{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.969537524231514,
  "global_step": 72000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 1.9861534201052343e-05,
      "loss": 2.0278,
      "step": 500
    },
    {
      "epoch": 0.14,
      "learning_rate": 1.972306840210468e-05,
      "loss": 1.6767,
      "step": 1000
    },
    {
      "epoch": 0.21,
      "learning_rate": 1.958460260315702e-05,
      "loss": 1.5473,
      "step": 1500
    },
    {
      "epoch": 0.28,
      "learning_rate": 1.9446136804209362e-05,
      "loss": 1.4575,
      "step": 2000
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.93076710052617e-05,
      "loss": 1.4112,
      "step": 2500
    },
    {
      "epoch": 0.42,
      "learning_rate": 1.916920520631404e-05,
      "loss": 1.3585,
      "step": 3000
    },
    {
      "epoch": 0.48,
      "learning_rate": 1.903073940736638e-05,
      "loss": 1.338,
      "step": 3500
    },
    {
      "epoch": 0.55,
      "learning_rate": 1.8892273608418723e-05,
      "loss": 1.2933,
      "step": 4000
    },
    {
      "epoch": 0.62,
      "learning_rate": 1.8753807809471064e-05,
      "loss": 1.2931,
      "step": 4500
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.86153420105234e-05,
      "loss": 1.2415,
      "step": 5000
    },
    {
      "epoch": 0.76,
      "learning_rate": 1.8476876211575742e-05,
      "loss": 1.2126,
      "step": 5500
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.8338410412628083e-05,
      "loss": 1.1998,
      "step": 6000
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.8199944613680424e-05,
      "loss": 1.1952,
      "step": 6500
    },
    {
      "epoch": 0.97,
      "learning_rate": 1.806147881473276e-05,
      "loss": 1.1406,
      "step": 7000
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.0695199966430664,
      "eval_runtime": 44.0148,
      "eval_samples_per_second": 659.142,
      "eval_steps_per_second": 82.404,
      "step": 7222
    },
    {
      "epoch": 1.04,
      "learning_rate": 1.7923013015785103e-05,
      "loss": 1.1321,
      "step": 7500
    },
    {
      "epoch": 1.11,
      "learning_rate": 1.7784547216837444e-05,
      "loss": 1.1054,
      "step": 8000
    },
    {
      "epoch": 1.18,
      "learning_rate": 1.764608141788978e-05,
      "loss": 1.1185,
      "step": 8500
    },
    {
      "epoch": 1.25,
      "learning_rate": 1.7507615618942122e-05,
      "loss": 1.0977,
      "step": 9000
    },
    {
      "epoch": 1.32,
      "learning_rate": 1.7369149819994463e-05,
      "loss": 1.1232,
      "step": 9500
    },
    {
      "epoch": 1.38,
      "learning_rate": 1.72306840210468e-05,
      "loss": 1.0793,
      "step": 10000
    },
    {
      "epoch": 1.45,
      "learning_rate": 1.7092218222099145e-05,
      "loss": 1.0588,
      "step": 10500
    },
    {
      "epoch": 1.52,
      "learning_rate": 1.6953752423151483e-05,
      "loss": 1.048,
      "step": 11000
    },
    {
      "epoch": 1.59,
      "learning_rate": 1.6815286624203824e-05,
      "loss": 1.036,
      "step": 11500
    },
    {
      "epoch": 1.66,
      "learning_rate": 1.6676820825256165e-05,
      "loss": 1.0414,
      "step": 12000
    },
    {
      "epoch": 1.73,
      "learning_rate": 1.6538355026308502e-05,
      "loss": 1.0109,
      "step": 12500
    },
    {
      "epoch": 1.8,
      "learning_rate": 1.6399889227360843e-05,
      "loss": 1.0186,
      "step": 13000
    },
    {
      "epoch": 1.87,
      "learning_rate": 1.6261423428413184e-05,
      "loss": 1.0456,
      "step": 13500
    },
    {
      "epoch": 1.94,
      "learning_rate": 1.612295762946552e-05,
      "loss": 0.999,
      "step": 14000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.9253028035163879,
      "eval_runtime": 44.5087,
      "eval_samples_per_second": 651.828,
      "eval_steps_per_second": 81.49,
      "step": 14444
    },
    {
      "epoch": 2.01,
