{
  "best_metric": 0.887685239315033,
  "best_model_checkpoint": "outputs/checkpoint-185",
  "epoch": 10.99055330634278,
  "eval_steps": 500,
  "global_step": 509,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4318488529014845,
      "grad_norm": 0.5924658179283142,
      "learning_rate": 4e-05,
      "loss": 1.9539,
      "step": 20
    },
    {
      "epoch": 0.863697705802969,
      "grad_norm": 0.495579332113266,
      "learning_rate": 8e-05,
      "loss": 1.8186,
      "step": 40
    },
    {
      "epoch": 0.9932523616734144,
      "eval_loss": 1.5013270378112793,
      "eval_runtime": 13.4628,
      "eval_samples_per_second": 27.632,
      "eval_steps_per_second": 3.491,
      "step": 46
    },
    {
      "epoch": 1.2955465587044535,
      "grad_norm": 0.6942113041877747,
      "learning_rate": 0.00012,
      "loss": 1.6882,
      "step": 60
    },
    {
      "epoch": 1.7273954116059378,
      "grad_norm": 0.8643310070037842,
      "learning_rate": 0.00016,
      "loss": 1.4623,
      "step": 80
    },
    {
      "epoch": 1.9865047233468287,
      "eval_loss": 1.1028813123703003,
      "eval_runtime": 13.4743,
      "eval_samples_per_second": 27.608,
      "eval_steps_per_second": 3.488,
      "step": 92
    },
    {
      "epoch": 2.1592442645074224,
      "grad_norm": 1.187238097190857,
      "learning_rate": 0.0002,
      "loss": 1.3448,
      "step": 100
    },
    {
      "epoch": 2.591093117408907,
      "grad_norm": 1.2552473545074463,
      "learning_rate": 0.00019943348002101371,
      "loss": 1.1681,
      "step": 120
    },
    {
      "epoch": 2.979757085020243,
      "eval_loss": 0.9362145066261292,
      "eval_runtime": 13.4657,
      "eval_samples_per_second": 27.626,
      "eval_steps_per_second": 3.49,
      "step": 138
    },
    {
      "epoch": 3.0229419703103915,
      "grad_norm": 1.0589548349380493,
      "learning_rate": 0.00019774033898178667,
      "loss": 1.1426,
      "step": 140
    },
    {
      "epoch": 3.454790823211876,
      "grad_norm": 1.1629831790924072,
      "learning_rate": 0.00019493976084683813,
      "loss": 0.9296,
      "step": 160
    },
    {
      "epoch": 3.8866396761133606,
      "grad_norm": 0.9937772154808044,
      "learning_rate": 0.00019106347728549135,
      "loss": 0.9304,
      "step": 180
    },
    {
      "epoch": 3.9946018893387314,
      "eval_loss": 0.887685239315033,
      "eval_runtime": 13.4691,
      "eval_samples_per_second": 27.619,
      "eval_steps_per_second": 3.489,
      "step": 185
    },
    {
      "epoch": 4.318488529014845,
      "grad_norm": 1.299687385559082,
      "learning_rate": 0.0001861554081393806,
      "loss": 0.7665,
      "step": 200
    },
    {
      "epoch": 4.75033738191633,
      "grad_norm": 1.375931978225708,
      "learning_rate": 0.00018027116379309638,
      "loss": 0.7884,
      "step": 220
    },
    {
      "epoch": 4.987854251012146,
      "eval_loss": 0.8908663392066956,
      "eval_runtime": 13.466,
      "eval_samples_per_second": 27.625,
      "eval_steps_per_second": 3.49,
      "step": 231
    },
    {
      "epoch": 5.182186234817814,
      "grad_norm": 1.134634256362915,
      "learning_rate": 0.00017347741508630672,
      "loss": 0.7193,
      "step": 240
    },
    {
      "epoch": 5.614035087719298,
      "grad_norm": 1.0707848072052002,
      "learning_rate": 0.00016585113790650388,
      "loss": 0.6113,
      "step": 260
    },
    {
      "epoch": 5.98110661268556,
