{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 19.607843137254903,
  "eval_steps": 250,
  "global_step": 12000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.41,
      "learning_rate": 1.2254901960784313e-05,
      "loss": 2.8121,
      "step": 250
    },
    {
      "epoch": 0.41,
      "eval_loss": 0.8146461248397827,
      "eval_runtime": 28.4909,
      "eval_samples_per_second": 50.086,
      "eval_steps_per_second": 3.159,
      "step": 250
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.4509803921568626e-05,
      "loss": 0.7537,
      "step": 500
    },
    {
      "epoch": 0.82,
      "eval_loss": 0.7322247624397278,
      "eval_runtime": 28.2904,
      "eval_samples_per_second": 50.441,
      "eval_steps_per_second": 3.181,
      "step": 500
    },
    {
      "epoch": 1.23,
      "learning_rate": 2.9643962848297217e-05,
      "loss": 0.6434,
      "step": 750
    },
    {
      "epoch": 1.23,
      "eval_loss": 0.6870973706245422,
      "eval_runtime": 28.2349,
      "eval_samples_per_second": 50.54,
      "eval_steps_per_second": 3.188,
      "step": 750
    },
    {
      "epoch": 1.63,
      "learning_rate": 2.8998968008255932e-05,
      "loss": 0.6207,
      "step": 1000
    },
    {
      "epoch": 1.63,
      "eval_loss": 0.6630086898803711,
      "eval_runtime": 28.2684,
      "eval_samples_per_second": 50.48,
      "eval_steps_per_second": 3.184,
      "step": 1000
    },
    {
      "epoch": 2.04,
      "learning_rate": 2.8353973168214654e-05,
      "loss": 0.5771,
      "step": 1250
    },
    {
      "epoch": 2.04,
      "eval_loss": 0.6516671776771545,
      "eval_runtime": 28.2447,
      "eval_samples_per_second": 50.523,
      "eval_steps_per_second": 3.186,
      "step": 1250
    },
    {
      "epoch": 2.45,
      "learning_rate": 2.7708978328173376e-05,
      "loss": 0.54,
      "step": 1500
    },
    {
      "epoch": 2.45,
      "eval_loss": 0.6349040269851685,
      "eval_runtime": 28.3054,
      "eval_samples_per_second": 50.414,
      "eval_steps_per_second": 3.18,
      "step": 1500
    },
    {
      "epoch": 2.86,
      "learning_rate": 2.7063983488132095e-05,
      "loss": 0.5286,
      "step": 1750
    },
    {
      "epoch": 2.86,
      "eval_loss": 0.6304510831832886,
      "eval_runtime": 28.2763,
      "eval_samples_per_second": 50.466,
      "eval_steps_per_second": 3.183,
      "step": 1750
    },
    {
      "epoch": 3.27,
      "learning_rate": 2.6418988648090816e-05,
      "loss": 0.4998,
      "step": 2000
    },
    {
      "epoch": 3.27,
      "eval_loss": 0.6260914206504822,
      "eval_runtime": 28.3237,
      "eval_samples_per_second": 50.382,
      "eval_steps_per_second": 3.178,
      "step": 2000
    },
    {
      "epoch": 3.68,
      "learning_rate": 2.577399380804954e-05,
      "loss": 0.4874,
      "step": 2250
    },
    {
      "epoch": 3.68,
      "eval_loss": 0.6250064969062805,
      "eval_runtime": 28.23,
      "eval_samples_per_second": 50.549,
      "eval_steps_per_second": 3.188,
      "step": 2250
    },
    {
      "epoch": 4.08,
      "learning_rate": 2.5128998968008257e-05,
      "loss": 0.4704,
      "step": 2500
    },
    {
      "epoch": 4.08,
      "eval_loss": 0.6252643465995789,
      "eval_runtime": 28.3447,
      "eval_samples_per_second": 50.344,
      "eval_steps_per_second": 3.175,
      "step": 2500
    },
    {
      "epoch": 4.49,
      "learning_rate": 2.4484004127966975e-05,
      "loss": 0.4431,
      "step": 2750
    },
    {
      "epoch": 4.49,
      "eval_loss": 0.6222314238548279,
      "eval_runtime": 28.3136,
      "eval_samples_per_second": 50.4,
      "eval_steps_per_second": 3.179,
      "step": 2750
    },
    {
      "epoch": 4.9,
      "learning_rate": 2.3839009287925697e-05,
      "loss": 0.4502,
      "step": 3000
    },
    {
      "epoch": 4.9,
      "eval_loss": 0.6138860583305359,
      "eval_runtime": 28.2755,
      "eval_samples_per_second": 50.468,
