{
  "best_metric": 1.0420600175857544,
  "best_model_checkpoint": "outputs/checkpoint-555",
  "epoch": 16.0,
  "eval_steps": 500,
  "global_step": 741,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4318488529014845,
      "grad_norm": 0.9230825901031494,
      "learning_rate": 6e-06,
      "loss": 2.3403,
      "step": 20
    },
    {
      "epoch": 0.863697705802969,
      "grad_norm": 0.656217098236084,
      "learning_rate": 1.2e-05,
      "loss": 2.2959,
      "step": 40
    },
    {
      "epoch": 0.9932523616734144,
      "eval_loss": 2.0449142456054688,
      "eval_runtime": 11.6816,
      "eval_samples_per_second": 31.845,
      "eval_steps_per_second": 4.023,
      "step": 46
    },
    {
      "epoch": 1.2955465587044535,
      "grad_norm": 0.5638908743858337,
      "learning_rate": 1.8e-05,
      "loss": 2.1979,
      "step": 60
    },
    {
      "epoch": 1.7273954116059378,
      "grad_norm": 0.51393723487854,
      "learning_rate": 2.4e-05,
      "loss": 2.0118,
      "step": 80
    },
    {
      "epoch": 1.9865047233468287,
      "eval_loss": 1.7561986446380615,
      "eval_runtime": 11.7086,
      "eval_samples_per_second": 31.772,
      "eval_steps_per_second": 4.014,
      "step": 92
    },
    {
      "epoch": 2.1592442645074224,
      "grad_norm": 0.6104076504707336,
      "learning_rate": 3e-05,
      "loss": 1.9541,
      "step": 100
    },
    {
      "epoch": 2.591093117408907,
      "grad_norm": 0.5912227034568787,
      "learning_rate": 2.9981931843077588e-05,
      "loss": 1.849,
      "step": 120
    },
    {
      "epoch": 2.979757085020243,
      "eval_loss": 1.5828752517700195,
      "eval_runtime": 11.7012,
      "eval_samples_per_second": 31.792,
      "eval_steps_per_second": 4.017,
      "step": 138
    },
    {
      "epoch": 3.0229419703103915,
      "grad_norm": 0.9186816215515137,
      "learning_rate": 2.9927770900082956e-05,
      "loss": 1.8383,
      "step": 140
    },
    {
      "epoch": 3.454790823211876,
      "grad_norm": 0.9306802153587341,
      "learning_rate": 2.9837647649471717e-05,
      "loss": 1.7307,
      "step": 160
    },
    {
      "epoch": 3.8866396761133606,
      "grad_norm": 0.9665244817733765,
      "learning_rate": 2.9711779206048457e-05,
      "loss": 1.695,
      "step": 180
    },
    {
      "epoch": 3.9946018893387314,
      "eval_loss": 1.4405076503753662,
      "eval_runtime": 11.7039,
      "eval_samples_per_second": 31.784,
      "eval_steps_per_second": 4.016,
      "step": 185
    },
    {
      "epoch": 4.318488529014845,
      "grad_norm": 1.23567795753479,
      "learning_rate": 2.9550468797918162e-05,
      "loss": 1.6044,
      "step": 200
    },
    {
      "epoch": 4.75033738191633,
      "grad_norm": 1.4878095388412476,
      "learning_rate": 2.9354105035983133e-05,
      "loss": 1.5957,
      "step": 220
    },
    {
      "epoch": 4.987854251012146,
      "eval_loss": 1.32224702835083,
      "eval_runtime": 11.7081,
      "eval_samples_per_second": 31.773,
      "eval_steps_per_second": 4.014,
      "step": 231
    },
    {
      "epoch": 5.182186234817814,
      "grad_norm": 1.3139092922210693,
      "learning_rate": 2.912316097774531e-05,
      "loss": 1.5493,
      "step": 240
    },
    {
      "epoch": 5.614035087719298,
      "grad_norm": 1.5993897914886475,
      "learning_rate": 2.8858192987669303e-05,
