{
  "best_metric": 1.0909315347671509,
  "best_model_checkpoint": "outputs/checkpoint-602",
  "epoch": 14.898785425101215,
  "eval_steps": 500,
  "global_step": 690,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.4318488529014845,
      "grad_norm": 0.9229329228401184,
      "learning_rate": 6e-06,
      "loss": 2.3402,
      "step": 20
    },
    {
      "epoch": 0.863697705802969,
      "grad_norm": 0.6567079424858093,
      "learning_rate": 1.2e-05,
      "loss": 2.2949,
      "step": 40
    },
    {
      "epoch": 0.9932523616734144,
      "eval_loss": 2.04097318649292,
      "eval_runtime": 12.0995,
      "eval_samples_per_second": 30.745,
      "eval_steps_per_second": 3.884,
      "step": 46
    },
    {
      "epoch": 1.2955465587044535,
      "grad_norm": 0.5602224469184875,
      "learning_rate": 1.8e-05,
      "loss": 2.1934,
      "step": 60
    },
    {
      "epoch": 1.7273954116059378,
      "grad_norm": 0.5213916301727295,
      "learning_rate": 2.4e-05,
      "loss": 2.0095,
      "step": 80
    },
    {
      "epoch": 1.9865047233468287,
      "eval_loss": 1.7542632818222046,
      "eval_runtime": 12.0873,
      "eval_samples_per_second": 30.776,
      "eval_steps_per_second": 3.888,
      "step": 92
    },
    {
      "epoch": 2.1592442645074224,
      "grad_norm": 0.6186114549636841,
      "learning_rate": 3e-05,
      "loss": 1.9523,
      "step": 100
    },
    {
      "epoch": 2.591093117408907,
      "grad_norm": 0.5964342951774597,
      "learning_rate": 2.9915022003152058e-05,
      "loss": 1.8469,
      "step": 120
    },
    {
      "epoch": 2.979757085020243,
      "eval_loss": 1.581282138824463,
      "eval_runtime": 12.0784,
      "eval_samples_per_second": 30.799,
      "eval_steps_per_second": 3.891,
      "step": 138
    },
    {
      "epoch": 3.0229419703103915,
      "grad_norm": 0.9454211592674255,
      "learning_rate": 2.9661050847268002e-05,
      "loss": 1.8366,
      "step": 140
    },
    {
      "epoch": 3.454790823211876,
      "grad_norm": 0.9509153962135315,
      "learning_rate": 2.924096412702572e-05,
      "loss": 1.7299,
      "step": 160
    },
    {
      "epoch": 3.8866396761133606,
      "grad_norm": 0.9890360832214355,
      "learning_rate": 2.8659521592823702e-05,
      "loss": 1.6954,
      "step": 180
    },
    {
      "epoch": 3.9946018893387314,
      "eval_loss": 1.4414902925491333,
      "eval_runtime": 12.0827,
      "eval_samples_per_second": 30.788,
      "eval_steps_per_second": 3.89,
      "step": 185
    },
    {
      "epoch": 4.318488529014845,
      "grad_norm": 1.2653647661209106,
      "learning_rate": 2.792331122090709e-05,
      "loss": 1.6064,
      "step": 200
    },
    {
      "epoch": 4.75033738191633,
      "grad_norm": 1.5335606336593628,
      "learning_rate": 2.7040674568964454e-05,
      "loss": 1.5995,
      "step": 220
    },
    {
      "epoch": 4.987854251012146,
      "eval_loss": 1.3295938968658447,
      "eval_runtime": 12.0736,
      "eval_samples_per_second": 30.811,
      "eval_steps_per_second": 3.893,
      "step": 231
    },
    {
      "epoch": 5.182186234817814,
      "grad_norm": 1.3195627927780151,
      "learning_rate": 2.6021612262946008e-05,
      "loss": 1.5563,
      "step": 240
    },
    {
      "epoch": 5.614035087719298,
      "grad_norm": 1.599996566772461,
      "learning_rate": 2.487767068597558e-05,
      "loss": 1.4811,
      "step": 260
    },
    {
      "epoch": 5.98110661268556,
      "eval_loss": 1.2465661764144897,
      "eval_runtime": 12.0805,
      "eval_samples_per_second": 30.793,
      "eval_steps_per_second": 3.891,
      "step": 277
    },
    {
      "epoch": 6.045883940620783,
      "grad_norm": 1.6401485204696655,
      "learning_rate": 2.3621811153216105e-05,
      "loss": 1.4402,
      "step": 280
    },
    {
      "epoch": 6.477732793522267,
      "grad_norm": 1.6716389656066895,
      "learning_rate": 2.2268263054989755e-05,
      "loss": 1.3571,
      "step": 300
    },
    {
      "epoch": 6.909581646423752,
      "grad_norm": 1.7910152673721313,
      "learning_rate": 2.0832362632099814e-05,
      "loss": 1.4024,
      "step": 320
    },
    {
      "epoch": 6.995951417004049,
      "eval_loss": 1.1861686706542969,
      "eval_runtime": 12.072,
      "eval_samples_per_second": 30.815,
      "eval_steps_per_second": 3.893,
      "step": 324
    },
    {
      "epoch": 7.341430499325236,
      "grad_norm": 2.179117441177368,
      "learning_rate": 1.9330379210094315e-05,
      "loss": 1.3156,
      "step": 340
    },
    {
      "epoch": 7.77327935222672,
      "grad_norm": 1.932990550994873,
      "learning_rate": 1.7779330861306716e-05,
      "loss": 1.3311,
