{
  "best_metric": 0.9241104125976562,
  "best_model_checkpoint": "outputs/checkpoint-231",
  "epoch": 11.983805668016194,
  "eval_steps": 500,
  "global_step": 555,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4318488529014845,
      "grad_norm": 1.787990689277649,
      "learning_rate": 4e-05,
      "loss": 2.7228,
      "step": 20
    },
    {
      "epoch": 0.863697705802969,
      "grad_norm": 0.5038349032402039,
      "learning_rate": 8e-05,
      "loss": 2.062,
      "step": 40
    },
    {
      "epoch": 0.9932523616734144,
      "eval_loss": 1.6799557209014893,
      "eval_runtime": 11.9387,
      "eval_samples_per_second": 31.159,
      "eval_steps_per_second": 3.937,
      "step": 46
    },
    {
      "epoch": 1.2955465587044535,
      "grad_norm": 0.6116560697555542,
      "learning_rate": 0.00012,
      "loss": 1.8813,
      "step": 60
    },
    {
      "epoch": 1.7273954116059378,
      "grad_norm": 0.8396438360214233,
      "learning_rate": 0.00016,
      "loss": 1.6421,
      "step": 80
    },
    {
      "epoch": 1.9865047233468287,
      "eval_loss": 1.2483179569244385,
      "eval_runtime": 11.921,
      "eval_samples_per_second": 31.205,
      "eval_steps_per_second": 3.943,
      "step": 92
    },
    {
      "epoch": 2.1592442645074224,
      "grad_norm": 1.1572123765945435,
      "learning_rate": 0.0002,
      "loss": 1.5211,
      "step": 100
    },
    {
      "epoch": 2.591093117408907,
      "grad_norm": 1.1013579368591309,
      "learning_rate": 0.00019943348002101371,
      "loss": 1.327,
      "step": 120
    },
    {
      "epoch": 2.979757085020243,
      "eval_loss": 1.0307273864746094,
      "eval_runtime": 11.9077,
      "eval_samples_per_second": 31.24,
      "eval_steps_per_second": 3.947,
      "step": 138
    },
    {
      "epoch": 3.0229419703103915,
      "grad_norm": 1.080468773841858,
      "learning_rate": 0.00019774033898178667,
      "loss": 1.2809,
      "step": 140
    },
    {
      "epoch": 3.454790823211876,
      "grad_norm": 1.2337204217910767,
      "learning_rate": 0.00019493976084683813,
      "loss": 1.0656,
      "step": 160
    },
    {
      "epoch": 3.8866396761133606,
      "grad_norm": 1.1269233226776123,
      "learning_rate": 0.00019106347728549135,
      "loss": 1.0563,
      "step": 180
    },
    {
      "epoch": 3.9946018893387314,
      "eval_loss": 0.9378284215927124,
      "eval_runtime": 11.9305,
      "eval_samples_per_second": 31.181,
      "eval_steps_per_second": 3.939,
      "step": 185
    },
    {
      "epoch": 4.318488529014845,
      "grad_norm": 1.3905510902404785,
      "learning_rate": 0.0001861554081393806,
      "loss": 0.8855,
      "step": 200
    },
    {
      "epoch": 4.75033738191633,
      "grad_norm": 1.3569377660751343,
      "learning_rate": 0.00018027116379309638,
      "loss": 0.9061,
      "step": 220
    },
    {
      "epoch": 4.987854251012146,
      "eval_loss": 0.9241104125976562,
      "eval_runtime": 11.9189,
      "eval_samples_per_second": 31.211,
      "eval_steps_per_second": 3.943,
      "step": 231
    },
    {
      "epoch": 5.182186234817814,
      "grad_norm": 1.3102238178253174,
      "learning_rate": 0.00017347741508630672,
      "loss": 0.8228,
      "step": 240
    },
    {
      "epoch": 5.614035087719298,
      "grad_norm": 1.1156977415084839,
      "learning_rate": 0.00016585113790650388,
      "loss": 0.711,
      "step": 260
    },
    {
      "epoch": 5.98110661268556,
      "eval_loss": 0.9796295762062073,
      "eval_runtime": 11.9256,
      "eval_samples_per_second": 31.194,
      "eval_steps_per_second": 3.941,
      "step": 277
    },
    {
      "epoch": 6.045883940620783,
      "grad_norm": 0.9861016869544983,
      "learning_rate": 0.0001574787410214407,
