{
  "best_metric": 0.8088124957369893,
  "best_model_checkpoint": "outputs/t5-mini/weak_tiny_poe/qqp_42/checkpoint-56855",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 56855,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "grad_norm": 2.30487060546875,
      "learning_rate": 4.956028493536189e-05,
      "loss": 0.7659,
      "step": 500
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.7330808639526367,
      "learning_rate": 4.9120569870723775e-05,
      "loss": 0.6875,
      "step": 1000
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.5595574378967285,
      "learning_rate": 4.868085480608566e-05,
      "loss": 0.6526,
      "step": 1500
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.4334590435028076,
      "learning_rate": 4.824113974144755e-05,
      "loss": 0.6392,
      "step": 2000
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.1763668060302734,
      "learning_rate": 4.780142467680943e-05,
      "loss": 0.6377,
      "step": 2500
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.2207213640213013,
      "learning_rate": 4.736170961217132e-05,
      "loss": 0.6186,
      "step": 3000
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.3439034223556519,
      "learning_rate": 4.6921994547533196e-05,
      "loss": 0.6166,
      "step": 3500
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.9837682247161865,
      "learning_rate": 4.6482279482895086e-05,
      "loss": 0.6171,
      "step": 4000
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.070117473602295,
      "learning_rate": 4.6042564418256975e-05,
      "loss": 0.602,
      "step": 4500
    },
    {
      "epoch": 0.44,
      "grad_norm": 2.691117525100708,
      "learning_rate": 4.560284935361886e-05,
      "loss": 0.5938,
      "step": 5000
    },
    {
      "epoch": 0.48,
      "grad_norm": 1.9327023029327393,
      "learning_rate": 4.516313428898074e-05,
      "loss": 0.5996,
      "step": 5500
    },
    {
      "epoch": 0.53,
      "grad_norm": 2.1416549682617188,
      "learning_rate": 4.472341922434263e-05,
      "loss": 0.5902,
      "step": 6000
    },
    {
      "epoch": 0.57,
      "grad_norm": 2.225383996963501,
      "learning_rate": 4.428370415970451e-05,
      "loss": 0.584,
      "step": 6500
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.4455324411392212,
      "learning_rate": 4.38439890950664e-05,
      "loss": 0.5719,
      "step": 7000
    },
    {
      "epoch": 0.66,
      "grad_norm": 2.7837624549865723,
      "learning_rate": 4.3404274030428286e-05,
      "loss": 0.5762,
      "step": 7500
    },
    {
      "epoch": 0.7,
      "grad_norm": 2.026057004928589,
      "learning_rate": 4.296455896579017e-05,
      "loss": 0.5707,
      "step": 8000
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.983589768409729,
      "learning_rate": 4.252484390115206e-05,
      "loss": 0.5651,
      "step": 8500
    },
    {
      "epoch": 0.79,
      "grad_norm": 1.826207160949707,
      "learning_rate": 4.208512883651394e-05,
      "loss": 0.5593,
      "step": 9000
    },
    {
      "epoch": 0.84,
      "grad_norm": 2.2966904640197754,
      "learning_rate": 4.1645413771875824e-05,
      "loss": 0.5658,
      "step": 9500
    },
    {
      "epoch": 0.88,
      "grad_norm": 2.696525812149048,
      "learning_rate": 4.1205698707237714e-05,
      "loss": 0.5664,
      "step": 10000
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.6921826601028442,
      "learning_rate": 4.0765983642599596e-05,
      "loss": 0.562,
      "step": 10500
    },
    {
      "epoch": 0.97,
      "grad_norm": 1.9230735301971436,
      "learning_rate": 4.0326268577961486e-05,
      "loss": 0.562,
      "step": 11000
    },
    {
      "epoch": 1.0,
      "eval_combined_score": 0.7611372754229897,
      "eval_f1": 0.7611372754229897,
      "eval_loss": 0.37701112031936646,
      "eval_runtime": 36.8648,
      "eval_samples_per_second": 1096.71,
      "eval_steps_per_second": 4.286,
      "step": 11371
    },
    {
      "epoch": 1.01,
      "grad_norm": 2.1575307846069336,
      "learning_rate": 3.988655351332337e-05,
      "loss": 0.5502,
      "step": 11500
    },
    {
      "epoch": 1.06,
      "grad_norm": 2.0553629398345947,
      "learning_rate": 3.944683844868525e-05,
      "loss": 0.5486,
      "step": 12000
    },
    {
      "epoch": 1.1,
      "grad_norm": 2.253108024597168,
      "learning_rate": 3.900712338404714e-05,
      "loss": 0.5457,
      "step": 12500
    },
    {
      "epoch": 1.14,
      "grad_norm": 2.2528011798858643,
      "learning_rate": 3.8567408319409024e-05,
      "loss": 0.5418,
      "step": 13000
    },
    {
      "epoch": 1.19,
      "grad_norm": 1.872086524963379,
      "learning_rate": 3.812769325477091e-05,
      "loss": 0.5405,
      "step": 13500
    },
    {
      "epoch": 1.23,