      "learning_rate": 1.5984491830517863e-05,
      "loss": 0.9973,
      "step": 14500
    },
    {
      "epoch": 2.08,
      "learning_rate": 1.5846026031570204e-05,
      "loss": 1.0,
      "step": 15000
    },
    {
      "epoch": 2.15,
      "learning_rate": 1.5707560232622545e-05,
      "loss": 0.9893,
      "step": 15500
    },
    {
      "epoch": 2.22,
      "learning_rate": 1.5569094433674882e-05,
      "loss": 0.9825,
      "step": 16000
    },
    {
      "epoch": 2.28,
      "learning_rate": 1.5430628634727223e-05,
      "loss": 0.9775,
      "step": 16500
    },
    {
      "epoch": 2.35,
      "learning_rate": 1.5292162835779564e-05,
      "loss": 0.9564,
      "step": 17000
    },
    {
      "epoch": 2.42,
      "learning_rate": 1.5153697036831905e-05,
      "loss": 0.9555,
      "step": 17500
    },
    {
      "epoch": 2.49,
      "learning_rate": 1.5015231237884244e-05,
      "loss": 0.9441,
      "step": 18000
    },
    {
      "epoch": 2.56,
      "learning_rate": 1.4876765438936585e-05,
      "loss": 0.957,
      "step": 18500
    },
    {
      "epoch": 2.63,
      "learning_rate": 1.4738299639988925e-05,
      "loss": 0.9341,
      "step": 19000
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.4599833841041264e-05,
      "loss": 0.9222,
      "step": 19500
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.4461368042093605e-05,
      "loss": 0.9204,
      "step": 20000
    },
    {
      "epoch": 2.84,
      "learning_rate": 1.4322902243145944e-05,
      "loss": 0.9379,
      "step": 20500
    },
    {
      "epoch": 2.91,
      "learning_rate": 1.4184436444198283e-05,
      "loss": 0.9113,
      "step": 21000
    },
    {
      "epoch": 2.98,
      "learning_rate": 1.4045970645250624e-05,
      "loss": 0.9015,
      "step": 21500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.8526855707168579,
      "eval_runtime": 44.1672,
      "eval_samples_per_second": 656.868,
      "eval_steps_per_second": 82.12,
      "step": 21666
    },
    {
      "epoch": 3.05,
      "learning_rate": 1.3907504846302964e-05,
      "loss": 0.8847,
      "step": 22000
    },
    {
      "epoch": 3.12,
      "learning_rate": 1.3769039047355303e-05,
      "loss": 0.9032,
      "step": 22500
    },
    {
      "epoch": 3.18,
      "learning_rate": 1.3630573248407644e-05,
      "loss": 0.9074,
      "step": 23000
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.3492107449459983e-05,
      "loss": 0.8774,
      "step": 23500
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.3353641650512326e-05,
      "loss": 0.9121,
      "step": 24000
    },
    {
      "epoch": 3.39,
      "learning_rate": 1.3215175851564665e-05,
      "loss": 0.8713,
      "step": 24500
    },
    {
      "epoch": 3.46,
      "learning_rate": 1.3076710052617006e-05,
      "loss": 0.8623,
      "step": 25000
    },
    {
      "epoch": 3.53,
      "learning_rate": 1.2938244253669345e-05,
      "loss": 0.899,
      "step": 25500
    },
    {
      "epoch": 3.6,
      "learning_rate": 1.2799778454721685e-05,
      "loss": 0.8702,
      "step": 26000
    },
    {
      "epoch": 3.67,
      "learning_rate": 1.2661312655774026e-05,
      "loss": 0.884,
      "step": 26500
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.2522846856826365e-05,
      "loss": 0.8518,
      "step": 27000
    },
    {
      "epoch": 3.81,
      "learning_rate": 1.2384381057878704e-05,
      "loss": 0.8816,
      "step": 27500
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.2245915258931045e-05,
      "loss": 0.8495,
      "step": 28000
    },
    {
      "epoch": 3.95,
      "learning_rate": 1.2107449459983384e-05,
      "loss": 0.8663,
      "step": 28500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.7954610586166382,
      "eval_runtime": 44.1292,
      "eval_samples_per_second": 657.434,
      "eval_steps_per_second": 82.191,
      "step": 28888
    },
    {
      "epoch": 4.02,