      "loss": 1.4703,
      "step": 260
    },
    {
      "epoch": 5.98110661268556,
      "eval_loss": 1.231151819229126,
      "eval_runtime": 11.7193,
      "eval_samples_per_second": 31.743,
      "eval_steps_per_second": 4.01,
      "step": 277
    },
    {
      "epoch": 6.045883940620783,
      "grad_norm": 1.660462737083435,
      "learning_rate": 2.8559839396851653e-05,
      "loss": 1.4259,
      "step": 280
    },
    {
      "epoch": 6.477732793522267,
      "grad_norm": 1.586065411567688,
      "learning_rate": 2.8228818965225325e-05,
      "loss": 1.334,
      "step": 300
    },
    {
      "epoch": 6.909581646423752,
      "grad_norm": 1.869524359703064,
      "learning_rate": 2.7865929150004083e-05,
      "loss": 1.3783,
      "step": 320
    },
    {
      "epoch": 6.995951417004049,
      "eval_loss": 1.1627566814422607,
      "eval_runtime": 11.7057,
      "eval_samples_per_second": 31.779,
      "eval_steps_per_second": 4.015,
      "step": 324
    },
    {
      "epoch": 7.341430499325236,
      "grad_norm": 2.316075325012207,
      "learning_rate": 2.747204418453818e-05,
      "loss": 1.2803,
      "step": 340
    },
    {
      "epoch": 7.77327935222672,
      "grad_norm": 2.061739921569824,
      "learning_rate": 2.7048112972209674e-05,
      "loss": 1.2944,
      "step": 360
    },
    {
      "epoch": 7.989203778677463,
      "eval_loss": 1.1163532733917236,
      "eval_runtime": 11.7029,
      "eval_samples_per_second": 31.787,
      "eval_steps_per_second": 4.016,
      "step": 370
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 2.3766236305236816,
      "learning_rate": 2.6595156800441055e-05,
      "loss": 1.2365,
      "step": 380
    },
    {
      "epoch": 8.63697705802969,
      "grad_norm": 2.696465015411377,
      "learning_rate": 2.611426688032439e-05,
      "loss": 1.2014,
      "step": 400
    },
    {
      "epoch": 8.982456140350877,
      "eval_loss": 1.0828282833099365,
      "eval_runtime": 11.7054,
      "eval_samples_per_second": 31.78,
      "eval_steps_per_second": 4.015,
      "step": 416
    },
    {
      "epoch": 9.068825910931174,
      "grad_norm": 2.374746322631836,
      "learning_rate": 2.5606601717798212e-05,
      "loss": 1.1982,
      "step": 420
    },
    {
      "epoch": 9.50067476383266,
      "grad_norm": 2.3256847858428955,
      "learning_rate": 2.5073384322705278e-05,
      "loss": 1.1262,
      "step": 440
    },
    {
      "epoch": 9.932523616734143,
      "grad_norm": 2.200650930404663,
      "learning_rate": 2.4515899262454684e-05,
      "loss": 1.1541,
      "step": 460
    },
    {
      "epoch": 9.997300944669366,
      "eval_loss": 1.0607314109802246,
      "eval_runtime": 11.7013,
      "eval_samples_per_second": 31.791,
      "eval_steps_per_second": 4.017,
      "step": 463
    },
    {
      "epoch": 10.364372469635628,
      "grad_norm": 2.9810495376586914,
      "learning_rate": 2.3935489567386502e-05,
      "loss": 1.0843,
      "step": 480
    },
    {
      "epoch": 10.796221322537113,
      "grad_norm": 2.9131200313568115,
      "learning_rate": 2.3333553495294033e-05,
      "loss": 1.0568,
      "step": 500
    },
    {
      "epoch": 10.99055330634278,
      "eval_loss": 1.0553909540176392,
      "eval_runtime": 11.6966,
      "eval_samples_per_second": 31.804,
