "eval_loss": 6.11335563659668,
      "eval_runtime": 3.9362,
      "eval_samples_per_second": 0.254,
      "eval_steps_per_second": 0.254,
      "step": 360
    },
    {
      "epoch": 41.0,
      "learning_rate": 7.95e-07,
      "loss": 6.3254,
      "step": 369
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.2873900293255132,
      "eval_loss": 6.095658779144287,
      "eval_runtime": 3.8555,
      "eval_samples_per_second": 0.259,
      "eval_steps_per_second": 0.259,
      "step": 369
    },
    {
      "epoch": 42.0,
      "learning_rate": 7.9e-07,
      "loss": 6.3077,
      "step": 378
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.2873900293255132,
      "eval_loss": 6.081845283508301,
      "eval_runtime": 3.4864,
      "eval_samples_per_second": 0.287,
      "eval_steps_per_second": 0.287,
      "step": 378
    },
    {
      "epoch": 43.0,
      "learning_rate": 7.85e-07,
      "loss": 6.301,
      "step": 387
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.2873900293255132,
      "eval_loss": 6.0687761306762695,
      "eval_runtime": 4.7364,
      "eval_samples_per_second": 0.211,
      "eval_steps_per_second": 0.211,
      "step": 387
    },
    {
      "epoch": 44.0,
      "learning_rate": 7.799999999999999e-07,
      "loss": 6.2846,
      "step": 396
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.2883675464320626,
      "eval_loss": 6.051051616668701,
      "eval_runtime": 3.4251,
      "eval_samples_per_second": 0.292,
      "eval_steps_per_second": 0.292,
      "step": 396
    },
    {
      "epoch": 45.0,
      "learning_rate": 7.75e-07,
      "loss": 6.2739,
      "step": 405
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.28934506353861195,
      "eval_loss": 6.039752006530762,
      "eval_runtime": 4.3563,
      "eval_samples_per_second": 0.23,
      "eval_steps_per_second": 0.23,
      "step": 405
    },
    {
      "epoch": 46.0,
      "learning_rate": 7.699999999999999e-07,
      "loss": 6.2569,
      "step": 414
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.2913000977517107,
      "eval_loss": 6.030134677886963,
      "eval_runtime": 3.8175,
      "eval_samples_per_second": 0.262,
      "eval_steps_per_second": 0.262,
      "step": 414
    },
    {
      "epoch": 47.0,
      "learning_rate": 7.65e-07,
      "loss": 6.258,
      "step": 423
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.2913000977517107,
      "eval_loss": 6.01760721206665,
      "eval_runtime": 4.1737,
      "eval_samples_per_second": 0.24,
      "eval_steps_per_second": 0.24,
      "step": 423
    },
    {
      "epoch": 48.0,
      "learning_rate": 7.599999999999999e-07,
      "loss": 6.2273,
      "step": 432
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.2932551319648094,
      "eval_loss": 6.003724575042725,
      "eval_runtime": 3.8811,
      "eval_samples_per_second": 0.258,
      "eval_steps_per_second": 0.258,
      "step": 432
    },
    {
      "epoch": 49.0,
      "learning_rate": 7.55e-07,
      "loss": 6.2256,
      "step": 441
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.2932551319648094,
      "eval_loss": 5.9961748123168945,
      "eval_runtime": 4.331,
      "eval_samples_per_second": 0.231,
      "eval_steps_per_second": 0.231,
      "step": 441
    },
    {
      "epoch": 50.0,
      "learning_rate": 7.5e-07,
      "loss": 6.2065,
      "step": 450
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.2932551319648094,
      "eval_loss": 5.986941337585449,
      "eval_runtime": 4.6185,
      "eval_samples_per_second": 0.217,
      "eval_steps_per_second": 0.217,
      "step": 450
    },
    {
      "epoch": 51.0,
      "learning_rate": 7.45e-07,
      "loss": 6.1991,
      "step": 459
    },
    {
      "epoch": 51.0,
      "eval_accuracy": 0.2932551319648094,
      "eval_loss": 5.975553512573242,
      "eval_runtime": 3.9164,
      "eval_samples_per_second": 0.255,
      "eval_steps_per_second": 0.255,
      "step": 459
    },
    {
      "epoch": 52.0,
      "learning_rate": 7.4e-07,
      "loss": 6.1895,
      "step": 468
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.968191623687744,
      "eval_runtime": 4.4356,
      "eval_samples_per_second": 0.225,
      "eval_steps_per_second": 0.225,
      "step": 468
    },
    {
      "epoch": 53.0,
      "learning_rate": 7.35e-07,
      "loss": 6.1763,
      "step": 477
    },
    {
      "epoch": 53.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.961073398590088,
      "eval_runtime": 4.4638,
      "eval_samples_per_second": 0.224,
      "eval_steps_per_second": 0.224,
      "step": 477
    },
    {
      "epoch": 54.0,
      "learning_rate": 7.3e-07,
      "loss": 6.1734,
      "step": 486
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.953503608703613,
      "eval_runtime": 3.7888,
      "eval_samples_per_second": 0.264,
      "eval_steps_per_second": 0.264,
      "step": 486
    },
    {
      "epoch": 55.0,
      "learning_rate": 7.249999999999999e-07,
      "loss": 6.1702,
      "step": 495
    },
    {
      "epoch": 55.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.94571590423584,
      "eval_runtime": 4.1672,
      "eval_samples_per_second": 0.24,
      "eval_steps_per_second": 0.24,
      "step": 495
    },
    {
      "epoch": 56.0,
      "learning_rate": 7.2e-07,
      "loss": 6.1556,
      "step": 504
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.29521016617790813,
      "eval_loss": 5.937567710876465,
      "eval_runtime": 4.244,
      "eval_samples_per_second": 0.236,
      "eval_steps_per_second": 0.236,
      "step": 504
    },
    {
      "epoch": 57.0,
      "learning_rate": 7.149999999999999e-07,
      "loss": 6.1481,
      "step": 513
    },
    {
      "epoch": 57.0,
      "eval_accuracy": 0.29521016617790813,
      "eval_loss": 5.930552005767822,
      "eval_runtime": 4.4958,
      "eval_samples_per_second": 0.222,
      "eval_steps_per_second": 0.222,
      "step": 513
    },
    {
      "epoch": 58.0,
      "learning_rate": 7.1e-07,
      "loss": 6.1425,
      "step": 522
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.922237396240234,
      "eval_runtime": 4.1228,
      "eval_samples_per_second": 0.243,
      "eval_steps_per_second": 0.243,
      "step": 522
    },
    {
      "epoch": 59.0,
      "learning_rate": 7.049999999999999e-07,
      "loss": 6.1416,
      "step": 531
    },
    {
      "epoch": 59.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.91697883605957,
      "eval_runtime": 4.4905,
      "eval_samples_per_second": 0.223,
      "eval_steps_per_second": 0.223,
      "step": 531
    },
    {
      "epoch": 60.0,
      "learning_rate": 7e-07,
      "loss": 6.1328,
      "step": 540
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.914181232452393,
      "eval_runtime": 4.5555,
      "eval_samples_per_second": 0.22,
      "eval_steps_per_second": 0.22,
      "step": 540
