{
  "best_metric": 1.0718672275543213,
  "best_model_checkpoint": "outputs/checkpoint-602",
  "epoch": 16.885290148448043,
  "eval_steps": 500,
  "global_step": 782,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4318488529014845,
      "grad_norm": 0.9060143232345581,
      "learning_rate": 6e-06,
      "loss": 2.3403,
      "step": 20
    },
    {
      "epoch": 0.863697705802969,
      "grad_norm": 0.6521434187889099,
      "learning_rate": 1.2e-05,
      "loss": 2.2954,
      "step": 40
    },
    {
      "epoch": 0.9932523616734144,
      "eval_loss": 2.0418646335601807,
      "eval_runtime": 11.606,
      "eval_samples_per_second": 32.052,
      "eval_steps_per_second": 4.05,
      "step": 46
    },
    {
      "epoch": 1.2955465587044535,
      "grad_norm": 0.5583148002624512,
      "learning_rate": 1.8e-05,
      "loss": 2.1939,
      "step": 60
    },
    {
      "epoch": 1.7273954116059378,
      "grad_norm": 0.5125463008880615,
      "learning_rate": 2.4e-05,
      "loss": 2.011,
      "step": 80
    },
    {
      "epoch": 1.9865047233468287,
      "eval_loss": 1.755987524986267,
      "eval_runtime": 11.6593,
      "eval_samples_per_second": 31.906,
      "eval_steps_per_second": 4.031,
      "step": 92
    },
    {
      "epoch": 2.1592442645074224,
      "grad_norm": 0.6073498725891113,
      "learning_rate": 3e-05,
      "loss": 1.9537,
      "step": 100
    },
    {
      "epoch": 2.591093117408907,
      "grad_norm": 0.5852423906326294,
      "learning_rate": 2.9936387121454473e-05,
      "loss": 1.848,
      "step": 120
    },
    {
      "epoch": 2.979757085020243,
      "eval_loss": 1.5818973779678345,
      "eval_runtime": 11.6353,
      "eval_samples_per_second": 31.972,
      "eval_steps_per_second": 4.039,
      "step": 138
    },
    {
      "epoch": 3.0229419703103915,
      "grad_norm": 0.9267003536224365,
      "learning_rate": 2.9746088032260134e-05,
      "loss": 1.8373,
      "step": 140
    },
    {
      "epoch": 3.454790823211876,
      "grad_norm": 0.929217517375946,
      "learning_rate": 2.9430716795463416e-05,
      "loss": 1.7299,
      "step": 160
    },
    {
      "epoch": 3.8866396761133606,
      "grad_norm": 0.9659837484359741,
      "learning_rate": 2.8992948300688734e-05,
      "loss": 1.6946,
      "step": 180
    },
    {
      "epoch": 3.9946018893387314,
      "eval_loss": 1.4404759407043457,
      "eval_runtime": 11.6425,
      "eval_samples_per_second": 31.952,
      "eval_steps_per_second": 4.037,
      "step": 185
    },
    {
      "epoch": 4.318488529014845,
      "grad_norm": 1.2371968030929565,
      "learning_rate": 2.843649557648131e-05,
      "loss": 1.6049,
      "step": 200
    },
    {
      "epoch": 4.75033738191633,
      "grad_norm": 1.4866975545883179,
      "learning_rate": 2.776607829744932e-05,
      "loss": 1.5971,
      "step": 220
    },
    {
      "epoch": 4.987854251012146,
      "eval_loss": 1.325922966003418,
      "eval_runtime": 11.6354,
      "eval_samples_per_second": 31.971,
      "eval_steps_per_second": 4.039,
      "step": 231
    },
    {
      "epoch": 5.182186234817814,
      "grad_norm": 1.290022611618042,
      "learning_rate": 2.6987382753318885e-05,
      "loss": 1.5523,
      "step": 240
    },
    {
      "epoch": 5.614035087719298,
      "grad_norm": 1.576865315437317,
      "learning_rate": 2.6107013619433027e-05,
      "loss": 1.4756,
      "step": 260
    },
    {
      "epoch": 5.98110661268556,
      "eval_loss": 1.2389355897903442,
      "eval_runtime": 11.6564,