      "eval_steps_per_second": 4.018,
      "step": 509
    },
    {
      "epoch": 11.228070175438596,
      "grad_norm": 2.6710944175720215,
      "learning_rate": 2.2711541162898324e-05,
      "loss": 1.0991,
      "step": 520
    },
    {
      "epoch": 11.65991902834008,
      "grad_norm": 3.4023935794830322,
      "learning_rate": 2.207095105238997e-05,
      "loss": 1.0467,
      "step": 540
    },
    {
      "epoch": 11.983805668016194,
      "eval_loss": 1.0420600175857544,
      "eval_runtime": 11.7007,
      "eval_samples_per_second": 31.793,
      "eval_steps_per_second": 4.017,
      "step": 555
    },
    {
      "epoch": 12.091767881241566,
      "grad_norm": 2.8432018756866455,
      "learning_rate": 2.141332640145423e-05,
      "loss": 1.0121,
      "step": 560
    },
    {
      "epoch": 12.523616734143049,
      "grad_norm": 2.7599430084228516,
      "learning_rate": 2.074025148547635e-05,
      "loss": 0.9882,
      "step": 580
    },
    {
      "epoch": 12.955465587044534,
      "grad_norm": 3.7460949420928955,
      "learning_rate": 2.00533478008833e-05,
      "loss": 0.9781,
      "step": 600
    },
    {
      "epoch": 12.998650472334683,
      "eval_loss": 1.0568532943725586,
      "eval_runtime": 11.6959,
      "eval_samples_per_second": 31.806,
      "eval_steps_per_second": 4.018,
      "step": 602
    },
    {
      "epoch": 13.387314439946019,
      "grad_norm": 2.7911863327026367,
      "learning_rate": 1.9354270158816936e-05,
      "loss": 0.9352,
      "step": 620
    },
    {
      "epoch": 13.819163292847504,
      "grad_norm": 3.0418496131896973,
      "learning_rate": 1.8644702698548963e-05,
      "loss": 0.9929,
      "step": 640
    },
    {
      "epoch": 13.991902834008098,
      "eval_loss": 1.0488306283950806,
      "eval_runtime": 11.7173,
      "eval_samples_per_second": 31.748,
      "eval_steps_per_second": 4.011,
      "step": 648
    },
    {
      "epoch": 14.251012145748987,
      "grad_norm": 3.146819829940796,
      "learning_rate": 1.7926354830241928e-05,
      "loss": 0.9557,
      "step": 660
    },
    {
      "epoch": 14.682860998650472,
      "grad_norm": 2.9473140239715576,
      "learning_rate": 1.7200957116830427e-05,
      "loss": 0.9102,
      "step": 680
    },
    {
      "epoch": 14.98515519568151,
      "eval_loss": 1.065933108329773,
      "eval_runtime": 11.728,
      "eval_samples_per_second": 31.719,
      "eval_steps_per_second": 4.007,
      "step": 694
    },
    {
      "epoch": 15.114709851551957,
      "grad_norm": 3.4028427600860596,
      "learning_rate": 1.6470257104943414e-05,
      "loss": 0.9319,
      "step": 700
    },
    {
      "epoch": 15.54655870445344,
      "grad_norm": 3.339811086654663,
      "learning_rate": 1.5736015114911272e-05,
      "loss": 0.8673,
      "step": 720
    },
    {
      "epoch": 15.978407557354926,
      "grad_norm": 3.3190574645996094,
      "learning_rate": 1.5e-05,
      "loss": 0.9472,
      "step": 740
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.0857369899749756,
      "eval_runtime": 11.7599,
      "eval_samples_per_second": 31.633,
      "eval_steps_per_second": 3.997,
      "step": 741
    }
  ],
  "logging_steps": 20,
  "max_steps": 1380,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 6.589532696154931e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}