    },
    {
      "epoch": 61.0,
      "learning_rate": 6.949999999999999e-07,
      "loss": 6.1176,
      "step": 549
    },
    {
      "epoch": 61.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.906389236450195,
      "eval_runtime": 4.1516,
      "eval_samples_per_second": 0.241,
      "eval_steps_per_second": 0.241,
      "step": 549
    },
    {
      "epoch": 62.0,
      "learning_rate": 6.9e-07,
      "loss": 6.1091,
      "step": 558
    },
    {
      "epoch": 62.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.901010513305664,
      "eval_runtime": 4.2358,
      "eval_samples_per_second": 0.236,
      "eval_steps_per_second": 0.236,
      "step": 558
    },
    {
      "epoch": 63.0,
      "learning_rate": 6.85e-07,
      "loss": 6.104,
      "step": 567
    },
    {
      "epoch": 63.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.896124362945557,
      "eval_runtime": 4.4857,
      "eval_samples_per_second": 0.223,
      "eval_steps_per_second": 0.223,
      "step": 567
    },
    {
      "epoch": 64.0,
      "learning_rate": 6.800000000000001e-07,
      "loss": 6.0986,
      "step": 576
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.891510486602783,
      "eval_runtime": 4.4644,
      "eval_samples_per_second": 0.224,
      "eval_steps_per_second": 0.224,
      "step": 576
    },
    {
      "epoch": 65.0,
      "learning_rate": 6.75e-07,
      "loss": 6.089,
      "step": 585
    },
    {
      "epoch": 65.0,
      "eval_accuracy": 0.29521016617790813,
      "eval_loss": 5.885429382324219,
      "eval_runtime": 3.9204,
      "eval_samples_per_second": 0.255,
      "eval_steps_per_second": 0.255,
      "step": 585
    },
    {
      "epoch": 66.0,
      "learning_rate": 6.7e-07,
      "loss": 6.0734,
      "step": 594
    },
    {
      "epoch": 66.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.880984783172607,
      "eval_runtime": 4.2715,
      "eval_samples_per_second": 0.234,
      "eval_steps_per_second": 0.234,
      "step": 594
    },
    {
      "epoch": 67.0,
      "learning_rate": 6.65e-07,
      "loss": 6.0905,
      "step": 603
    },
    {
      "epoch": 67.0,
      "eval_accuracy": 0.29423264907135877,
      "eval_loss": 5.876211166381836,
      "eval_runtime": 4.4141,
      "eval_samples_per_second": 0.227,
      "eval_steps_per_second": 0.227,
      "step": 603
    },
    {
      "epoch": 68.0,
      "learning_rate": 6.6e-07,
      "loss": 6.0701,
      "step": 612
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.2961876832844575,
      "eval_loss": 5.873714447021484,
      "eval_runtime": 4.3879,
      "eval_samples_per_second": 0.228,
      "eval_steps_per_second": 0.228,
      "step": 612
    },
    {
      "epoch": 69.0,
      "learning_rate": 6.55e-07,
      "loss": 6.0595,
      "step": 621
    },
    {
      "epoch": 69.0,
      "eval_accuracy": 0.2961876832844575,
      "eval_loss": 5.869368553161621,
      "eval_runtime": 4.2313,
      "eval_samples_per_second": 0.236,
      "eval_steps_per_second": 0.236,
      "step": 621
    },
    {
      "epoch": 70.0,
      "learning_rate": 6.5e-07,
      "loss": 6.0616,
      "step": 630
    },
    {
      "epoch": 70.0,
      "eval_accuracy": 0.2961876832844575,
      "eval_loss": 5.8661274909973145,
      "eval_runtime": 4.1245,
      "eval_samples_per_second": 0.242,
      "eval_steps_per_second": 0.242,
      "step": 630
    },
    {
      "epoch": 71.0,
      "learning_rate": 6.45e-07,
      "loss": 6.0512,
      "step": 639
    },
    {
      "epoch": 71.0,
      "eval_accuracy": 0.2961876832844575,
      "eval_loss": 5.863522052764893,
      "eval_runtime": 4.2876,
      "eval_samples_per_second": 0.233,
      "eval_steps_per_second": 0.233,
      "step": 639
    },
    {
      "epoch": 72.0,
      "learning_rate": 6.4e-07,
      "loss": 6.0415,
      "step": 648
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.2961876832844575,
      "eval_loss": 5.861262321472168,
      "eval_runtime": 4.932,
      "eval_samples_per_second": 0.203,
      "eval_steps_per_second": 0.203,
      "step": 648
    },
    {
      "epoch": 73.0,
      "learning_rate": 6.35e-07,
      "loss": 6.0391,
      "step": 657
    },
    {
      "epoch": 73.0,
      "eval_accuracy": 0.2961876832844575,
      "eval_loss": 5.858279228210449,
      "eval_runtime": 4.4327,
      "eval_samples_per_second": 0.226,
      "eval_steps_per_second": 0.226,
      "step": 657
    },
    {
      "epoch": 74.0,
      "learning_rate": 6.3e-07,
      "loss": 6.032,
      "step": 666
    },
    {
      "epoch": 74.0,
      "eval_accuracy": 0.2961876832844575,
      "eval_loss": 5.854796886444092,
      "eval_runtime": 4.4253,
      "eval_samples_per_second": 0.226,
      "eval_steps_per_second": 0.226,
      "step": 666
    },
    {
      "epoch": 75.0,
      "learning_rate": 6.249999999999999e-07,
      "loss": 6.0317,
      "step": 675
    },
    {
      "epoch": 75.0,
      "eval_accuracy": 0.2961876832844575,
      "eval_loss": 5.851076126098633,
      "eval_runtime": 4.5175,
      "eval_samples_per_second": 0.221,
      "eval_steps_per_second": 0.221,
      "step": 675
    },
    {
      "epoch": 76.0,
      "learning_rate": 6.2e-07,
      "loss": 6.0343,
      "step": 684
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.29716520039100686,
      "eval_loss": 5.847884178161621,
      "eval_runtime": 4.5177,
      "eval_samples_per_second": 0.221,
      "eval_steps_per_second": 0.221,
      "step": 684
    },
    {
      "epoch": 77.0,
      "learning_rate": 6.149999999999999e-07,
      "loss": 6.0156,
      "step": 693
    },
    {
      "epoch": 77.0,
      "eval_accuracy": 0.2981427174975562,
      "eval_loss": 5.843489646911621,
      "eval_runtime": 4.7261,
      "eval_samples_per_second": 0.212,
      "eval_steps_per_second": 0.212,
      "step": 693
    },
    {
      "epoch": 78.0,
      "learning_rate": 6.1e-07,
      "loss": 6.0167,
      "step": 702
    },
    {
      "epoch": 78.0,
      "eval_accuracy": 0.2981427174975562,
      "eval_loss": 5.8403215408325195,
      "eval_runtime": 4.3157,
      "eval_samples_per_second": 0.232,
      "eval_steps_per_second": 0.232,
      "step": 702
    },
    {
      "epoch": 79.0,
      "learning_rate": 6.049999999999999e-07,
      "loss": 6.0052,
      "step": 711
    },
    {
      "epoch": 79.0,
      "eval_accuracy": 0.2981427174975562,
      "eval_loss": 5.83635950088501,
      "eval_runtime": 4.3221,
      "eval_samples_per_second": 0.231,
      "eval_steps_per_second": 0.231,
      "step": 711
    },
    {
      "epoch": 80.0,
      "learning_rate": 6e-07,
      "loss": 6.0057,
      "step": 720
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.2981427174975562,
      "eval_loss": 5.8336262702941895,
      "eval_runtime": 4.8317,
      "eval_samples_per_second": 0.207,
      "eval_steps_per_second": 0.207,
      "step": 720
    },
    {
