{
  "best_metric": 0.9539479613304138,
  "best_model_checkpoint": "outputs/checkpoint-231",
  "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": 0.7265238165855408,
      "learning_rate": 4e-05,
      "loss": 2.2901,
      "step": 20
    },
    {
      "epoch": 0.863697705802969,
      "grad_norm": 0.4902186989784241,
      "learning_rate": 8e-05,
      "loss": 2.0553,
      "step": 40
    },
    {
      "epoch": 0.9932523616734144,
      "eval_loss": 1.7082113027572632,
      "eval_runtime": 11.6752,
      "eval_samples_per_second": 31.862,
      "eval_steps_per_second": 4.026,
      "step": 46
    },
    {
      "epoch": 1.2955465587044535,
      "grad_norm": 0.5778974294662476,
      "learning_rate": 0.00012,
      "loss": 1.9034,
      "step": 60
    },
    {
      "epoch": 1.7273954116059378,
      "grad_norm": 0.818723201751709,
      "learning_rate": 0.00016,
      "loss": 1.6895,
      "step": 80
    },
    {
      "epoch": 1.9865047233468287,
      "eval_loss": 1.3229522705078125,
      "eval_runtime": 11.7014,
      "eval_samples_per_second": 31.791,
      "eval_steps_per_second": 4.017,
      "step": 92
    },
    {
      "epoch": 2.1592442645074224,
      "grad_norm": 1.2037190198898315,
      "learning_rate": 0.0002,
      "loss": 1.5763,
      "step": 100
    },
    {
      "epoch": 2.591093117408907,
      "grad_norm": 1.2296253442764282,
      "learning_rate": 0.00019978499773373596,
      "loss": 1.3997,
      "step": 120
    },
    {
      "epoch": 2.979757085020243,
      "eval_loss": 1.0936287641525269,
      "eval_runtime": 11.7036,
      "eval_samples_per_second": 31.785,
      "eval_steps_per_second": 4.016,
      "step": 138
    },
    {
      "epoch": 3.0229419703103915,
      "grad_norm": 1.3699569702148438,
      "learning_rate": 0.0001991409154544338,
      "loss": 1.3491,
      "step": 140
    },
    {
      "epoch": 3.454790823211876,
      "grad_norm": 1.451266884803772,
      "learning_rate": 0.00019807052274508773,
      "loss": 1.1626,
      "step": 160
    },
    {
      "epoch": 3.8866396761133606,
      "grad_norm": 1.3704801797866821,
      "learning_rate": 0.0001965784223428638,
      "loss": 1.1558,
      "step": 180
    },
    {
      "epoch": 3.9946018893387314,
      "eval_loss": 0.9916501641273499,
      "eval_runtime": 11.7248,
      "eval_samples_per_second": 31.728,
      "eval_steps_per_second": 4.009,
      "step": 185
    },
    {
      "epoch": 4.318488529014845,
      "grad_norm": 1.6158828735351562,
      "learning_rate": 0.0001946710303471214,
      "loss": 1.0048,
      "step": 200
    },
    {
      "epoch": 4.75033738191633,
      "grad_norm": 1.7279417514801025,
      "learning_rate": 0.00019235654862989537,
      "loss": 1.0293,
      "step": 220
    },
    {
      "epoch": 4.987854251012146,
      "eval_loss": 0.9539479613304138,
      "eval_runtime": 11.7041,
      "eval_samples_per_second": 31.784,
      "eval_steps_per_second": 4.016,
      "step": 231
    },
    {
      "epoch": 5.182186234817814,
      "grad_norm": 1.5809475183486938,
      "learning_rate": 0.00018964492956747425,
      "loss": 0.966,
      "step": 240
    },
    {
      "epoch": 5.614035087719298,
      "grad_norm": 1.6080421209335327,
      "learning_rate": 0.00018654783324473137,
      "loss": 0.8655,
      "step": 260
    },
    {
      "epoch": 5.98110661268556,
      "eval_loss": 0.979802131652832,
      "eval_runtime": 11.7244,