      "step": 360
    },
    {
      "epoch": 7.989203778677463,
      "eval_loss": 1.1523733139038086,
      "eval_runtime": 12.0776,
      "eval_samples_per_second": 30.801,
      "eval_steps_per_second": 3.892,
      "step": 370
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 2.1301991939544678,
      "learning_rate": 1.6196791583296248e-05,
      "loss": 1.2848,
      "step": 380
    },
    {
      "epoch": 8.63697705802969,
      "grad_norm": 2.5514206886291504,
      "learning_rate": 1.460069217843338e-05,
      "loss": 1.2557,
      "step": 400
    },
    {
      "epoch": 8.982456140350877,
      "eval_loss": 1.1161961555480957,
      "eval_runtime": 12.0718,
      "eval_samples_per_second": 30.816,
      "eval_steps_per_second": 3.893,
      "step": 416
    },
    {
      "epoch": 9.068825910931174,
      "grad_norm": 1.910114049911499,
      "learning_rate": 1.3009117090744171e-05,
      "loss": 1.254,
      "step": 420
    },
    {
      "epoch": 9.50067476383266,
      "grad_norm": 2.328728437423706,
      "learning_rate": 1.1440099501933278e-05,
      "loss": 1.2017,
      "step": 440
    },
    {
      "epoch": 9.932523616734143,
      "grad_norm": 2.1410059928894043,
      "learning_rate": 9.911417008229545e-06,
      "loss": 1.2285,
      "step": 460
    },
    {
      "epoch": 9.997300944669366,
      "eval_loss": 1.1040430068969727,
      "eval_runtime": 12.0718,
      "eval_samples_per_second": 30.816,
      "eval_steps_per_second": 3.893,
      "step": 463
    },
    {
      "epoch": 10.364372469635628,
      "grad_norm": 2.478320837020874,
      "learning_rate": 8.44039019311717e-06,
      "loss": 1.1817,
      "step": 480
    },
    {
      "epoch": 10.796221322537113,
      "grad_norm": 2.545358180999756,
      "learning_rate": 7.043686378203864e-06,
      "loss": 1.1507,
      "step": 500
    },
    {
      "epoch": 10.99055330634278,
      "eval_loss": 1.0961867570877075,
      "eval_runtime": 12.0643,
      "eval_samples_per_second": 30.835,
      "eval_steps_per_second": 3.896,
      "step": 509
    },
    {
      "epoch": 11.228070175438596,
      "grad_norm": 2.47627854347229,
      "learning_rate": 5.7371307758071225e-06,
      "loss": 1.207,
      "step": 520
    },
    {
      "epoch": 11.65991902834008,
      "grad_norm": 2.8737947940826416,
      "learning_rate": 4.535527182975231e-06,
      "loss": 1.169,
      "step": 540
    },
    {
      "epoch": 11.983805668016194,
      "eval_loss": 1.094117522239685,
      "eval_runtime": 12.0615,
      "eval_samples_per_second": 30.842,
      "eval_steps_per_second": 3.897,
      "step": 555
    },
    {
      "epoch": 12.091767881241566,
      "grad_norm": 2.6336514949798584,
      "learning_rate": 3.4524902485514042e-06,
      "loss": 1.1437,
      "step": 560
    },
    {
      "epoch": 12.523616734143049,
      "grad_norm": 2.0161590576171875,
      "learning_rate": 2.500291213762274e-06,
      "loss": 1.1425,
      "step": 580
    },
    {
      "epoch": 12.955465587044534,
      "grad_norm": 3.3482580184936523,
      "learning_rate": 1.6897188741514285e-06,
      "loss": 1.1293,
      "step": 600
    },
    {
      "epoch": 12.998650472334683,
      "eval_loss": 1.0909315347671509,
      "eval_runtime": 12.0609,
      "eval_samples_per_second": 30.844,
      "eval_steps_per_second": 3.897,
      "step": 602
    },
    {
      "epoch": 13.387314439946019,
      "grad_norm": 2.3085474967956543,
      "learning_rate": 1.0299573382149235e-06,
      "loss": 1.1141,
      "step": 620
    },
    {
      "epoch": 13.819163292847504,
      "grad_norm": 2.5478107929229736,
      "learning_rate": 5.284819677822611e-07,
      "loss": 1.18,
      "step": 640
    },
    {
      "epoch": 13.991902834008098,
      "eval_loss": 1.0923157930374146,
      "eval_runtime": 12.0657,
      "eval_samples_per_second": 30.831,
      "eval_steps_per_second": 3.895,
      "step": 648
    },
    {
      "epoch": 14.251012145748987,
      "grad_norm": 2.4361941814422607,
      "learning_rate": 1.909746791798317e-07,
      "loss": 1.151,
      "step": 660
    },
    {
      "epoch": 14.682860998650472,
      "grad_norm": 2.4427309036254883,
      "learning_rate": 2.1259564848570835e-08,
      "loss": 1.1257,
      "step": 680
    },
    {
      "epoch": 14.898785425101215,
      "eval_loss": 1.0921934843063354,
      "eval_runtime": 12.0392,
      "eval_samples_per_second": 30.899,
      "eval_steps_per_second": 3.904,
      "step": 690
    }
  ],
  "logging_steps": 20,
  "max_steps": 690,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 15,
  "save_steps": 500,
  "total_flos": 6.133360120720589e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}