      "epoch": 81.0,
      "learning_rate": 5.949999999999999e-07,
      "loss": 6.0001,
      "step": 729
    },
    {
      "epoch": 81.0,
      "eval_accuracy": 0.2981427174975562,
      "eval_loss": 5.8319411277771,
      "eval_runtime": 4.5299,
      "eval_samples_per_second": 0.221,
      "eval_steps_per_second": 0.221,
      "step": 729
    },
    {
      "epoch": 82.0,
      "learning_rate": 5.9e-07,
      "loss": 6.001,
      "step": 738
    },
    {
      "epoch": 82.0,
      "eval_accuracy": 0.2991202346041056,
      "eval_loss": 5.828815937042236,
      "eval_runtime": 4.4118,
      "eval_samples_per_second": 0.227,
      "eval_steps_per_second": 0.227,
      "step": 738
    },
    {
      "epoch": 83.0,
      "learning_rate": 5.849999999999999e-07,
      "loss": 5.9905,
      "step": 747
    },
    {
      "epoch": 83.0,
      "eval_accuracy": 0.3010752688172043,
      "eval_loss": 5.826550006866455,
      "eval_runtime": 4.3318,
      "eval_samples_per_second": 0.231,
      "eval_steps_per_second": 0.231,
      "step": 747
    },
    {
      "epoch": 84.0,
      "learning_rate": 5.8e-07,
      "loss": 5.9906,
      "step": 756
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.3020527859237537,
      "eval_loss": 5.824170112609863,
      "eval_runtime": 4.4876,
      "eval_samples_per_second": 0.223,
      "eval_steps_per_second": 0.223,
      "step": 756
    },
    {
      "epoch": 85.0,
      "learning_rate": 5.749999999999999e-07,
      "loss": 5.9862,
      "step": 765
    },
    {
      "epoch": 85.0,
      "eval_accuracy": 0.3020527859237537,
      "eval_loss": 5.821638584136963,
      "eval_runtime": 4.7768,
      "eval_samples_per_second": 0.209,
      "eval_steps_per_second": 0.209,
      "step": 765
    },
    {
      "epoch": 86.0,
      "learning_rate": 5.699999999999999e-07,
      "loss": 5.9829,
      "step": 774
    },
    {
      "epoch": 86.0,
      "eval_accuracy": 0.30498533724340177,
      "eval_loss": 5.819102764129639,
      "eval_runtime": 4.2697,
      "eval_samples_per_second": 0.234,
      "eval_steps_per_second": 0.234,
      "step": 774
    },
    {
      "epoch": 87.0,
      "learning_rate": 5.649999999999999e-07,
      "loss": 5.9725,
      "step": 783
    },
    {
      "epoch": 87.0,
      "eval_accuracy": 0.3069403714565005,
      "eval_loss": 5.817332744598389,
      "eval_runtime": 4.3341,
      "eval_samples_per_second": 0.231,
      "eval_steps_per_second": 0.231,
      "step": 783
    },
    {
      "epoch": 88.0,
      "learning_rate": 5.6e-07,
      "loss": 5.9795,
      "step": 792
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.814181804656982,
      "eval_runtime": 4.759,
      "eval_samples_per_second": 0.21,
      "eval_steps_per_second": 0.21,
      "step": 792
    },
    {
      "epoch": 89.0,
      "learning_rate": 5.55e-07,
      "loss": 5.9695,
      "step": 801
    },
    {
      "epoch": 89.0,
      "eval_accuracy": 0.3069403714565005,
      "eval_loss": 5.811527729034424,
      "eval_runtime": 4.3738,
      "eval_samples_per_second": 0.229,
      "eval_steps_per_second": 0.229,
      "step": 801
    },
    {
      "epoch": 90.0,
      "learning_rate": 5.5e-07,
      "loss": 5.9607,
      "step": 810
    },
    {
      "epoch": 90.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.808058261871338,
      "eval_runtime": 3.4007,
      "eval_samples_per_second": 0.294,
      "eval_steps_per_second": 0.294,
      "step": 810
    },
    {
      "epoch": 91.0,
      "learning_rate": 5.45e-07,
      "loss": 5.9605,
      "step": 819
    },
    {
      "epoch": 91.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.805708885192871,
      "eval_runtime": 3.8195,
      "eval_samples_per_second": 0.262,
      "eval_steps_per_second": 0.262,
      "step": 819
    },
    {
      "epoch": 92.0,
      "learning_rate": 5.4e-07,
      "loss": 5.9591,
      "step": 828
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.80366849899292,
      "eval_runtime": 4.5002,
      "eval_samples_per_second": 0.222,
      "eval_steps_per_second": 0.222,
      "step": 828
    },
    {
      "epoch": 93.0,
      "learning_rate": 5.35e-07,
      "loss": 5.9481,
      "step": 837
    },
    {
      "epoch": 93.0,
      "eval_accuracy": 0.3069403714565005,
      "eval_loss": 5.801074981689453,
      "eval_runtime": 4.3082,
      "eval_samples_per_second": 0.232,
      "eval_steps_per_second": 0.232,
      "step": 837
    },
    {
      "epoch": 94.0,
      "learning_rate": 5.3e-07,
      "loss": 5.9501,
      "step": 846
    },
    {
      "epoch": 94.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.798312664031982,
      "eval_runtime": 4.9824,
      "eval_samples_per_second": 0.201,
      "eval_steps_per_second": 0.201,
      "step": 846
    },
    {
      "epoch": 95.0,
      "learning_rate": 5.25e-07,
      "loss": 5.948,
      "step": 855
    },
    {
      "epoch": 95.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.794301509857178,
      "eval_runtime": 3.717,
      "eval_samples_per_second": 0.269,
      "eval_steps_per_second": 0.269,
      "step": 855
    },
    {
      "epoch": 96.0,
      "learning_rate": 5.2e-07,
      "loss": 5.9488,
      "step": 864
    },
    {
      "epoch": 96.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.790666580200195,
      "eval_runtime": 3.5092,
      "eval_samples_per_second": 0.285,
      "eval_steps_per_second": 0.285,
      "step": 864
    },
    {
      "epoch": 97.0,
      "learning_rate": 5.149999999999999e-07,
      "loss": 5.9449,
      "step": 873
    },
    {
      "epoch": 97.0,
      "eval_accuracy": 0.3118279569892473,
      "eval_loss": 5.788764476776123,
      "eval_runtime": 4.3467,
      "eval_samples_per_second": 0.23,
      "eval_steps_per_second": 0.23,
      "step": 873
    },
    {
      "epoch": 98.0,
      "learning_rate": 5.1e-07,
      "loss": 5.9357,
      "step": 882
    },
    {
      "epoch": 98.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.787247180938721,
      "eval_runtime": 4.2514,
      "eval_samples_per_second": 0.235,
      "eval_steps_per_second": 0.235,
      "step": 882
    },
    {
      "epoch": 99.0,
      "learning_rate": 5.049999999999999e-07,
      "loss": 5.9363,
      "step": 891
    },
    {
      "epoch": 99.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.783356666564941,
      "eval_runtime": 3.6762,
      "eval_samples_per_second": 0.272,
      "eval_steps_per_second": 0.272,
      "step": 891
    },
    {
      "epoch": 100.0,
      "learning_rate": 5e-07,
      "loss": 5.9368,
      "step": 900
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.781695365905762,
      "eval_runtime": 4.5979,
      "eval_samples_per_second": 0.217,
      "eval_steps_per_second": 0.217,
      "step": 900
    },
    {
      "epoch": 101.0,
      "learning_rate": 4.95e-07,
      "loss": 5.9215,
      "step": 909
    },