      "eval_samples_per_second": 31.729,
      "eval_steps_per_second": 4.009,
      "step": 277
    },
    {
      "epoch": 6.045883940620783,
      "grad_norm": 1.3081737756729126,
      "learning_rate": 0.00018307857731623132,
      "loss": 0.8181,
      "step": 280
    },
    {
      "epoch": 6.477732793522267,
      "grad_norm": 1.4086545705795288,
      "learning_rate": 0.0001792520797397116,
      "loss": 0.689,
      "step": 300
    },
    {
      "epoch": 6.909581646423752,
      "grad_norm": 1.4386905431747437,
      "learning_rate": 0.00017508479462818833,
      "loss": 0.768,
      "step": 320
    },
    {
      "epoch": 6.995951417004049,
      "eval_loss": 1.0414738655090332,
      "eval_runtime": 11.7025,
      "eval_samples_per_second": 31.788,
      "eval_steps_per_second": 4.016,
      "step": 324
    },
    {
      "epoch": 7.341430499325236,
      "grad_norm": 1.4608936309814453,
      "learning_rate": 0.00017059464149652448,
      "loss": 0.6351,
      "step": 340
    },
    {
      "epoch": 7.77327935222672,
      "grad_norm": 1.4568973779678345,
      "learning_rate": 0.0001658009282067036,
      "loss": 0.6611,
      "step": 360
    },
    {
      "epoch": 7.989203778677463,
      "eval_loss": 1.0716439485549927,
      "eval_runtime": 11.702,
      "eval_samples_per_second": 31.789,
      "eval_steps_per_second": 4.016,
      "step": 370
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 1.7085548639297485,
      "learning_rate": 0.00016072426794314836,
      "loss": 0.5859,
      "step": 380
    },
    {
      "epoch": 8.63697705802969,
      "grad_norm": 1.7092846632003784,
      "learning_rate": 0.00015538649057509306,
      "loss": 0.5377,
      "step": 400
    },
    {
      "epoch": 8.982456140350877,
      "eval_loss": 1.1437946557998657,
      "eval_runtime": 11.7191,
      "eval_samples_per_second": 31.743,
      "eval_steps_per_second": 4.011,
      "step": 416
    },
    {
      "epoch": 9.068825910931174,
      "grad_norm": 1.2362638711929321,
      "learning_rate": 0.0001498105487871566,
      "loss": 0.5412,
      "step": 420
    },
    {
      "epoch": 9.50067476383266,
      "grad_norm": 1.3204894065856934,
      "learning_rate": 0.00014402041938175826,
      "loss": 0.454,
      "step": 440
    },
    {
      "epoch": 9.932523616734143,
      "grad_norm": 1.385745882987976,
      "learning_rate": 0.00013804100017778047,
      "loss": 0.4787,
      "step": 460
    },
    {
      "epoch": 9.997300944669366,
      "eval_loss": 1.2027604579925537,
      "eval_runtime": 11.7626,
      "eval_samples_per_second": 31.626,
      "eval_steps_per_second": 3.996,
      "step": 463
    },
    {
      "epoch": 10.364372469635628,
      "grad_norm": 1.4514780044555664,
      "learning_rate": 0.00013189800294881925,
      "loss": 0.3897,
      "step": 480
    },
    {
      "epoch": 10.796221322537113,
      "grad_norm": 1.575695276260376,
      "learning_rate": 0.00012561784286139213,
      "loss": 0.3861,
      "step": 500
    },
    {
      "epoch": 10.99055330634278,
      "eval_loss": 1.2678759098052979,
      "eval_runtime": 11.7185,
      "eval_samples_per_second": 31.745,
      "eval_steps_per_second": 4.011,
      "step": 509
    }
  ],
  "logging_steps": 20,
  "max_steps": 1058,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 23,
  "save_steps": 500,
  "total_flos": 4.524072585417523e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}