{
  "best_metric": 3.9967169761657715,
  "best_model_checkpoint": "./models/gpt_test6/checkpoint-10000",
  "epoch": 5.0,
  "global_step": 13240,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 4.2296072507552865e-06,
      "loss": 3.8813,
      "step": 100
    },
    {
      "epoch": 0.08,
      "learning_rate": 8.459214501510573e-06,
      "loss": 3.4212,
      "step": 200
    },
    {
      "epoch": 0.11,
      "learning_rate": 1.2688821752265861e-05,
      "loss": 3.2721,
      "step": 300
    },
    {
      "epoch": 0.15,
      "learning_rate": 1.6918429003021146e-05,
      "loss": 3.2861,
      "step": 400
    },
    {
      "epoch": 0.19,
      "learning_rate": 2.1148036253776434e-05,
      "loss": 3.2347,
      "step": 500
    },
    {
      "epoch": 0.23,
      "learning_rate": 2.5377643504531723e-05,
      "loss": 3.1735,
      "step": 600
    },
    {
      "epoch": 0.26,
      "learning_rate": 2.960725075528701e-05,
      "loss": 3.2013,
      "step": 700
    },
    {
      "epoch": 0.3,
      "learning_rate": 3.383685800604229e-05,
      "loss": 3.1709,
      "step": 800
    },
    {
      "epoch": 0.34,
      "learning_rate": 3.8066465256797584e-05,
      "loss": 3.1716,
      "step": 900
    },
    {
      "epoch": 0.38,
      "learning_rate": 4.229607250755287e-05,
      "loss": 3.1407,
      "step": 1000
    },
    {
      "epoch": 0.38,
      "eval_loss": 4.086385250091553,
      "eval_runtime": 4.2648,
      "eval_samples_per_second": 100.59,
      "eval_steps_per_second": 6.331,
      "step": 1000
    },
    {
      "epoch": 0.42,
      "learning_rate": 4.6525679758308154e-05,
      "loss": 3.1629,
      "step": 1100
    },
    {
      "epoch": 0.45,
      "learning_rate": 5.0755287009063445e-05,
      "loss": 3.1884,
      "step": 1200
    },
    {
      "epoch": 0.49,
      "learning_rate": 5.498489425981873e-05,
      "loss": 3.2097,
      "step": 1300
    },
    {
      "epoch": 0.53,
      "learning_rate": 5.564283316549178e-05,
      "loss": 3.2055,
      "step": 1400
    },
    {
      "epoch": 0.57,
      "learning_rate": 5.517287680429674e-05,
      "loss": 3.1601,
      "step": 1500
    },
    {
      "epoch": 0.6,
      "learning_rate": 5.4702920443101714e-05,
      "loss": 3.138,
      "step": 1600
    },
    {
      "epoch": 0.64,
      "learning_rate": 5.423296408190668e-05,
      "loss": 3.1692,
      "step": 1700
    },
    {
      "epoch": 0.68,
      "learning_rate": 5.376300772071164e-05,
      "loss": 3.1219,
      "step": 1800
    },
    {
      "epoch": 0.72,
      "learning_rate": 5.3293051359516615e-05,
      "loss": 3.1056,
      "step": 1900
    },
    {
      "epoch": 0.76,
      "learning_rate": 5.2823094998321586e-05,
      "loss": 3.1639,
      "step": 2000
    },
    {
      "epoch": 0.76,
      "eval_loss": 4.086682319641113,
      "eval_runtime": 4.2525,
      "eval_samples_per_second": 100.882,
      "eval_steps_per_second": 6.349,
      "step": 2000
    },
    {
      "epoch": 0.79,
      "learning_rate": 5.235313863712655e-05,
      "loss": 3.1307,
      "step": 2100
    },
    {
      "epoch": 0.83,
      "learning_rate": 5.188318227593152e-05,
      "loss": 3.1382,
      "step": 2200
    },
    {
      "epoch": 0.87,
      "learning_rate": 5.1413225914736487e-05,
      "loss": 3.0843,
      "step": 2300
    },
    {
      "epoch": 0.91,
      "learning_rate": 5.094326955354146e-05,
      "loss": 3.094,
      "step": 2400
    },
    {
      "epoch": 0.94,
      "learning_rate": 5.047331319234642e-05,
      "loss": 3.0955,
      "step": 2500
    },
    {
      "epoch": 0.98,
      "learning_rate": 5.000335683115139e-05,
      "loss": 3.0562,
      "step": 2600
    },
    {
      "epoch": 1.02,
      "learning_rate": 4.9533400469956365e-05,
      "loss": 2.9527,
      "step": 2700
    },
    {