      "eval_steps_per_second": 3.183,
      "step": 3000
    },
    {
      "epoch": 5.31,
      "learning_rate": 2.3194014447884416e-05,
      "loss": 0.4216,
      "step": 3250
    },
    {
      "epoch": 5.31,
      "eval_loss": 0.6158970594406128,
      "eval_runtime": 28.2142,
      "eval_samples_per_second": 50.577,
      "eval_steps_per_second": 3.19,
      "step": 3250
    },
    {
      "epoch": 5.72,
      "learning_rate": 2.2549019607843138e-05,
      "loss": 0.4146,
      "step": 3500
    },
    {
      "epoch": 5.72,
      "eval_loss": 0.6141221523284912,
      "eval_runtime": 28.2223,
      "eval_samples_per_second": 50.563,
      "eval_steps_per_second": 3.189,
      "step": 3500
    },
    {
      "epoch": 6.13,
      "learning_rate": 2.190402476780186e-05,
      "loss": 0.4079,
      "step": 3750
    },
    {
      "epoch": 6.13,
      "eval_loss": 0.6209476590156555,
      "eval_runtime": 28.2459,
      "eval_samples_per_second": 50.521,
      "eval_steps_per_second": 3.186,
      "step": 3750
    },
    {
      "epoch": 6.54,
      "learning_rate": 2.1259029927760578e-05,
      "loss": 0.3882,
      "step": 4000
    },
    {
      "epoch": 6.54,
      "eval_loss": 0.6084998250007629,
      "eval_runtime": 28.195,
      "eval_samples_per_second": 50.612,
      "eval_steps_per_second": 3.192,
      "step": 4000
    },
    {
      "epoch": 6.94,
      "learning_rate": 2.06140350877193e-05,
      "loss": 0.3917,
      "step": 4250
    },
    {
      "epoch": 6.94,
      "eval_loss": 0.6080483198165894,
      "eval_runtime": 28.2021,
      "eval_samples_per_second": 50.599,
      "eval_steps_per_second": 3.191,
      "step": 4250
    },
    {
      "epoch": 7.35,
      "learning_rate": 1.9969040247678018e-05,
      "loss": 0.3654,
      "step": 4500
    },
    {
      "epoch": 7.35,
      "eval_loss": 0.616654634475708,
      "eval_runtime": 28.1521,
      "eval_samples_per_second": 50.689,
      "eval_steps_per_second": 3.197,
      "step": 4500
    },
    {
      "epoch": 7.76,
      "learning_rate": 1.9324045407636737e-05,
      "loss": 0.3672,
      "step": 4750
    },
    {
      "epoch": 7.76,
      "eval_loss": 0.6128713488578796,
      "eval_runtime": 28.1986,
      "eval_samples_per_second": 50.605,
      "eval_steps_per_second": 3.192,
      "step": 4750
    },
    {
      "epoch": 8.17,
      "learning_rate": 1.867905056759546e-05,
      "loss": 0.3562,
      "step": 5000
    },
    {
      "epoch": 8.17,
      "eval_loss": 0.6237680315971375,
      "eval_runtime": 28.209,
      "eval_samples_per_second": 50.587,
      "eval_steps_per_second": 3.19,
      "step": 5000
    },
    {
      "epoch": 8.58,
      "learning_rate": 1.803405572755418e-05,
      "loss": 0.3383,
      "step": 5250
    },
    {
      "epoch": 8.58,
      "eval_loss": 0.6211274266242981,
      "eval_runtime": 28.1959,
      "eval_samples_per_second": 50.61,
      "eval_steps_per_second": 3.192,
      "step": 5250
    },
    {
      "epoch": 8.99,
      "learning_rate": 1.73890608875129e-05,
      "loss": 0.347,
      "step": 5500
    },
    {
      "epoch": 8.99,
      "eval_loss": 0.6140613555908203,
      "eval_runtime": 28.169,
      "eval_samples_per_second": 50.659,
      "eval_steps_per_second": 3.195,
      "step": 5500
    },
    {
      "epoch": 9.4,
      "learning_rate": 1.674406604747162e-05,
      "loss": 0.3209,
      "step": 5750
    },
    {
      "epoch": 9.4,
      "eval_loss": 0.6265636682510376,
      "eval_runtime": 28.1893,
      "eval_samples_per_second": 50.622,
      "eval_steps_per_second": 3.193,
      "step": 5750
    },
    {
      "epoch": 9.8,
      "learning_rate": 1.6099071207430343e-05,
      "loss": 0.3268,
      "step": 6000
    },
    {
      "epoch": 9.8,
      "eval_loss": 0.6181433796882629,
      "eval_runtime": 28.1932,
      "eval_samples_per_second": 50.615,
      "eval_steps_per_second": 3.192,
      "step": 6000
    },
    {
      "epoch": 10.21,
      "learning_rate": 1.545407636738906e-05,