      "learning_rate": 1.1968983661035725e-05,
      "loss": 0.852,
      "step": 29000
    },
    {
      "epoch": 4.08,
      "learning_rate": 1.1830517862088065e-05,
      "loss": 0.8425,
      "step": 29500
    },
    {
      "epoch": 4.15,
      "learning_rate": 1.1692052063140404e-05,
      "loss": 0.8379,
      "step": 30000
    },
    {
      "epoch": 4.22,
      "learning_rate": 1.1553586264192745e-05,
      "loss": 0.859,
      "step": 30500
    },
    {
      "epoch": 4.29,
      "learning_rate": 1.1415120465245086e-05,
      "loss": 0.8449,
      "step": 31000
    },
    {
      "epoch": 4.36,
      "learning_rate": 1.1276654666297427e-05,
      "loss": 0.8333,
      "step": 31500
    },
    {
      "epoch": 4.43,
      "learning_rate": 1.1138188867349766e-05,
      "loss": 0.8225,
      "step": 32000
    },
    {
      "epoch": 4.5,
      "learning_rate": 1.0999723068402105e-05,
      "loss": 0.8356,
      "step": 32500
    },
    {
      "epoch": 4.57,
      "learning_rate": 1.0861257269454446e-05,
      "loss": 0.839,
      "step": 33000
    },
    {
      "epoch": 4.64,
      "learning_rate": 1.0722791470506786e-05,
      "loss": 0.8511,
      "step": 33500
    },
    {
      "epoch": 4.71,
      "learning_rate": 1.0584325671559127e-05,
      "loss": 0.8518,
      "step": 34000
    },
    {
      "epoch": 4.78,
      "learning_rate": 1.0445859872611466e-05,
      "loss": 0.8101,
      "step": 34500
    },
    {
      "epoch": 4.85,
      "learning_rate": 1.0307394073663805e-05,
      "loss": 0.8172,
      "step": 35000
    },
    {
      "epoch": 4.92,
      "learning_rate": 1.0168928274716146e-05,
      "loss": 0.8387,
      "step": 35500
    },
    {
      "epoch": 4.98,
      "learning_rate": 1.0030462475768485e-05,
      "loss": 0.7919,
      "step": 36000
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.7614109516143799,
      "eval_runtime": 24.141,
      "eval_samples_per_second": 1201.775,
      "eval_steps_per_second": 150.243,
      "step": 36110
    },
    {
      "epoch": 5.05,
      "learning_rate": 9.891996676820826e-06,
      "loss": 0.806,
      "step": 36500
    },
    {
      "epoch": 5.12,
      "learning_rate": 9.753530877873166e-06,
      "loss": 0.8068,
      "step": 37000
    },
    {
      "epoch": 5.19,
      "learning_rate": 9.615065078925507e-06,
      "loss": 0.8124,
      "step": 37500
    },
    {
      "epoch": 5.26,
      "learning_rate": 9.476599279977846e-06,
      "loss": 0.7742,
      "step": 38000
    },
    {
      "epoch": 5.33,
      "learning_rate": 9.338133481030187e-06,
      "loss": 0.7928,
      "step": 38500
    },
    {
      "epoch": 5.4,
      "learning_rate": 9.199667682082526e-06,
      "loss": 0.8034,
      "step": 39000
    },
    {
      "epoch": 5.47,
      "learning_rate": 9.061201883134867e-06,
      "loss": 0.802,
      "step": 39500
    },
    {
      "epoch": 5.54,
      "learning_rate": 8.922736084187206e-06,
      "loss": 0.7834,
      "step": 40000
    },
    {
      "epoch": 5.61,
      "learning_rate": 8.784270285239547e-06,
      "loss": 0.7892,
      "step": 40500
    },
    {
      "epoch": 5.68,
      "learning_rate": 8.645804486291887e-06,
      "loss": 0.7783,
      "step": 41000
    },
    {
      "epoch": 5.75,
      "learning_rate": 8.507338687344226e-06,
      "loss": 0.7757,
      "step": 41500
    },
    {
      "epoch": 5.82,
      "learning_rate": 8.368872888396567e-06,
      "loss": 0.7774,
      "step": 42000
    },
    {
      "epoch": 5.88,
      "learning_rate": 8.230407089448906e-06,
      "loss": 0.7684,
      "step": 42500
    },
    {
      "epoch": 5.95,
      "learning_rate": 8.091941290501247e-06,
      "loss": 0.7933,
      "step": 43000
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.7296460866928101,
      "eval_runtime": 22.896,
      "eval_samples_per_second": 1267.119,
      "eval_steps_per_second": 158.412,
      "step": 43332