    {
      "epoch": 101.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.779068946838379,
      "eval_runtime": 4.3432,
      "eval_samples_per_second": 0.23,
      "eval_steps_per_second": 0.23,
      "step": 909
    },
    {
      "epoch": 102.0,
      "learning_rate": 4.9e-07,
      "loss": 5.9264,
      "step": 918
    },
    {
      "epoch": 102.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.77487325668335,
      "eval_runtime": 3.8835,
      "eval_samples_per_second": 0.258,
      "eval_steps_per_second": 0.258,
      "step": 918
    },
    {
      "epoch": 103.0,
      "learning_rate": 4.85e-07,
      "loss": 5.9126,
      "step": 927
    },
    {
      "epoch": 103.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.772488594055176,
      "eval_runtime": 4.0354,
      "eval_samples_per_second": 0.248,
      "eval_steps_per_second": 0.248,
      "step": 927
    },
    {
      "epoch": 104.0,
      "learning_rate": 4.8e-07,
      "loss": 5.9163,
      "step": 936
    },
    {
      "epoch": 104.0,
      "eval_accuracy": 0.3069403714565005,
      "eval_loss": 5.769767761230469,
      "eval_runtime": 3.6621,
      "eval_samples_per_second": 0.273,
      "eval_steps_per_second": 0.273,
      "step": 936
    },
    {
      "epoch": 105.0,
      "learning_rate": 4.7499999999999995e-07,
      "loss": 5.9183,
      "step": 945
    },
    {
      "epoch": 105.0,
      "eval_accuracy": 0.3069403714565005,
      "eval_loss": 5.767302989959717,
      "eval_runtime": 4.3085,
      "eval_samples_per_second": 0.232,
      "eval_steps_per_second": 0.232,
      "step": 945
    },
    {
      "epoch": 106.0,
      "learning_rate": 4.6999999999999995e-07,
      "loss": 5.9154,
      "step": 954
    },
    {
      "epoch": 106.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.763917446136475,
      "eval_runtime": 4.2316,
      "eval_samples_per_second": 0.236,
      "eval_steps_per_second": 0.236,
      "step": 954
    },
    {
      "epoch": 107.0,
      "learning_rate": 4.65e-07,
      "loss": 5.9149,
      "step": 963
    },
    {
      "epoch": 107.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.76121187210083,
      "eval_runtime": 4.5023,
      "eval_samples_per_second": 0.222,
      "eval_steps_per_second": 0.222,
      "step": 963
    },
    {
      "epoch": 108.0,
      "learning_rate": 4.6e-07,
      "loss": 5.9082,
      "step": 972
    },
    {
      "epoch": 108.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.757934093475342,
      "eval_runtime": 4.5081,
      "eval_samples_per_second": 0.222,
      "eval_steps_per_second": 0.222,
      "step": 972
    },
    {
      "epoch": 109.0,
      "learning_rate": 4.55e-07,
      "loss": 5.9051,
      "step": 981
    },
    {
      "epoch": 109.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.755875110626221,
      "eval_runtime": 4.207,
      "eval_samples_per_second": 0.238,
      "eval_steps_per_second": 0.238,
      "step": 981
    },
    {
      "epoch": 110.0,
      "learning_rate": 4.5e-07,
      "loss": 5.908,
      "step": 990
    },
    {
      "epoch": 110.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.753073692321777,
      "eval_runtime": 4.5603,
      "eval_samples_per_second": 0.219,
      "eval_steps_per_second": 0.219,
      "step": 990
    },
    {
      "epoch": 111.0,
      "learning_rate": 4.45e-07,
      "loss": 5.901,
      "step": 999
    },
    {
      "epoch": 111.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.750200271606445,
      "eval_runtime": 4.3073,
      "eval_samples_per_second": 0.232,
      "eval_steps_per_second": 0.232,
      "step": 999
    },
    {
      "epoch": 112.0,
      "learning_rate": 4.3999999999999997e-07,
      "loss": 5.9064,
      "step": 1008
    },
    {
      "epoch": 112.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.748438835144043,
      "eval_runtime": 4.0892,
      "eval_samples_per_second": 0.245,
      "eval_steps_per_second": 0.245,
      "step": 1008
    },
    {
      "epoch": 113.0,
      "learning_rate": 4.3499999999999996e-07,
      "loss": 5.895,
      "step": 1017
    },
    {
      "epoch": 113.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.745972156524658,
      "eval_runtime": 3.6273,
      "eval_samples_per_second": 0.276,
      "eval_steps_per_second": 0.276,
      "step": 1017
    },
    {
      "epoch": 114.0,
      "learning_rate": 4.2999999999999996e-07,
      "loss": 5.894,
      "step": 1026
    },
    {
      "epoch": 114.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.743812561035156,
      "eval_runtime": 3.373,
      "eval_samples_per_second": 0.296,
      "eval_steps_per_second": 0.296,
      "step": 1026
    },
    {
      "epoch": 115.0,
      "learning_rate": 4.2499999999999995e-07,
      "loss": 5.8809,
      "step": 1035
    },
    {
      "epoch": 115.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.741939067840576,
      "eval_runtime": 3.9496,
      "eval_samples_per_second": 0.253,
      "eval_steps_per_second": 0.253,
      "step": 1035
    },
    {
      "epoch": 116.0,
      "learning_rate": 4.1999999999999995e-07,
      "loss": 5.8893,
      "step": 1044
    },
    {
      "epoch": 116.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.74192476272583,
      "eval_runtime": 4.7772,
      "eval_samples_per_second": 0.209,
      "eval_steps_per_second": 0.209,
      "step": 1044
    },
    {
      "epoch": 117.0,
      "learning_rate": 4.1499999999999994e-07,
      "loss": 5.8874,
      "step": 1053
    },
    {
      "epoch": 117.0,
      "eval_accuracy": 0.3069403714565005,
      "eval_loss": 5.739171504974365,
      "eval_runtime": 4.0083,
      "eval_samples_per_second": 0.249,
      "eval_steps_per_second": 0.249,
      "step": 1053
    },
    {
      "epoch": 118.0,
      "learning_rate": 4.0999999999999994e-07,
      "loss": 5.8798,
      "step": 1062
    },
    {
      "epoch": 118.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.73491907119751,
      "eval_runtime": 4.1773,
      "eval_samples_per_second": 0.239,
      "eval_steps_per_second": 0.239,
      "step": 1062
    },
    {
      "epoch": 119.0,
      "learning_rate": 4.05e-07,
      "loss": 5.8826,
      "step": 1071
    },
    {
      "epoch": 119.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.732446193695068,
      "eval_runtime": 4.4141,
      "eval_samples_per_second": 0.227,
      "eval_steps_per_second": 0.227,
      "step": 1071
    },
    {
      "epoch": 120.0,
      "learning_rate": 4e-07,
      "loss": 5.8736,
      "step": 1080
    },
    {
      "epoch": 120.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.729852199554443,
      "eval_runtime": 4.4117,
      "eval_samples_per_second": 0.227,
      "eval_steps_per_second": 0.227,
      "step": 1080
    },
    {
      "epoch": 121.0,
      "learning_rate": 3.95e-07,
      "loss": 5.8751,
      "step": 1089
    },
    {