      "epoch": 1.06,
      "learning_rate": 4.906344410876133e-05,
      "loss": 2.8783,
      "step": 2800
    },
    {
      "epoch": 1.1,
      "learning_rate": 4.8593487747566294e-05,
      "loss": 2.8926,
      "step": 2900
    },
    {
      "epoch": 1.13,
      "learning_rate": 4.8123531386371266e-05,
      "loss": 2.87,
      "step": 3000
    },
    {
      "epoch": 1.13,
      "eval_loss": 4.06110143661499,
      "eval_runtime": 4.2609,
      "eval_samples_per_second": 100.683,
      "eval_steps_per_second": 6.337,
      "step": 3000
    },
    {
      "epoch": 1.17,
      "learning_rate": 4.765357502517623e-05,
      "loss": 2.8732,
      "step": 3100
    },
    {
      "epoch": 1.21,
      "learning_rate": 4.71836186639812e-05,
      "loss": 2.8985,
      "step": 3200
    },
    {
      "epoch": 1.25,
      "learning_rate": 4.6713662302786166e-05,
      "loss": 2.8827,
      "step": 3300
    },
    {
      "epoch": 1.28,
      "learning_rate": 4.624370594159114e-05,
      "loss": 2.8261,
      "step": 3400
    },
    {
      "epoch": 1.32,
      "learning_rate": 4.577374958039611e-05,
      "loss": 2.9161,
      "step": 3500
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.5303793219201074e-05,
      "loss": 2.8551,
      "step": 3600
    },
    {
      "epoch": 1.4,
      "learning_rate": 4.483383685800604e-05,
      "loss": 2.88,
      "step": 3700
    },
    {
      "epoch": 1.44,
      "learning_rate": 4.436388049681101e-05,
      "loss": 2.8753,
      "step": 3800
    },
    {
      "epoch": 1.47,
      "learning_rate": 4.3893924135615974e-05,
      "loss": 2.8922,
      "step": 3900
    },
    {
      "epoch": 1.51,
      "learning_rate": 4.3423967774420946e-05,
      "loss": 2.8591,
      "step": 4000
    },
    {
      "epoch": 1.51,
      "eval_loss": 4.031496524810791,
      "eval_runtime": 4.2566,
      "eval_samples_per_second": 100.784,
      "eval_steps_per_second": 6.343,
      "step": 4000
    },
    {
      "epoch": 1.55,
      "learning_rate": 4.295401141322592e-05,
      "loss": 2.8911,
      "step": 4100
    },
    {
      "epoch": 1.59,
      "learning_rate": 4.248405505203088e-05,
      "loss": 2.8629,
      "step": 4200
    },
    {
      "epoch": 1.62,
      "learning_rate": 4.201409869083585e-05,
      "loss": 2.8655,
      "step": 4300
    },
    {
      "epoch": 1.66,
      "learning_rate": 4.154414232964082e-05,
      "loss": 2.9125,
      "step": 4400
    },
    {
      "epoch": 1.7,
      "learning_rate": 4.107418596844578e-05,
      "loss": 2.9118,
      "step": 4500
    },
    {
      "epoch": 1.74,
      "learning_rate": 4.0604229607250753e-05,
      "loss": 2.8677,
      "step": 4600
    },
    {
      "epoch": 1.77,
      "learning_rate": 4.0134273246055725e-05,
      "loss": 2.8446,
      "step": 4700
    },
    {
      "epoch": 1.81,
      "learning_rate": 3.9664316884860696e-05,
      "loss": 2.8516,
      "step": 4800
    },
    {
      "epoch": 1.85,
      "learning_rate": 3.919436052366566e-05,
      "loss": 2.8291,
      "step": 4900
    },
    {
      "epoch": 1.89,
      "learning_rate": 3.8724404162470625e-05,
      "loss": 2.8866,
      "step": 5000
    },
    {
      "epoch": 1.89,
      "eval_loss": 4.007871150970459,
      "eval_runtime": 4.253,
      "eval_samples_per_second": 100.869,
      "eval_steps_per_second": 6.348,
      "step": 5000
    },
    {
      "epoch": 1.93,
      "learning_rate": 3.82544478012756e-05,
      "loss": 2.8411,
      "step": 5100
    },
    {
      "epoch": 1.96,
      "learning_rate": 3.778449144008056e-05,
      "loss": 2.8922,
      "step": 5200
    },
    {
      "epoch": 2.0,
      "learning_rate": 3.7314535078885526e-05,
      "loss": 2.8015,
      "step": 5300
    },
    {
      "epoch": 2.04,
      "learning_rate": 3.6844578717690504e-05,
      "loss": 2.591,