    },
    {
      "epoch": 6.02,
      "learning_rate": 7.953475491553588e-06,
      "loss": 0.7829,
      "step": 43500
    },
    {
      "epoch": 6.09,
      "learning_rate": 7.815009692605927e-06,
      "loss": 0.7581,
      "step": 44000
    },
    {
      "epoch": 6.16,
      "learning_rate": 7.676543893658267e-06,
      "loss": 0.7389,
      "step": 44500
    },
    {
      "epoch": 6.23,
      "learning_rate": 7.538078094710607e-06,
      "loss": 0.767,
      "step": 45000
    },
    {
      "epoch": 6.3,
      "learning_rate": 7.399612295762947e-06,
      "loss": 0.7517,
      "step": 45500
    },
    {
      "epoch": 6.37,
      "learning_rate": 7.261146496815287e-06,
      "loss": 0.7572,
      "step": 46000
    },
    {
      "epoch": 6.44,
      "learning_rate": 7.122680697867628e-06,
      "loss": 0.7501,
      "step": 46500
    },
    {
      "epoch": 6.51,
      "learning_rate": 6.984214898919968e-06,
      "loss": 0.7603,
      "step": 47000
    },
    {
      "epoch": 6.58,
      "learning_rate": 6.845749099972307e-06,
      "loss": 0.7432,
      "step": 47500
    },
    {
      "epoch": 6.65,
      "learning_rate": 6.707283301024647e-06,
      "loss": 0.7523,
      "step": 48000
    },
    {
      "epoch": 6.72,
      "learning_rate": 6.5688175020769875e-06,
      "loss": 0.7652,
      "step": 48500
    },
    {
      "epoch": 6.78,
      "learning_rate": 6.430351703129327e-06,
      "loss": 0.7707,
      "step": 49000
    },
    {
      "epoch": 6.85,
      "learning_rate": 6.291885904181667e-06,
      "loss": 0.7512,
      "step": 49500
    },
    {
      "epoch": 6.92,
      "learning_rate": 6.153420105234008e-06,
      "loss": 0.7407,
      "step": 50000
    },
    {
      "epoch": 6.99,
      "learning_rate": 6.014954306286348e-06,
      "loss": 0.7713,
      "step": 50500
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.6923113465309143,
      "eval_runtime": 24.1974,
      "eval_samples_per_second": 1198.969,
      "eval_steps_per_second": 149.892,
      "step": 50554
    },
    {
      "epoch": 7.06,
      "learning_rate": 5.876488507338688e-06,
      "loss": 0.7631,
      "step": 51000
    },
    {
      "epoch": 7.13,
      "learning_rate": 5.738022708391027e-06,
      "loss": 0.7597,
      "step": 51500
    },
    {
      "epoch": 7.2,
      "learning_rate": 5.5995569094433675e-06,
      "loss": 0.7353,
      "step": 52000
    },
    {
      "epoch": 7.27,
      "learning_rate": 5.461091110495708e-06,
      "loss": 0.7588,
      "step": 52500
    },
    {
      "epoch": 7.34,
      "learning_rate": 5.322625311548048e-06,
      "loss": 0.7333,
      "step": 53000
    },
    {
      "epoch": 7.41,
      "learning_rate": 5.184159512600389e-06,
      "loss": 0.7562,
      "step": 53500
    },
    {
      "epoch": 7.48,
      "learning_rate": 5.045693713652728e-06,
      "loss": 0.7262,
      "step": 54000
    },
    {
      "epoch": 7.55,
      "learning_rate": 4.907227914705068e-06,
      "loss": 0.7627,
      "step": 54500
    },
    {
      "epoch": 7.62,
      "learning_rate": 4.768762115757408e-06,
      "loss": 0.7443,
      "step": 55000
    },
    {
      "epoch": 7.68,
      "learning_rate": 4.630296316809748e-06,
      "loss": 0.758,
      "step": 55500
    },
    {
      "epoch": 7.75,
      "learning_rate": 4.4918305178620885e-06,
      "loss": 0.7295,
      "step": 56000
    },
    {
      "epoch": 7.82,
      "learning_rate": 4.353364718914429e-06,
      "loss": 0.726,
      "step": 56500
    },
    {
      "epoch": 7.89,
      "learning_rate": 4.214898919966769e-06,
      "loss": 0.7335,
      "step": 57000
    },
    {
      "epoch": 7.96,
      "learning_rate": 4.076433121019109e-06,
      "loss": 0.7249,
      "step": 57500
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.6828723549842834,
      "eval_runtime": 22.9367,
      "eval_samples_per_second": 1264.871,
      "eval_steps_per_second": 158.131,