      "loss": 0.3126,
      "step": 6250
    },
    {
      "epoch": 10.21,
      "eval_loss": 0.6234831809997559,
      "eval_runtime": 28.1977,
      "eval_samples_per_second": 50.607,
      "eval_steps_per_second": 3.192,
      "step": 6250
    },
    {
      "epoch": 10.62,
      "learning_rate": 1.4809081527347781e-05,
      "loss": 0.3075,
      "step": 6500
    },
    {
      "epoch": 10.62,
      "eval_loss": 0.6242398023605347,
      "eval_runtime": 28.2355,
      "eval_samples_per_second": 50.539,
      "eval_steps_per_second": 3.187,
      "step": 6500
    },
    {
      "epoch": 11.03,
      "learning_rate": 1.4164086687306502e-05,
      "loss": 0.3113,
      "step": 6750
    },
    {
      "epoch": 11.03,
      "eval_loss": 0.6290929913520813,
      "eval_runtime": 28.2357,
      "eval_samples_per_second": 50.539,
      "eval_steps_per_second": 3.187,
      "step": 6750
    },
    {
      "epoch": 11.44,
      "learning_rate": 1.3519091847265222e-05,
      "loss": 0.2921,
      "step": 7000
    },
    {
      "epoch": 11.44,
      "eval_loss": 0.6291936039924622,
      "eval_runtime": 28.3781,
      "eval_samples_per_second": 50.285,
      "eval_steps_per_second": 3.171,
      "step": 7000
    },
    {
      "epoch": 11.85,
      "learning_rate": 1.2874097007223944e-05,
      "loss": 0.2933,
      "step": 7250
    },
    {
      "epoch": 11.85,
      "eval_loss": 0.6279457807540894,
      "eval_runtime": 28.2812,
      "eval_samples_per_second": 50.457,
      "eval_steps_per_second": 3.182,
      "step": 7250
    },
    {
      "epoch": 12.25,
      "learning_rate": 1.2229102167182664e-05,
      "loss": 0.2808,
      "step": 7500
    },
    {
      "epoch": 12.25,
      "eval_loss": 0.6309792995452881,
      "eval_runtime": 28.2292,
      "eval_samples_per_second": 50.55,
      "eval_steps_per_second": 3.188,
      "step": 7500
    },
    {
      "epoch": 12.66,
      "learning_rate": 1.1584107327141382e-05,
      "loss": 0.2809,
      "step": 7750
    },
    {
      "epoch": 12.66,
      "eval_loss": 0.6356364488601685,
      "eval_runtime": 28.2615,
      "eval_samples_per_second": 50.493,
      "eval_steps_per_second": 3.185,
      "step": 7750
    },
    {
      "epoch": 13.07,
      "learning_rate": 1.0939112487100104e-05,
      "loss": 0.2778,
      "step": 8000
    },
    {
      "epoch": 13.07,
      "eval_loss": 0.6354484558105469,
      "eval_runtime": 28.3136,
      "eval_samples_per_second": 50.4,
      "eval_steps_per_second": 3.179,
      "step": 8000
    },
    {
      "epoch": 13.48,
      "learning_rate": 1.0294117647058824e-05,
      "loss": 0.2707,
      "step": 8250
    },
    {
      "epoch": 13.48,
      "eval_loss": 0.6366216540336609,
      "eval_runtime": 28.3106,
      "eval_samples_per_second": 50.405,
      "eval_steps_per_second": 3.179,
      "step": 8250
    },
    {
      "epoch": 13.89,
      "learning_rate": 9.649122807017543e-06,
      "loss": 0.2644,
      "step": 8500
    },
    {
      "epoch": 13.89,
      "eval_loss": 0.6386236548423767,
      "eval_runtime": 28.3478,
      "eval_samples_per_second": 50.339,
      "eval_steps_per_second": 3.175,
      "step": 8500
    },
    {
      "epoch": 14.3,
      "learning_rate": 9.004127966976265e-06,
      "loss": 0.2574,
      "step": 8750
    },
    {
      "epoch": 14.3,
      "eval_loss": 0.6385951638221741,
      "eval_runtime": 28.3245,
      "eval_samples_per_second": 50.38,
      "eval_steps_per_second": 3.177,
      "step": 8750
    },
    {
      "epoch": 14.71,
      "learning_rate": 8.359133126934985e-06,
      "loss": 0.2564,
      "step": 9000
    },
    {
      "epoch": 14.71,
      "eval_loss": 0.6378169059753418,
      "eval_runtime": 28.3461,
      "eval_samples_per_second": 50.342,
      "eval_steps_per_second": 3.175,
      "step": 9000
    },
    {
      "epoch": 15.11,
      "learning_rate": 7.714138286893705e-06,
      "loss": 0.2555,
      "step": 9250
    },