      "step": 5400
    },
    {
      "epoch": 2.08,
      "learning_rate": 3.637462235649547e-05,
      "loss": 2.6147,
      "step": 5500
    },
    {
      "epoch": 2.11,
      "learning_rate": 3.590466599530044e-05,
      "loss": 2.6515,
      "step": 5600
    },
    {
      "epoch": 2.15,
      "learning_rate": 3.5434709634105405e-05,
      "loss": 2.6206,
      "step": 5700
    },
    {
      "epoch": 2.19,
      "learning_rate": 3.496475327291037e-05,
      "loss": 2.5989,
      "step": 5800
    },
    {
      "epoch": 2.23,
      "learning_rate": 3.449479691171534e-05,
      "loss": 2.6168,
      "step": 5900
    },
    {
      "epoch": 2.27,
      "learning_rate": 3.4024840550520305e-05,
      "loss": 2.6146,
      "step": 6000
    },
    {
      "epoch": 2.27,
      "eval_loss": 4.015293598175049,
      "eval_runtime": 4.2602,
      "eval_samples_per_second": 100.7,
      "eval_steps_per_second": 6.338,
      "step": 6000
    },
    {
      "epoch": 2.3,
      "learning_rate": 3.3554884189325276e-05,
      "loss": 2.6718,
      "step": 6100
    },
    {
      "epoch": 2.34,
      "learning_rate": 3.308492782813025e-05,
      "loss": 2.6144,
      "step": 6200
    },
    {
      "epoch": 2.38,
      "learning_rate": 3.261497146693521e-05,
      "loss": 2.6326,
      "step": 6300
    },
    {
      "epoch": 2.42,
      "learning_rate": 3.2145015105740184e-05,
      "loss": 2.6431,
      "step": 6400
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.167505874454515e-05,
      "loss": 2.6734,
      "step": 6500
    },
    {
      "epoch": 2.49,
      "learning_rate": 3.120510238335011e-05,
      "loss": 2.6928,
      "step": 6600
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.0735146022155084e-05,
      "loss": 2.642,
      "step": 6700
    },
    {
      "epoch": 2.57,
      "learning_rate": 3.0265189660960056e-05,
      "loss": 2.6116,
      "step": 6800
    },
    {
      "epoch": 2.61,
      "learning_rate": 2.9795233299765024e-05,
      "loss": 2.663,
      "step": 6900
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.9325276938569988e-05,
      "loss": 2.6576,
      "step": 7000
    },
    {
      "epoch": 2.64,
      "eval_loss": 4.002060890197754,
      "eval_runtime": 4.2577,
      "eval_samples_per_second": 100.759,
      "eval_steps_per_second": 6.341,
      "step": 7000
    },
    {
      "epoch": 2.68,
      "learning_rate": 2.8855320577374956e-05,
      "loss": 2.6327,
      "step": 7100
    },
    {
      "epoch": 2.72,
      "learning_rate": 2.8385364216179924e-05,
      "loss": 2.6684,
      "step": 7200
    },
    {
      "epoch": 2.76,
      "learning_rate": 2.7915407854984896e-05,
      "loss": 2.6826,
      "step": 7300
    },
    {
      "epoch": 2.79,
      "learning_rate": 2.744545149378986e-05,
      "loss": 2.6216,
      "step": 7400
    },
    {
      "epoch": 2.83,
      "learning_rate": 2.6975495132594828e-05,
      "loss": 2.62,
      "step": 7500
    },
    {
      "epoch": 2.87,
      "learning_rate": 2.65055387713998e-05,
      "loss": 2.6401,
      "step": 7600
    },
    {
      "epoch": 2.91,
      "learning_rate": 2.6035582410204767e-05,
      "loss": 2.6276,
      "step": 7700
    },
    {
      "epoch": 2.95,
      "learning_rate": 2.5565626049009732e-05,
      "loss": 2.6809,
      "step": 7800
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.5095669687814703e-05,
      "loss": 2.6544,
      "step": 7900
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.462571332661967e-05,
      "loss": 2.5361,
      "step": 8000
    },
    {
      "epoch": 3.02,
      "eval_loss": 4.010578632354736,
      "eval_runtime": 4.2557,
      "eval_samples_per_second": 100.806,
      "eval_steps_per_second": 6.344,
      "step": 8000
    },
    {
      "epoch": 3.06,
      "learning_rate": 2.416045652903659e-05,