      "epoch": 121.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.726986408233643,
      "eval_runtime": 4.9388,
      "eval_samples_per_second": 0.202,
      "eval_steps_per_second": 0.202,
      "step": 1089
    },
    {
      "epoch": 122.0,
      "learning_rate": 3.8999999999999997e-07,
      "loss": 5.8699,
      "step": 1098
    },
    {
      "epoch": 122.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.72526216506958,
      "eval_runtime": 4.0541,
      "eval_samples_per_second": 0.247,
      "eval_steps_per_second": 0.247,
      "step": 1098
    },
    {
      "epoch": 123.0,
      "learning_rate": 3.8499999999999997e-07,
      "loss": 5.8802,
      "step": 1107
    },
    {
      "epoch": 123.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.723138809204102,
      "eval_runtime": 4.1908,
      "eval_samples_per_second": 0.239,
      "eval_steps_per_second": 0.239,
      "step": 1107
    },
    {
      "epoch": 124.0,
      "learning_rate": 3.7999999999999996e-07,
      "loss": 5.8707,
      "step": 1116
    },
    {
      "epoch": 124.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.7239861488342285,
      "eval_runtime": 4.4694,
      "eval_samples_per_second": 0.224,
      "eval_steps_per_second": 0.224,
      "step": 1116
    },
    {
      "epoch": 125.0,
      "learning_rate": 3.75e-07,
      "loss": 5.8653,
      "step": 1125
    },
    {
      "epoch": 125.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.723176002502441,
      "eval_runtime": 4.2543,
      "eval_samples_per_second": 0.235,
      "eval_steps_per_second": 0.235,
      "step": 1125
    },
    {
      "epoch": 126.0,
      "learning_rate": 3.7e-07,
      "loss": 5.8693,
      "step": 1134
    },
    {
      "epoch": 126.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.7180495262146,
      "eval_runtime": 4.3031,
      "eval_samples_per_second": 0.232,
      "eval_steps_per_second": 0.232,
      "step": 1134
    },
    {
      "epoch": 127.0,
      "learning_rate": 3.65e-07,
      "loss": 5.8662,
      "step": 1143
    },
    {
      "epoch": 127.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.714703559875488,
      "eval_runtime": 4.1167,
      "eval_samples_per_second": 0.243,
      "eval_steps_per_second": 0.243,
      "step": 1143
    },
    {
      "epoch": 128.0,
      "learning_rate": 3.6e-07,
      "loss": 5.8539,
      "step": 1152
    },
    {
      "epoch": 128.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.713204383850098,
      "eval_runtime": 4.0648,
      "eval_samples_per_second": 0.246,
      "eval_steps_per_second": 0.246,
      "step": 1152
    },
    {
      "epoch": 129.0,
      "learning_rate": 3.55e-07,
      "loss": 5.8611,
      "step": 1161
    },
    {
      "epoch": 129.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.712745189666748,
      "eval_runtime": 4.0643,
      "eval_samples_per_second": 0.246,
      "eval_steps_per_second": 0.246,
      "step": 1161
    },
    {
      "epoch": 130.0,
      "learning_rate": 3.5e-07,
      "loss": 5.8495,
      "step": 1170
    },
    {
      "epoch": 130.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.713523864746094,
      "eval_runtime": 4.0466,
      "eval_samples_per_second": 0.247,
      "eval_steps_per_second": 0.247,
      "step": 1170
    },
    {
      "epoch": 131.0,
      "learning_rate": 3.45e-07,
      "loss": 5.8602,
      "step": 1179
    },
    {
      "epoch": 131.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.711104393005371,
      "eval_runtime": 3.9889,
      "eval_samples_per_second": 0.251,
      "eval_steps_per_second": 0.251,
      "step": 1179
    },
    {
      "epoch": 132.0,
      "learning_rate": 3.4000000000000003e-07,
      "loss": 5.8512,
      "step": 1188
    },
    {
      "epoch": 132.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.7077484130859375,
      "eval_runtime": 3.6772,
      "eval_samples_per_second": 0.272,
      "eval_steps_per_second": 0.272,
      "step": 1188
    },
    {
      "epoch": 133.0,
      "learning_rate": 3.35e-07,
      "loss": 5.8493,
      "step": 1197
    },
    {
      "epoch": 133.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.704965591430664,
      "eval_runtime": 4.3621,
      "eval_samples_per_second": 0.229,
      "eval_steps_per_second": 0.229,
      "step": 1197
    },
    {
      "epoch": 134.0,
      "learning_rate": 3.3e-07,
      "loss": 5.8477,
      "step": 1206
    },
    {
      "epoch": 134.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.704052925109863,
      "eval_runtime": 4.4972,
      "eval_samples_per_second": 0.222,
      "eval_steps_per_second": 0.222,
      "step": 1206
    },
    {
      "epoch": 135.0,
      "learning_rate": 3.25e-07,
      "loss": 5.8464,
      "step": 1215
    },
    {
      "epoch": 135.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.704152584075928,
      "eval_runtime": 3.8106,
      "eval_samples_per_second": 0.262,
      "eval_steps_per_second": 0.262,
      "step": 1215
    },
    {
      "epoch": 136.0,
      "learning_rate": 3.2e-07,
      "loss": 5.8459,
      "step": 1224
    },
    {
      "epoch": 136.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.702281951904297,
      "eval_runtime": 4.7548,
      "eval_samples_per_second": 0.21,
      "eval_steps_per_second": 0.21,
      "step": 1224
    },
    {
      "epoch": 137.0,
      "learning_rate": 3.15e-07,
      "loss": 5.8475,
      "step": 1233
    },
    {
      "epoch": 137.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.7000274658203125,
      "eval_runtime": 4.1681,
      "eval_samples_per_second": 0.24,
      "eval_steps_per_second": 0.24,
      "step": 1233
    },
    {
      "epoch": 138.0,
      "learning_rate": 3.1e-07,
      "loss": 5.8384,
      "step": 1242
    },
    {
      "epoch": 138.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.698208332061768,
      "eval_runtime": 4.6023,
      "eval_samples_per_second": 0.217,
      "eval_steps_per_second": 0.217,
      "step": 1242
    },
    {
      "epoch": 139.0,
      "learning_rate": 3.05e-07,
      "loss": 5.8453,
      "step": 1251
    },
    {
      "epoch": 139.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.697631359100342,
      "eval_runtime": 3.8429,
      "eval_samples_per_second": 0.26,
      "eval_steps_per_second": 0.26,
      "step": 1251
    },
    {
      "epoch": 140.0,
      "learning_rate": 3e-07,
      "loss": 5.8441,
      "step": 1260
    },
    {
      "epoch": 140.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.697261810302734,
      "eval_runtime": 4.2789,
      "eval_samples_per_second": 0.234,
      "eval_steps_per_second": 0.234,
      "step": 1260
    },
    {
      "epoch": 141.0,
      "learning_rate": 2.95e-07,
      "loss": 5.838,
      "step": 1269
    },
    {
      "epoch": 141.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.697081565856934,