      "eval_loss": 0.9561511874198914,
      "eval_runtime": 13.4701,
      "eval_samples_per_second": 27.617,
      "eval_steps_per_second": 3.489,
      "step": 277
    },
    {
      "epoch": 6.045883940620783,
      "grad_norm": 0.8976731300354004,
      "learning_rate": 0.0001574787410214407,
      "loss": 0.5777,
      "step": 280
    },
    {
      "epoch": 6.477732793522267,
      "grad_norm": 1.1106950044631958,
      "learning_rate": 0.00014845508703326504,
      "loss": 0.4493,
      "step": 300
    },
    {
      "epoch": 6.909581646423752,
      "grad_norm": 1.1484525203704834,
      "learning_rate": 0.00013888241754733208,
      "loss": 0.503,
      "step": 320
    },
    {
      "epoch": 6.995951417004049,
      "eval_loss": 1.0866990089416504,
      "eval_runtime": 13.4618,
      "eval_samples_per_second": 27.634,
      "eval_steps_per_second": 3.491,
      "step": 324
    },
    {
      "epoch": 7.341430499325236,
      "grad_norm": 1.0823029279708862,
      "learning_rate": 0.0001288691947339621,
      "loss": 0.3874,
      "step": 340
    },
    {
      "epoch": 7.77327935222672,
      "grad_norm": 1.199161410331726,
      "learning_rate": 0.00011852887240871145,
      "loss": 0.4033,
      "step": 360
    },
    {
      "epoch": 7.989203778677463,
      "eval_loss": 1.1609514951705933,
      "eval_runtime": 13.4575,
      "eval_samples_per_second": 27.643,
      "eval_steps_per_second": 3.492,
      "step": 370
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 1.689239263534546,
      "learning_rate": 0.00010797861055530831,
      "loss": 0.3307,
      "step": 380
    },
    {
      "epoch": 8.63697705802969,
      "grad_norm": 1.1823320388793945,
      "learning_rate": 9.733794785622253e-05,
      "loss": 0.2835,
      "step": 400
    },
    {
      "epoch": 8.982456140350877,
      "eval_loss": 1.2469967603683472,
      "eval_runtime": 13.507,
      "eval_samples_per_second": 27.541,
      "eval_steps_per_second": 3.48,
      "step": 416
    },
    {
      "epoch": 9.068825910931174,
      "grad_norm": 0.8874559998512268,
      "learning_rate": 8.672744727162781e-05,
      "loss": 0.2876,
      "step": 420
    },
    {
      "epoch": 9.50067476383266,
      "grad_norm": 1.01341712474823,
      "learning_rate": 7.626733001288851e-05,
      "loss": 0.2161,
      "step": 440
    },
    {
      "epoch": 9.932523616734143,
      "grad_norm": 0.9801278710365295,
      "learning_rate": 6.607611338819697e-05,
      "loss": 0.2191,
      "step": 460
    },
    {
      "epoch": 9.997300944669366,
      "eval_loss": 1.3673460483551025,
      "eval_runtime": 13.4884,
      "eval_samples_per_second": 27.579,
      "eval_steps_per_second": 3.484,
      "step": 463
    },
    {
      "epoch": 10.364372469635628,
      "grad_norm": 0.9759404063224792,
      "learning_rate": 5.626926795411447e-05,
      "loss": 0.1568,
      "step": 480
    },
    {
      "epoch": 10.796221322537113,
      "grad_norm": 1.0447685718536377,
      "learning_rate": 4.695790918802576e-05,
      "loss": 0.1527,
      "step": 500
    },
    {
      "epoch": 10.99055330634278,
      "eval_loss": 1.4647294282913208,
      "eval_runtime": 13.4915,
      "eval_samples_per_second": 27.573,
      "eval_steps_per_second": 3.484,
      "step": 509
    }
  ],
  "logging_steps": 20,
  "max_steps": 690,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "total_flos": 3.921061227431117e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}