      "loss": 2.4391,
      "step": 8100
    },
    {
      "epoch": 3.1,
      "learning_rate": 2.3690500167841557e-05,
      "loss": 2.4457,
      "step": 8200
    },
    {
      "epoch": 3.13,
      "learning_rate": 2.3220543806646525e-05,
      "loss": 2.4267,
      "step": 8300
    },
    {
      "epoch": 3.17,
      "learning_rate": 2.2750587445451493e-05,
      "loss": 2.4724,
      "step": 8400
    },
    {
      "epoch": 3.21,
      "learning_rate": 2.228063108425646e-05,
      "loss": 2.4534,
      "step": 8500
    },
    {
      "epoch": 3.25,
      "learning_rate": 2.1810674723061432e-05,
      "loss": 2.446,
      "step": 8600
    },
    {
      "epoch": 3.29,
      "learning_rate": 2.1340718361866397e-05,
      "loss": 2.4415,
      "step": 8700
    },
    {
      "epoch": 3.32,
      "learning_rate": 2.0870762000671365e-05,
      "loss": 2.4938,
      "step": 8800
    },
    {
      "epoch": 3.36,
      "learning_rate": 2.0400805639476333e-05,
      "loss": 2.4444,
      "step": 8900
    },
    {
      "epoch": 3.4,
      "learning_rate": 1.9935548841893253e-05,
      "loss": 2.4552,
      "step": 9000
    },
    {
      "epoch": 3.4,
      "eval_loss": 4.0137410163879395,
      "eval_runtime": 4.2578,
      "eval_samples_per_second": 100.757,
      "eval_steps_per_second": 6.341,
      "step": 9000
    },
    {
      "epoch": 3.44,
      "learning_rate": 1.946559248069822e-05,
      "loss": 2.4298,
      "step": 9100
    },
    {
      "epoch": 3.47,
      "learning_rate": 1.899563611950319e-05,
      "loss": 2.4714,
      "step": 9200
    },
    {
      "epoch": 3.51,
      "learning_rate": 1.8525679758308157e-05,
      "loss": 2.4545,
      "step": 9300
    },
    {
      "epoch": 3.55,
      "learning_rate": 1.8055723397113125e-05,
      "loss": 2.4551,
      "step": 9400
    },
    {
      "epoch": 3.59,
      "learning_rate": 1.7585767035918093e-05,
      "loss": 2.4403,
      "step": 9500
    },
    {
      "epoch": 3.63,
      "learning_rate": 1.711581067472306e-05,
      "loss": 2.4297,
      "step": 9600
    },
    {
      "epoch": 3.66,
      "learning_rate": 1.664585431352803e-05,
      "loss": 2.4709,
      "step": 9700
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.6175897952332997e-05,
      "loss": 2.4781,
      "step": 9800
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.5705941591137965e-05,
      "loss": 2.4719,
      "step": 9900
    },
    {
      "epoch": 3.78,
      "learning_rate": 1.5235985229942935e-05,
      "loss": 2.4595,
      "step": 10000
    },
    {
      "epoch": 3.78,
      "eval_loss": 3.9967169761657715,
      "eval_runtime": 4.254,
      "eval_samples_per_second": 100.845,
      "eval_steps_per_second": 6.347,
      "step": 10000
    },
    {
      "epoch": 3.81,
      "learning_rate": 1.4766028868747901e-05,
      "loss": 2.4654,
      "step": 10100
    },
    {
      "epoch": 3.85,
      "learning_rate": 1.4296072507552869e-05,
      "loss": 2.4776,
      "step": 10200
    },
    {
      "epoch": 3.89,
      "learning_rate": 1.3826116146357837e-05,
      "loss": 2.4636,
      "step": 10300
    },
    {
      "epoch": 3.93,
      "learning_rate": 1.3356159785162807e-05,
      "loss": 2.45,
      "step": 10400
    },
    {
      "epoch": 3.97,
      "learning_rate": 1.2886203423967773e-05,
      "loss": 2.4622,
      "step": 10500
    },
    {
      "epoch": 4.0,
      "learning_rate": 1.2416247062772743e-05,
      "loss": 2.4645,
      "step": 10600
    },
    {
      "epoch": 4.04,
      "learning_rate": 1.194629070157771e-05,
      "loss": 2.3412,
      "step": 10700
    },
    {
      "epoch": 4.08,
      "learning_rate": 1.1476334340382679e-05,
      "loss": 2.3713,
      "step": 10800
    },
    {
      "epoch": 4.12,
      "learning_rate": 1.1006377979187647e-05,
      "loss": 2.3126,