    {
      "epoch": 15.11,
      "eval_loss": 0.6417073011398315,
      "eval_runtime": 28.3435,
      "eval_samples_per_second": 50.347,
      "eval_steps_per_second": 3.175,
      "step": 9250
    },
    {
      "epoch": 15.52,
      "learning_rate": 7.069143446852426e-06,
      "loss": 0.2478,
      "step": 9500
    },
    {
      "epoch": 15.52,
      "eval_loss": 0.6406723260879517,
      "eval_runtime": 28.3209,
      "eval_samples_per_second": 50.387,
      "eval_steps_per_second": 3.178,
      "step": 9500
    },
    {
      "epoch": 15.93,
      "learning_rate": 6.424148606811145e-06,
      "loss": 0.2504,
      "step": 9750
    },
    {
      "epoch": 15.93,
      "eval_loss": 0.6424481272697449,
      "eval_runtime": 28.3698,
      "eval_samples_per_second": 50.3,
      "eval_steps_per_second": 3.172,
      "step": 9750
    },
    {
      "epoch": 16.34,
      "learning_rate": 5.779153766769866e-06,
      "loss": 0.2437,
      "step": 10000
    },
    {
      "epoch": 16.34,
      "eval_loss": 0.6423542499542236,
      "eval_runtime": 28.3599,
      "eval_samples_per_second": 50.318,
      "eval_steps_per_second": 3.173,
      "step": 10000
    },
    {
      "epoch": 16.75,
      "learning_rate": 5.134158926728587e-06,
      "loss": 0.2399,
      "step": 10250
    },
    {
      "epoch": 16.75,
      "eval_loss": 0.645331621170044,
      "eval_runtime": 28.3144,
      "eval_samples_per_second": 50.398,
      "eval_steps_per_second": 3.179,
      "step": 10250
    },
    {
      "epoch": 17.16,
      "learning_rate": 4.489164086687306e-06,
      "loss": 0.238,
      "step": 10500
    },
    {
      "epoch": 17.16,
      "eval_loss": 0.6478310227394104,
      "eval_runtime": 28.3979,
      "eval_samples_per_second": 50.25,
      "eval_steps_per_second": 3.169,
      "step": 10500
    },
    {
      "epoch": 17.57,
      "learning_rate": 3.844169246646027e-06,
      "loss": 0.2359,
      "step": 10750
    },
    {
      "epoch": 17.57,
      "eval_loss": 0.6469157934188843,
      "eval_runtime": 28.3531,
      "eval_samples_per_second": 50.33,
      "eval_steps_per_second": 3.174,
      "step": 10750
    },
    {
      "epoch": 17.97,
      "learning_rate": 3.1991744066047476e-06,
      "loss": 0.2325,
      "step": 11000
    },
    {
      "epoch": 17.97,
      "eval_loss": 0.6457414031028748,
      "eval_runtime": 28.3364,
      "eval_samples_per_second": 50.359,
      "eval_steps_per_second": 3.176,
      "step": 11000
    },
    {
      "epoch": 18.38,
      "learning_rate": 2.5541795665634673e-06,
      "loss": 0.2294,
      "step": 11250
    },
    {
      "epoch": 18.38,
      "eval_loss": 0.6480973958969116,
      "eval_runtime": 28.308,
      "eval_samples_per_second": 50.41,
      "eval_steps_per_second": 3.179,
      "step": 11250
    },
    {
      "epoch": 18.79,
      "learning_rate": 1.909184726522188e-06,
      "loss": 0.2306,
      "step": 11500
    },
    {
      "epoch": 18.79,
      "eval_loss": 0.6470589637756348,
      "eval_runtime": 28.2173,
      "eval_samples_per_second": 50.572,
      "eval_steps_per_second": 3.19,
      "step": 11500
    },
    {
      "epoch": 19.2,
      "learning_rate": 1.264189886480908e-06,
      "loss": 0.2289,
      "step": 11750
    },
    {
      "epoch": 19.2,
      "eval_loss": 0.6486860513687134,
      "eval_runtime": 28.2611,
      "eval_samples_per_second": 50.493,
      "eval_steps_per_second": 3.185,
      "step": 11750
    },
    {
      "epoch": 19.61,
      "learning_rate": 6.191950464396285e-07,
      "loss": 0.2269,
      "step": 12000
    },
    {
      "epoch": 19.61,
      "eval_loss": 0.6482470631599426,
      "eval_runtime": 28.2376,
      "eval_samples_per_second": 50.535,
      "eval_steps_per_second": 3.187,
      "step": 12000
    }
  ],
  "logging_steps": 250,
  "max_steps": 12240,
  "num_train_epochs": 20,
  "save_steps": 250,
  "total_flos": 5.845452187030733e+16,
  "trial_name": null,
  "trial_params": null
}