      "eval_samples_per_second": 31.914,
      "eval_steps_per_second": 4.032,
      "step": 277
    },
    {
      "epoch": 6.045883940620783,
      "grad_norm": 1.606406569480896,
      "learning_rate": 2.513243793776364e-05,
      "loss": 1.4332,
      "step": 280
    },
    {
      "epoch": 6.477732793522267,
      "grad_norm": 1.5990298986434937,
      "learning_rate": 2.407192178357357e-05,
      "loss": 1.3462,
      "step": 300
    },
    {
      "epoch": 6.909581646423752,
      "grad_norm": 1.7616294622421265,
      "learning_rate": 2.2934460154904436e-05,
      "loss": 1.3916,
      "step": 320
    },
    {
      "epoch": 6.995951417004049,
      "eval_loss": 1.1750843524932861,
      "eval_runtime": 11.6363,
      "eval_samples_per_second": 31.969,
      "eval_steps_per_second": 4.039,
      "step": 324
    },
    {
      "epoch": 7.341430499325236,
      "grad_norm": 2.2005562782287598,
      "learning_rate": 2.1729700679547537e-05,
      "loss": 1.3007,
      "step": 340
    },
    {
      "epoch": 7.77327935222672,
      "grad_norm": 2.0084428787231445,
      "learning_rate": 2.0467861786593858e-05,
      "loss": 1.3156,
      "step": 360
    },
    {
      "epoch": 7.989203778677463,
      "eval_loss": 1.1366231441497803,
      "eval_runtime": 11.652,
      "eval_samples_per_second": 31.926,
      "eval_steps_per_second": 4.034,
      "step": 370
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 2.2589313983917236,
      "learning_rate": 1.915964603660893e-05,
      "loss": 1.265,
      "step": 380
    },
    {
      "epoch": 8.63697705802969,
      "grad_norm": 2.530799627304077,
      "learning_rate": 1.7816149345541454e-05,
      "loss": 1.2353,
      "step": 400
    },
    {
      "epoch": 8.982456140350877,
      "eval_loss": 1.09785795211792,
      "eval_runtime": 11.6325,
      "eval_samples_per_second": 31.979,
      "eval_steps_per_second": 4.04,
      "step": 416
    },
    {
      "epoch": 9.068825910931174,
      "grad_norm": 1.9402692317962646,
      "learning_rate": 1.6448766872302767e-05,
      "loss": 1.2324,
      "step": 420
    },
    {
      "epoch": 9.50067476383266,
      "grad_norm": 2.265495538711548,
      "learning_rate": 1.5069096368252277e-05,
      "loss": 1.1746,
      "step": 440
    },
    {
      "epoch": 9.932523616734143,
      "grad_norm": 2.078183174133301,
      "learning_rate": 1.368883980835091e-05,
      "loss": 1.2004,
      "step": 460
    },
    {
      "epoch": 9.997300944669366,
      "eval_loss": 1.0847485065460205,
      "eval_runtime": 11.6602,
      "eval_samples_per_second": 31.903,
      "eval_steps_per_second": 4.031,
      "step": 463
    },
    {
      "epoch": 10.364372469635628,
      "grad_norm": 2.5572941303253174,
      "learning_rate": 1.2319704138319558e-05,
      "loss": 1.1473,
      "step": 480
    },
    {
      "epoch": 10.796221322537113,
      "grad_norm": 2.599616050720215,
      "learning_rate": 1.0973301979636888e-05,
      "loss": 1.1167,
      "step": 500
    },
    {
      "epoch": 10.99055330634278,
      "eval_loss": 1.075912356376648,
      "eval_runtime": 11.6432,
      "eval_samples_per_second": 31.95,
      "eval_steps_per_second": 4.037,
      "step": 509
    },
    {
      "epoch": 11.228070175438596,
      "grad_norm": 2.4529025554656982,
      "learning_rate": 9.66105313456874e-06,
      "loss": 1.169,
      "step": 520
    },
    {
      "epoch": 11.65991902834008,
      "grad_norm": 2.9774246215820312,
      "learning_rate": 8.394087726635483e-06,
      "loss": 1.1259,
      "step": 540
    },