      "eval_runtime": 4.6381,
      "eval_samples_per_second": 0.216,
      "eval_steps_per_second": 0.216,
      "step": 1269
    },
    {
      "epoch": 142.0,
      "learning_rate": 2.9e-07,
      "loss": 5.8463,
      "step": 1278
    },
    {
      "epoch": 142.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.695047855377197,
      "eval_runtime": 4.2254,
      "eval_samples_per_second": 0.237,
      "eval_steps_per_second": 0.237,
      "step": 1278
    },
    {
      "epoch": 143.0,
      "learning_rate": 2.8499999999999997e-07,
      "loss": 5.8385,
      "step": 1287
    },
    {
      "epoch": 143.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.692078113555908,
      "eval_runtime": 3.9314,
      "eval_samples_per_second": 0.254,
      "eval_steps_per_second": 0.254,
      "step": 1287
    },
    {
      "epoch": 144.0,
      "learning_rate": 2.8e-07,
      "loss": 5.8354,
      "step": 1296
    },
    {
      "epoch": 144.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.69093656539917,
      "eval_runtime": 4.4756,
      "eval_samples_per_second": 0.223,
      "eval_steps_per_second": 0.223,
      "step": 1296
    },
    {
      "epoch": 145.0,
      "learning_rate": 2.75e-07,
      "loss": 5.8283,
      "step": 1305
    },
    {
      "epoch": 145.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.690831184387207,
      "eval_runtime": 4.2673,
      "eval_samples_per_second": 0.234,
      "eval_steps_per_second": 0.234,
      "step": 1305
    },
    {
      "epoch": 146.0,
      "learning_rate": 2.7e-07,
      "loss": 5.8363,
      "step": 1314
    },
    {
      "epoch": 146.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.6901655197143555,
      "eval_runtime": 4.2991,
      "eval_samples_per_second": 0.233,
      "eval_steps_per_second": 0.233,
      "step": 1314
    },
    {
      "epoch": 147.0,
      "learning_rate": 2.65e-07,
      "loss": 5.8433,
      "step": 1323
    },
    {
      "epoch": 147.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.689047813415527,
      "eval_runtime": 4.5899,
      "eval_samples_per_second": 0.218,
      "eval_steps_per_second": 0.218,
      "step": 1323
    },
    {
      "epoch": 148.0,
      "learning_rate": 2.6e-07,
      "loss": 5.8302,
      "step": 1332
    },
    {
      "epoch": 148.0,
      "eval_accuracy": 0.30791788856304986,
      "eval_loss": 5.689029216766357,
      "eval_runtime": 4.2306,
      "eval_samples_per_second": 0.236,
      "eval_steps_per_second": 0.236,
      "step": 1332
    },
    {
      "epoch": 149.0,
      "learning_rate": 2.55e-07,
      "loss": 5.8276,
      "step": 1341
    },
    {
      "epoch": 149.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.688091278076172,
      "eval_runtime": 4.1426,
      "eval_samples_per_second": 0.241,
      "eval_steps_per_second": 0.241,
      "step": 1341
    },
    {
      "epoch": 150.0,
      "learning_rate": 2.5e-07,
      "loss": 5.8366,
      "step": 1350
    },
    {
      "epoch": 150.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.686432838439941,
      "eval_runtime": 4.4478,
      "eval_samples_per_second": 0.225,
      "eval_steps_per_second": 0.225,
      "step": 1350
    },
    {
      "epoch": 151.0,
      "learning_rate": 2.45e-07,
      "loss": 5.826,
      "step": 1359
    },
    {
      "epoch": 151.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.685244083404541,
      "eval_runtime": 4.0347,
      "eval_samples_per_second": 0.248,
      "eval_steps_per_second": 0.248,
      "step": 1359
    },
    {
      "epoch": 152.0,
      "learning_rate": 2.4e-07,
      "loss": 5.8293,
      "step": 1368
    },
    {
      "epoch": 152.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.68437385559082,
      "eval_runtime": 3.7004,
      "eval_samples_per_second": 0.27,
      "eval_steps_per_second": 0.27,
      "step": 1368
    },
    {
      "epoch": 153.0,
      "learning_rate": 2.3499999999999997e-07,
      "loss": 5.8278,
      "step": 1377
    },
    {
      "epoch": 153.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.683380603790283,
      "eval_runtime": 4.0294,
      "eval_samples_per_second": 0.248,
      "eval_steps_per_second": 0.248,
      "step": 1377
    },
    {
      "epoch": 154.0,
      "learning_rate": 2.3e-07,
      "loss": 5.8239,
      "step": 1386
    },
    {
      "epoch": 154.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.683019638061523,
      "eval_runtime": 4.1012,
      "eval_samples_per_second": 0.244,
      "eval_steps_per_second": 0.244,
      "step": 1386
    },
    {
      "epoch": 155.0,
      "learning_rate": 2.25e-07,
      "loss": 5.8262,
      "step": 1395
    },
    {
      "epoch": 155.0,
      "eval_accuracy": 0.3088954056695992,
      "eval_loss": 5.681789875030518,
      "eval_runtime": 4.3436,
      "eval_samples_per_second": 0.23,
      "eval_steps_per_second": 0.23,
      "step": 1395
    },
    {
      "epoch": 156.0,
      "learning_rate": 2.1999999999999998e-07,
      "loss": 5.8253,
      "step": 1404
    },
    {
      "epoch": 156.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.680798530578613,
      "eval_runtime": 3.5916,
      "eval_samples_per_second": 0.278,
      "eval_steps_per_second": 0.278,
      "step": 1404
    },
    {
      "epoch": 157.0,
      "learning_rate": 2.1499999999999998e-07,
      "loss": 5.8169,
      "step": 1413
    },
    {
      "epoch": 157.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.679258346557617,
      "eval_runtime": 5.377,
      "eval_samples_per_second": 0.186,
      "eval_steps_per_second": 0.186,
      "step": 1413
    },
    {
      "epoch": 158.0,
      "learning_rate": 2.0999999999999997e-07,
      "loss": 5.8201,
      "step": 1422
    },
    {
      "epoch": 158.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.679547309875488,
      "eval_runtime": 3.4728,
      "eval_samples_per_second": 0.288,
      "eval_steps_per_second": 0.288,
      "step": 1422
    },
    {
      "epoch": 159.0,
      "learning_rate": 2.0499999999999997e-07,
      "loss": 5.8077,
      "step": 1431
    },
    {
      "epoch": 159.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.6798624992370605,
      "eval_runtime": 4.4591,
      "eval_samples_per_second": 0.224,
      "eval_steps_per_second": 0.224,
      "step": 1431
    },
    {
      "epoch": 160.0,
      "learning_rate": 2e-07,
      "loss": 5.8222,
      "step": 1440
    },
    {
      "epoch": 160.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.678928852081299,
      "eval_runtime": 3.9956,
      "eval_samples_per_second": 0.25,
      "eval_steps_per_second": 0.25,
      "step": 1440
    },
    {
      "epoch": 161.0,
      "learning_rate": 1.9499999999999999e-07,
      "loss": 5.8191,
      "step": 1449
    },
    {
      "epoch": 161.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.677757263183594,
      "eval_runtime": 3.5865,