      "step": 57776
    },
    {
      "epoch": 8.03,
      "learning_rate": 3.937967322071449e-06,
      "loss": 0.719,
      "step": 58000
    },
    {
      "epoch": 8.1,
      "learning_rate": 3.7995015231237887e-06,
      "loss": 0.7539,
      "step": 58500
    },
    {
      "epoch": 8.17,
      "learning_rate": 3.6610357241761292e-06,
      "loss": 0.7311,
      "step": 59000
    },
    {
      "epoch": 8.24,
      "learning_rate": 3.522569925228469e-06,
      "loss": 0.7186,
      "step": 59500
    },
    {
      "epoch": 8.31,
      "learning_rate": 3.384104126280809e-06,
      "loss": 0.722,
      "step": 60000
    },
    {
      "epoch": 8.38,
      "learning_rate": 3.2456383273331488e-06,
      "loss": 0.6963,
      "step": 60500
    },
    {
      "epoch": 8.45,
      "learning_rate": 3.1071725283854893e-06,
      "loss": 0.7177,
      "step": 61000
    },
    {
      "epoch": 8.52,
      "learning_rate": 2.968706729437829e-06,
      "loss": 0.7121,
      "step": 61500
    },
    {
      "epoch": 8.58,
      "learning_rate": 2.830240930490169e-06,
      "loss": 0.7165,
      "step": 62000
    },
    {
      "epoch": 8.65,
      "learning_rate": 2.6917751315425097e-06,
      "loss": 0.6935,
      "step": 62500
    },
    {
      "epoch": 8.72,
      "learning_rate": 2.5533093325948494e-06,
      "loss": 0.6841,
      "step": 63000
    },
    {
      "epoch": 8.79,
      "learning_rate": 2.414843533647189e-06,
      "loss": 0.7214,
      "step": 63500
    },
    {
      "epoch": 8.86,
      "learning_rate": 2.2763777346995296e-06,
      "loss": 0.7061,
      "step": 64000
    },
    {
      "epoch": 8.93,
      "learning_rate": 2.1379119357518693e-06,
      "loss": 0.7083,
      "step": 64500
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.6712043881416321,
      "eval_runtime": 22.8269,
      "eval_samples_per_second": 1270.954,
      "eval_steps_per_second": 158.891,
      "step": 64998
    },
    {
      "epoch": 9.0,
      "learning_rate": 1.9994461368042094e-06,
      "loss": 0.709,
      "step": 65000
    },
    {
      "epoch": 9.07,
      "learning_rate": 1.8609803378565498e-06,
      "loss": 0.6984,
      "step": 65500
    },
    {
      "epoch": 9.14,
      "learning_rate": 1.7225145389088897e-06,
      "loss": 0.7085,
      "step": 66000
    },
    {
      "epoch": 9.21,
      "learning_rate": 1.5840487399612298e-06,
      "loss": 0.7019,
      "step": 66500
    },
    {
      "epoch": 9.28,
      "learning_rate": 1.4455829410135697e-06,
      "loss": 0.7084,
      "step": 67000
    },
    {
      "epoch": 9.35,
      "learning_rate": 1.3071171420659098e-06,
      "loss": 0.6973,
      "step": 67500
    },
    {
      "epoch": 9.42,
      "learning_rate": 1.16865134311825e-06,
      "loss": 0.698,
      "step": 68000
    },
    {
      "epoch": 9.48,
      "learning_rate": 1.0301855441705899e-06,
      "loss": 0.7298,
      "step": 68500
    },
    {
      "epoch": 9.55,
      "learning_rate": 8.9171974522293e-07,
      "loss": 0.7163,
      "step": 69000
    },
    {
      "epoch": 9.62,
      "learning_rate": 7.532539462752701e-07,
      "loss": 0.6955,
      "step": 69500
    },
    {
      "epoch": 9.69,
      "learning_rate": 6.147881473276101e-07,
      "loss": 0.6855,
      "step": 70000
    },
    {
      "epoch": 9.76,
      "learning_rate": 4.763223483799502e-07,
      "loss": 0.7026,
      "step": 70500
    },
    {
      "epoch": 9.83,
      "learning_rate": 3.3785654943229026e-07,
      "loss": 0.7055,
      "step": 71000
    },
    {
      "epoch": 9.9,
      "learning_rate": 1.993907504846303e-07,
      "loss": 0.6855,
      "step": 71500
    },
    {
      "epoch": 9.97,
      "learning_rate": 6.092495153697037e-08,
      "loss": 0.717,
      "step": 72000
    }
  ],
  "max_steps": 72220,
  "num_train_epochs": 10,
  "total_flos": 3729696905063250.0,
  "trial_name": null,
  "trial_params": null
}