      "loss": 0.6591,
      "step": 280
    },
    {
      "epoch": 6.477732793522267,
      "grad_norm": 1.1385196447372437,
      "learning_rate": 0.00014845508703326504,
      "loss": 0.5255,
      "step": 300
    },
    {
      "epoch": 6.909581646423752,
      "grad_norm": 1.0708714723587036,
      "learning_rate": 0.00013888241754733208,
      "loss": 0.5846,
      "step": 320
    },
    {
      "epoch": 6.995951417004049,
      "eval_loss": 1.0800774097442627,
      "eval_runtime": 11.9183,
      "eval_samples_per_second": 31.212,
      "eval_steps_per_second": 3.944,
      "step": 324
    },
    {
      "epoch": 7.341430499325236,
      "grad_norm": 1.0299906730651855,
      "learning_rate": 0.0001288691947339621,
      "loss": 0.4571,
      "step": 340
    },
    {
      "epoch": 7.77327935222672,
      "grad_norm": 1.1635417938232422,
      "learning_rate": 0.00011852887240871145,
      "loss": 0.465,
      "step": 360
    },
    {
      "epoch": 7.989203778677463,
      "eval_loss": 1.1291037797927856,
      "eval_runtime": 11.9151,
      "eval_samples_per_second": 31.221,
      "eval_steps_per_second": 3.945,
      "step": 370
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 0.9930064082145691,
      "learning_rate": 0.00010797861055530831,
      "loss": 0.3898,
      "step": 380
    },
    {
      "epoch": 8.63697705802969,
      "grad_norm": 1.086180329322815,
      "learning_rate": 9.733794785622253e-05,
      "loss": 0.341,
      "step": 400
    },
    {
      "epoch": 8.982456140350877,
      "eval_loss": 1.2221601009368896,
      "eval_runtime": 11.9463,
      "eval_samples_per_second": 31.139,
      "eval_steps_per_second": 3.934,
      "step": 416
    },
    {
      "epoch": 9.068825910931174,
      "grad_norm": 0.881294310092926,
      "learning_rate": 8.672744727162781e-05,
      "loss": 0.3388,
      "step": 420
    },
    {
      "epoch": 9.50067476383266,
      "grad_norm": 0.9569498896598816,
      "learning_rate": 7.626733001288851e-05,
      "loss": 0.2579,
      "step": 440
    },
    {
      "epoch": 9.932523616734143,
      "grad_norm": 0.9368858337402344,
      "learning_rate": 6.607611338819697e-05,
      "loss": 0.2657,
      "step": 460
    },
    {
      "epoch": 9.997300944669366,
      "eval_loss": 1.310530424118042,
      "eval_runtime": 11.9264,
      "eval_samples_per_second": 31.191,
      "eval_steps_per_second": 3.941,
      "step": 463
    },
    {
      "epoch": 10.364372469635628,
      "grad_norm": 0.8425939679145813,
      "learning_rate": 5.626926795411447e-05,
      "loss": 0.1949,
      "step": 480
    },
    {
      "epoch": 10.796221322537113,
      "grad_norm": 0.9611827731132507,
      "learning_rate": 4.695790918802576e-05,
      "loss": 0.1918,
      "step": 500
    },
    {
      "epoch": 10.99055330634278,
      "eval_loss": 1.393843412399292,
      "eval_runtime": 11.9319,
      "eval_samples_per_second": 31.177,
      "eval_steps_per_second": 3.939,
      "step": 509
    },
    {
      "epoch": 11.228070175438596,
      "grad_norm": 0.75007563829422,
      "learning_rate": 3.824753850538082e-05,
      "loss": 0.1854,
      "step": 520
    },
    {
      "epoch": 11.65991902834008,
      "grad_norm": 0.710945725440979,
      "learning_rate": 3.0236847886501542e-05,
      "loss": 0.162,
      "step": 540
    },
    {
      "epoch": 11.983805668016194,
      "eval_loss": 1.4677594900131226,
      "eval_runtime": 11.9217,
      "eval_samples_per_second": 31.204,
      "eval_steps_per_second": 3.942,
      "step": 555
    }
  ],
  "logging_steps": 20,
  "max_steps": 690,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "total_flos": 4.962667014222643e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}