      "eval_samples_per_second": 0.279,
      "eval_steps_per_second": 0.279,
      "step": 1449
    },
    {
      "epoch": 162.0,
      "learning_rate": 1.8999999999999998e-07,
      "loss": 5.83,
      "step": 1458
    },
    {
      "epoch": 162.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.676819801330566,
      "eval_runtime": 4.7244,
      "eval_samples_per_second": 0.212,
      "eval_steps_per_second": 0.212,
      "step": 1458
    },
    {
      "epoch": 163.0,
      "learning_rate": 1.85e-07,
      "loss": 5.8183,
      "step": 1467
    },
    {
      "epoch": 163.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.675673484802246,
      "eval_runtime": 4.423,
      "eval_samples_per_second": 0.226,
      "eval_steps_per_second": 0.226,
      "step": 1467
    },
    {
      "epoch": 164.0,
      "learning_rate": 1.8e-07,
      "loss": 5.8124,
      "step": 1476
    },
    {
      "epoch": 164.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.674676418304443,
      "eval_runtime": 4.507,
      "eval_samples_per_second": 0.222,
      "eval_steps_per_second": 0.222,
      "step": 1476
    },
    {
      "epoch": 165.0,
      "learning_rate": 1.75e-07,
      "loss": 5.8119,
      "step": 1485
    },
    {
      "epoch": 165.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.674499034881592,
      "eval_runtime": 4.5793,
      "eval_samples_per_second": 0.218,
      "eval_steps_per_second": 0.218,
      "step": 1485
    },
    {
      "epoch": 166.0,
      "learning_rate": 1.7000000000000001e-07,
      "loss": 5.821,
      "step": 1494
    },
    {
      "epoch": 166.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.674384117126465,
      "eval_runtime": 4.1895,
      "eval_samples_per_second": 0.239,
      "eval_steps_per_second": 0.239,
      "step": 1494
    },
    {
      "epoch": 167.0,
      "learning_rate": 1.65e-07,
      "loss": 5.807,
      "step": 1503
    },
    {
      "epoch": 167.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.6736602783203125,
      "eval_runtime": 3.6058,
      "eval_samples_per_second": 0.277,
      "eval_steps_per_second": 0.277,
      "step": 1503
    },
    {
      "epoch": 168.0,
      "learning_rate": 1.6e-07,
      "loss": 5.8177,
      "step": 1512
    },
    {
      "epoch": 168.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.672515392303467,
      "eval_runtime": 4.5742,
      "eval_samples_per_second": 0.219,
      "eval_steps_per_second": 0.219,
      "step": 1512
    },
    {
      "epoch": 169.0,
      "learning_rate": 1.55e-07,
      "loss": 5.8046,
      "step": 1521
    },
    {
      "epoch": 169.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.671017169952393,
      "eval_runtime": 3.599,
      "eval_samples_per_second": 0.278,
      "eval_steps_per_second": 0.278,
      "step": 1521
    },
    {
      "epoch": 170.0,
      "learning_rate": 1.5e-07,
      "loss": 5.8093,
      "step": 1530
    },
    {
      "epoch": 170.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.670760154724121,
      "eval_runtime": 3.7495,
      "eval_samples_per_second": 0.267,
      "eval_steps_per_second": 0.267,
      "step": 1530
    },
    {
      "epoch": 171.0,
      "learning_rate": 1.45e-07,
      "loss": 5.8145,
      "step": 1539
    },
    {
      "epoch": 171.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.6709699630737305,
      "eval_runtime": 4.4875,
      "eval_samples_per_second": 0.223,
      "eval_steps_per_second": 0.223,
      "step": 1539
    },
    {
      "epoch": 172.0,
      "learning_rate": 1.4e-07,
      "loss": 5.803,
      "step": 1548
    },
    {
      "epoch": 172.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.67042350769043,
      "eval_runtime": 4.2512,
      "eval_samples_per_second": 0.235,
      "eval_steps_per_second": 0.235,
      "step": 1548
    },
    {
      "epoch": 173.0,
      "learning_rate": 1.35e-07,
      "loss": 5.8038,
      "step": 1557
    },
    {
      "epoch": 173.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.6697306632995605,
      "eval_runtime": 3.7575,
      "eval_samples_per_second": 0.266,
      "eval_steps_per_second": 0.266,
      "step": 1557
    },
    {
      "epoch": 174.0,
      "learning_rate": 1.3e-07,
      "loss": 5.807,
      "step": 1566
    },
    {
      "epoch": 174.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.668872356414795,
      "eval_runtime": 3.653,
      "eval_samples_per_second": 0.274,
      "eval_steps_per_second": 0.274,
      "step": 1566
    },
    {
      "epoch": 175.0,
      "learning_rate": 1.25e-07,
      "loss": 5.7974,
      "step": 1575
    },
    {
      "epoch": 175.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.668323993682861,
      "eval_runtime": 3.3471,
      "eval_samples_per_second": 0.299,
      "eval_steps_per_second": 0.299,
      "step": 1575
    },
    {
      "epoch": 176.0,
      "learning_rate": 1.2e-07,
      "loss": 5.8089,
      "step": 1584
    },
    {
      "epoch": 176.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.66883659362793,
      "eval_runtime": 3.7471,
      "eval_samples_per_second": 0.267,
      "eval_steps_per_second": 0.267,
      "step": 1584
    },
    {
      "epoch": 177.0,
      "learning_rate": 1.15e-07,
      "loss": 5.8067,
      "step": 1593
    },
    {
      "epoch": 177.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.669277667999268,
      "eval_runtime": 3.325,
      "eval_samples_per_second": 0.301,
      "eval_steps_per_second": 0.301,
      "step": 1593
    },
    {
      "epoch": 178.0,
      "learning_rate": 1.0999999999999999e-07,
      "loss": 5.8092,
      "step": 1602
    },
    {
      "epoch": 178.0,
      "eval_accuracy": 0.31085043988269795,
      "eval_loss": 5.669507026672363,
      "eval_runtime": 3.3536,
      "eval_samples_per_second": 0.298,
      "eval_steps_per_second": 0.298,
      "step": 1602
    },
    {
      "epoch": 179.0,
      "learning_rate": 1.0499999999999999e-07,
      "loss": 5.8047,
      "step": 1611
    },
    {
      "epoch": 179.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.668713092803955,
      "eval_runtime": 4.0451,
      "eval_samples_per_second": 0.247,
      "eval_steps_per_second": 0.247,
      "step": 1611
    },
    {
      "epoch": 180.0,
      "learning_rate": 1e-07,
      "loss": 5.8007,
      "step": 1620
    },
    {
      "epoch": 180.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.667922019958496,
      "eval_runtime": 4.4531,
      "eval_samples_per_second": 0.225,
      "eval_steps_per_second": 0.225,
      "step": 1620
    },
    {
      "epoch": 181.0,
      "learning_rate": 9.499999999999999e-08,
      "loss": 5.8041,
      "step": 1629
    },
    {
      "epoch": 181.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.667240619659424,
      "eval_runtime": 4.9291,
      "eval_samples_per_second": 0.203,