      "step": 10900
    },
    {
      "epoch": 4.15,
      "learning_rate": 1.0536421617992615e-05,
      "loss": 2.3286,
      "step": 11000
    },
    {
      "epoch": 4.15,
      "eval_loss": 4.020402431488037,
      "eval_runtime": 4.2569,
      "eval_samples_per_second": 100.778,
      "eval_steps_per_second": 6.343,
      "step": 11000
    },
    {
      "epoch": 4.19,
      "learning_rate": 1.0066465256797583e-05,
      "loss": 2.3153,
      "step": 11100
    },
    {
      "epoch": 4.23,
      "learning_rate": 9.6012084592145e-06,
      "loss": 2.3465,
      "step": 11200
    },
    {
      "epoch": 4.27,
      "learning_rate": 9.13125209801947e-06,
      "loss": 2.3226,
      "step": 11300
    },
    {
      "epoch": 4.31,
      "learning_rate": 8.661295736824438e-06,
      "loss": 2.3332,
      "step": 11400
    },
    {
      "epoch": 4.34,
      "learning_rate": 8.191339375629406e-06,
      "loss": 2.2948,
      "step": 11500
    },
    {
      "epoch": 4.38,
      "learning_rate": 7.721383014434374e-06,
      "loss": 2.2821,
      "step": 11600
    },
    {
      "epoch": 4.42,
      "learning_rate": 7.251426653239342e-06,
      "loss": 2.3112,
      "step": 11700
    },
    {
      "epoch": 4.46,
      "learning_rate": 6.7814702920443104e-06,
      "loss": 2.2696,
      "step": 11800
    },
    {
      "epoch": 4.49,
      "learning_rate": 6.311513930849278e-06,
      "loss": 2.3236,
      "step": 11900
    },
    {
      "epoch": 4.53,
      "learning_rate": 5.841557569654246e-06,
      "loss": 2.2805,
      "step": 12000
    },
    {
      "epoch": 4.53,
      "eval_loss": 4.015416622161865,
      "eval_runtime": 4.2597,
      "eval_samples_per_second": 100.711,
      "eval_steps_per_second": 6.338,
      "step": 12000
    },
    {
      "epoch": 4.57,
      "learning_rate": 5.371601208459214e-06,
      "loss": 2.3157,
      "step": 12100
    },
    {
      "epoch": 4.61,
      "learning_rate": 4.901644847264182e-06,
      "loss": 2.3227,
      "step": 12200
    },
    {
      "epoch": 4.65,
      "learning_rate": 4.43168848606915e-06,
      "loss": 2.3303,
      "step": 12300
    },
    {
      "epoch": 4.68,
      "learning_rate": 3.961732124874118e-06,
      "loss": 2.3228,
      "step": 12400
    },
    {
      "epoch": 4.72,
      "learning_rate": 3.491775763679087e-06,
      "loss": 2.3152,
      "step": 12500
    },
    {
      "epoch": 4.76,
      "learning_rate": 3.021819402484055e-06,
      "loss": 2.3273,
      "step": 12600
    },
    {
      "epoch": 4.8,
      "learning_rate": 2.551863041289023e-06,
      "loss": 2.3569,
      "step": 12700
    },
    {
      "epoch": 4.83,
      "learning_rate": 2.0819066800939914e-06,
      "loss": 2.3205,
      "step": 12800
    },
    {
      "epoch": 4.87,
      "learning_rate": 1.6119503188989594e-06,
      "loss": 2.2739,
      "step": 12900
    },
    {
      "epoch": 4.91,
      "learning_rate": 1.1419939577039274e-06,
      "loss": 2.3098,
      "step": 13000
    },
    {
      "epoch": 4.91,
      "eval_loss": 4.012601852416992,
      "eval_runtime": 4.2579,
      "eval_samples_per_second": 100.754,
      "eval_steps_per_second": 6.341,
      "step": 13000
    },
    {
      "epoch": 4.95,
      "learning_rate": 6.720375965088956e-07,
      "loss": 2.2791,
      "step": 13100
    },
    {
      "epoch": 4.98,
      "learning_rate": 2.0208123531386372e-07,
      "loss": 2.2955,
      "step": 13200
    },
    {
      "epoch": 5.0,
      "step": 13240,
      "total_flos": 5.535079759872e+16,
      "train_loss": 2.6959276412548974,
      "train_runtime": 6590.4176,
      "train_samples_per_second": 32.143,
      "train_steps_per_second": 2.009
    }
  ],
  "max_steps": 13240,
  "num_train_epochs": 5,
  "total_flos": 5.535079759872e+16,
  "trial_name": null,
  "trial_params": null
}