    {
      "epoch": 11.983805668016194,
      "eval_loss": 1.0757372379302979,
      "eval_runtime": 11.6211,
      "eval_samples_per_second": 32.011,
      "eval_steps_per_second": 4.044,
      "step": 555
    },
    {
      "epoch": 12.091767881241566,
      "grad_norm": 2.654956340789795,
      "learning_rate": 7.183151798052628e-06,
      "loss": 1.098,
      "step": 560
    },
    {
      "epoch": 12.523616734143049,
      "grad_norm": 2.110180139541626,
      "learning_rate": 6.038516164840341e-06,
      "loss": 1.0905,
      "step": 580
    },
    {
      "epoch": 12.955465587044534,
      "grad_norm": 3.4599268436431885,
      "learning_rate": 4.969889302667217e-06,
      "loss": 1.0781,
      "step": 600
    },
    {
      "epoch": 12.998650472334683,
      "eval_loss": 1.0718672275543213,
      "eval_runtime": 11.6347,
      "eval_samples_per_second": 31.973,
      "eval_steps_per_second": 4.04,
      "step": 602
    },
    {
      "epoch": 13.387314439946019,
      "grad_norm": 2.49025821685791,
      "learning_rate": 3.986335002305783e-06,
      "loss": 1.0559,
      "step": 620
    },
    {
      "epoch": 13.819163292847504,
      "grad_norm": 2.6091041564941406,
      "learning_rate": 3.0961954931229385e-06,
      "loss": 1.1183,
      "step": 640
    },
    {
      "epoch": 13.991902834008098,
      "eval_loss": 1.0730143785476685,
      "eval_runtime": 11.6275,
      "eval_samples_per_second": 31.993,
      "eval_steps_per_second": 4.042,
      "step": 648
    },
    {
      "epoch": 14.251012145748987,
      "grad_norm": 2.651367425918579,
      "learning_rate": 2.307020686650187e-06,
      "loss": 1.0891,
      "step": 660
    },
    {
      "epoch": 14.682860998650472,
      "grad_norm": 2.513788938522339,
      "learning_rate": 1.6255041403695748e-06,
      "loss": 1.058,
      "step": 680
    },
    {
      "epoch": 14.98515519568151,
      "eval_loss": 1.074668288230896,
      "eval_runtime": 11.6329,
      "eval_samples_per_second": 31.978,
      "eval_steps_per_second": 4.04,
      "step": 694
    },
    {
      "epoch": 15.114709851551957,
      "grad_norm": 2.880951166152954,
      "learning_rate": 1.0574262848524725e-06,
      "loss": 1.0858,
      "step": 700
    },
    {
      "epoch": 15.54655870445344,
      "grad_norm": 2.6492371559143066,
      "learning_rate": 6.076053957825411e-07,
      "loss": 1.0445,
      "step": 720
    },
    {
      "epoch": 15.978407557354926,
      "grad_norm": 2.578446626663208,
      "learning_rate": 2.798567267042601e-07,
      "loss": 1.1116,
      "step": 740
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.0725425481796265,
      "eval_runtime": 11.6476,
      "eval_samples_per_second": 31.938,
      "eval_steps_per_second": 4.035,
      "step": 741
    },
    {
      "epoch": 16.41025641025641,
      "grad_norm": 2.4758195877075195,
      "learning_rate": 7.696014912157267e-08,
      "loss": 1.0335,
      "step": 760
    },
    {
      "epoch": 16.842105263157894,
      "grad_norm": 2.5701375007629395,
      "learning_rate": 6.365744140879315e-10,
      "loss": 1.0869,
      "step": 780
    },
    {
      "epoch": 16.885290148448043,
      "eval_loss": 1.0733098983764648,
      "eval_runtime": 11.6223,
      "eval_samples_per_second": 32.008,
      "eval_steps_per_second": 4.044,
      "step": 782
    }
  ],
  "logging_steps": 20,
  "max_steps": 782,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 17,
  "save_steps": 500,
  "total_flos": 6.955404634570752e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}