      "eval_steps_per_second": 0.203,
      "step": 1629
    },
    {
      "epoch": 182.0,
      "learning_rate": 9e-08,
      "loss": 5.8072,
      "step": 1638
    },
    {
      "epoch": 182.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.666664123535156,
      "eval_runtime": 3.1517,
      "eval_samples_per_second": 0.317,
      "eval_steps_per_second": 0.317,
      "step": 1638
    },
    {
      "epoch": 183.0,
      "learning_rate": 8.500000000000001e-08,
      "loss": 5.8093,
      "step": 1647
    },
    {
      "epoch": 183.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.66619873046875,
      "eval_runtime": 3.528,
      "eval_samples_per_second": 0.283,
      "eval_steps_per_second": 0.283,
      "step": 1647
    },
    {
      "epoch": 184.0,
      "learning_rate": 8e-08,
      "loss": 5.7948,
      "step": 1656
    },
    {
      "epoch": 184.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.665794372558594,
      "eval_runtime": 3.2036,
      "eval_samples_per_second": 0.312,
      "eval_steps_per_second": 0.312,
      "step": 1656
    },
    {
      "epoch": 185.0,
      "learning_rate": 7.5e-08,
      "loss": 5.7968,
      "step": 1665
    },
    {
      "epoch": 185.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.665581703186035,
      "eval_runtime": 4.064,
      "eval_samples_per_second": 0.246,
      "eval_steps_per_second": 0.246,
      "step": 1665
    },
    {
      "epoch": 186.0,
      "learning_rate": 7e-08,
      "loss": 5.8033,
      "step": 1674
    },
    {
      "epoch": 186.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.665287494659424,
      "eval_runtime": 3.7071,
      "eval_samples_per_second": 0.27,
      "eval_steps_per_second": 0.27,
      "step": 1674
    },
    {
      "epoch": 187.0,
      "learning_rate": 6.5e-08,
      "loss": 5.8031,
      "step": 1683
    },
    {
      "epoch": 187.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.665118217468262,
      "eval_runtime": 3.8502,
      "eval_samples_per_second": 0.26,
      "eval_steps_per_second": 0.26,
      "step": 1683
    },
    {
      "epoch": 188.0,
      "learning_rate": 6e-08,
      "loss": 5.7953,
      "step": 1692
    },
    {
      "epoch": 188.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.664963722229004,
      "eval_runtime": 3.5099,
      "eval_samples_per_second": 0.285,
      "eval_steps_per_second": 0.285,
      "step": 1692
    },
    {
      "epoch": 189.0,
      "learning_rate": 5.4999999999999996e-08,
      "loss": 5.8085,
      "step": 1701
    },
    {
      "epoch": 189.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.6647257804870605,
      "eval_runtime": 3.1431,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 1701
    },
    {
      "epoch": 190.0,
      "learning_rate": 5e-08,
      "loss": 5.8021,
      "step": 1710
    },
    {
      "epoch": 190.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.664588928222656,
      "eval_runtime": 3.9583,
      "eval_samples_per_second": 0.253,
      "eval_steps_per_second": 0.253,
      "step": 1710
    },
    {
      "epoch": 191.0,
      "learning_rate": 4.5e-08,
      "loss": 5.7995,
      "step": 1719
    },
    {
      "epoch": 191.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.664261817932129,
      "eval_runtime": 3.9361,
      "eval_samples_per_second": 0.254,
      "eval_steps_per_second": 0.254,
      "step": 1719
    },
    {
      "epoch": 192.0,
      "learning_rate": 4e-08,
      "loss": 5.8057,
      "step": 1728
    },
    {
      "epoch": 192.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.664216041564941,
      "eval_runtime": 3.9765,
      "eval_samples_per_second": 0.251,
      "eval_steps_per_second": 0.251,
      "step": 1728
    },
    {
      "epoch": 193.0,
      "learning_rate": 3.5e-08,
      "loss": 5.7989,
      "step": 1737
    },
    {
      "epoch": 193.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.664193153381348,
      "eval_runtime": 3.0391,
      "eval_samples_per_second": 0.329,
      "eval_steps_per_second": 0.329,
      "step": 1737
    },
    {
      "epoch": 194.0,
      "learning_rate": 3e-08,
      "loss": 5.7977,
      "step": 1746
    },
    {
      "epoch": 194.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.664215564727783,
      "eval_runtime": 3.3818,
      "eval_samples_per_second": 0.296,
      "eval_steps_per_second": 0.296,
      "step": 1746
    },
    {
      "epoch": 195.0,
      "learning_rate": 2.5e-08,
      "loss": 5.8009,
      "step": 1755
    },
    {
      "epoch": 195.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.664393901824951,
      "eval_runtime": 3.3657,
      "eval_samples_per_second": 0.297,
      "eval_steps_per_second": 0.297,
      "step": 1755
    },
    {
      "epoch": 196.0,
      "learning_rate": 2e-08,
      "loss": 5.7988,
      "step": 1764
    },
    {
      "epoch": 196.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.6645026206970215,
      "eval_runtime": 3.9312,
      "eval_samples_per_second": 0.254,
      "eval_steps_per_second": 0.254,
      "step": 1764
    },
    {
      "epoch": 197.0,
      "learning_rate": 1.5e-08,
      "loss": 5.8016,
      "step": 1773
    },
    {
      "epoch": 197.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.664463043212891,
      "eval_runtime": 3.7271,
      "eval_samples_per_second": 0.268,
      "eval_steps_per_second": 0.268,
      "step": 1773
    },
    {
      "epoch": 198.0,
      "learning_rate": 1e-08,
      "loss": 5.7929,
      "step": 1782
    },
    {
      "epoch": 198.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.664458751678467,
      "eval_runtime": 3.5464,
      "eval_samples_per_second": 0.282,
      "eval_steps_per_second": 0.282,
      "step": 1782
    },
    {
      "epoch": 199.0,
      "learning_rate": 5e-09,
      "loss": 5.7973,
      "step": 1791
    },
    {
      "epoch": 199.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.664454936981201,
      "eval_runtime": 3.6778,
      "eval_samples_per_second": 0.272,
      "eval_steps_per_second": 0.272,
      "step": 1791
    },
    {
      "epoch": 200.0,
      "learning_rate": 0.0,
      "loss": 5.8022,
      "step": 1800
    },
    {
      "epoch": 200.0,
      "eval_accuracy": 0.3098729227761486,
      "eval_loss": 5.664452075958252,
      "eval_runtime": 3.5398,
      "eval_samples_per_second": 0.283,
      "eval_steps_per_second": 0.283,
      "step": 1800
    },
    {
      "epoch": 200.0,
      "step": 1800,
      "total_flos": 7834231111680000.0,
      "train_loss": 6.335995424058702,
      "train_runtime": 30730.8013,
      "train_samples_per_second": 0.059,
      "train_steps_per_second": 0.059
    }
  ],
  "logging_steps": 500,
  "max_steps": 1800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 200,
  "save_steps": 500,
  "total_flos": 7834231111680000.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}