      "grad_norm": 1.549558401107788,
      "learning_rate": 3.76879781901328e-05,
      "loss": 0.5432,
      "step": 14000
    },
    {
      "epoch": 1.28,
      "grad_norm": 2.1427700519561768,
      "learning_rate": 3.724826312549468e-05,
      "loss": 0.5363,
      "step": 14500
    },
    {
      "epoch": 1.32,
      "grad_norm": 2.0889768600463867,
      "learning_rate": 3.680854806085657e-05,
      "loss": 0.5401,
      "step": 15000
    },
    {
      "epoch": 1.36,
      "grad_norm": 1.9004706144332886,
      "learning_rate": 3.636883299621845e-05,
      "loss": 0.538,
      "step": 15500
    },
    {
      "epoch": 1.41,
      "grad_norm": 2.568284273147583,
      "learning_rate": 3.5929117931580335e-05,
      "loss": 0.5364,
      "step": 16000
    },
    {
      "epoch": 1.45,
      "grad_norm": 1.1210036277770996,
      "learning_rate": 3.5489402866942224e-05,
      "loss": 0.5278,
      "step": 16500
    },
    {
      "epoch": 1.5,
      "grad_norm": 2.536067247390747,
      "learning_rate": 3.504968780230411e-05,
      "loss": 0.5338,
      "step": 17000
    },
    {
      "epoch": 1.54,
      "grad_norm": 2.0306429862976074,
      "learning_rate": 3.4609972737666e-05,
      "loss": 0.5238,
      "step": 17500
    },
    {
      "epoch": 1.58,
      "grad_norm": 2.097214460372925,
      "learning_rate": 3.417025767302788e-05,
      "loss": 0.5273,
      "step": 18000
    },
    {
      "epoch": 1.63,
      "grad_norm": 2.1282901763916016,
      "learning_rate": 3.373054260838976e-05,
      "loss": 0.5249,
      "step": 18500
    },
    {
      "epoch": 1.67,
      "grad_norm": 2.8680260181427,
      "learning_rate": 3.329082754375165e-05,
      "loss": 0.533,
      "step": 19000
    },
    {
      "epoch": 1.71,
      "grad_norm": 2.5516855716705322,
      "learning_rate": 3.2851112479113535e-05,
      "loss": 0.5254,
      "step": 19500
    },
    {
      "epoch": 1.76,
      "grad_norm": 2.8096489906311035,
      "learning_rate": 3.241139741447542e-05,
      "loss": 0.5282,
      "step": 20000
    },
    {
      "epoch": 1.8,
      "grad_norm": 3.4034230709075928,
      "learning_rate": 3.197168234983731e-05,
      "loss": 0.5219,
      "step": 20500
    },
    {
      "epoch": 1.85,
      "grad_norm": 2.213874340057373,
      "learning_rate": 3.153196728519919e-05,
      "loss": 0.5277,
      "step": 21000
    },
    {
      "epoch": 1.89,
      "grad_norm": 2.4699976444244385,
      "learning_rate": 3.109225222056108e-05,
      "loss": 0.5261,
      "step": 21500
    },
    {
      "epoch": 1.93,
      "grad_norm": 1.352188229560852,
      "learning_rate": 3.065253715592296e-05,
      "loss": 0.5267,
      "step": 22000
    },
    {
      "epoch": 1.98,
      "grad_norm": 2.8092727661132812,
      "learning_rate": 3.021282209128485e-05,
      "loss": 0.5282,
      "step": 22500
    },
    {
      "epoch": 2.0,
      "eval_combined_score": 0.7765263781861292,
      "eval_f1": 0.7765263781861292,
      "eval_loss": 0.3535827398300171,
      "eval_runtime": 37.1503,
      "eval_samples_per_second": 1088.283,
      "eval_steps_per_second": 4.253,
      "step": 22742
    },
    {
      "epoch": 2.02,
      "grad_norm": 2.087219476699829,
      "learning_rate": 2.9773107026646735e-05,
      "loss": 0.5222,
      "step": 23000
    },
    {
      "epoch": 2.07,
      "grad_norm": 3.6042699813842773,
      "learning_rate": 2.933339196200862e-05,
      "loss": 0.5149,
      "step": 23500
    },
    {
      "epoch": 2.11,
      "grad_norm": 2.8945512771606445,
      "learning_rate": 2.8893676897370504e-05,
      "loss": 0.5143,
      "step": 24000
    },
    {
      "epoch": 2.15,
      "grad_norm": 3.639920234680176,
      "learning_rate": 2.845396183273239e-05,
      "loss": 0.5142,
      "step": 24500
    },
    {
      "epoch": 2.2,
      "grad_norm": 1.5696040391921997,
      "learning_rate": 2.8014246768094277e-05,
      "loss": 0.5109,
      "step": 25000
    },
    {
      "epoch": 2.24,
      "grad_norm": 2.502941608428955,
      "learning_rate": 2.7574531703456163e-05,
      "loss": 0.5041,
      "step": 25500
    },
    {
      "epoch": 2.29,
      "grad_norm": 1.7532044649124146,
      "learning_rate": 2.7134816638818046e-05,
      "loss": 0.5099,
      "step": 26000
    },
    {
      "epoch": 2.33,
      "grad_norm": 2.583704948425293,
      "learning_rate": 2.6695101574179932e-05,
      "loss": 0.5109,
      "step": 26500
    },
    {
      "epoch": 2.37,
      "grad_norm": 1.9148293733596802,
      "learning_rate": 2.625538650954182e-05,
      "loss": 0.5088,
      "step": 27000
    },
    {
      "epoch": 2.42,
      "grad_norm": 1.2226241827011108,
      "learning_rate": 2.5815671444903705e-05,
      "loss": 0.5046,
      "step": 27500
    },
    {
      "epoch": 2.46,
      "grad_norm": 2.276338815689087,
      "learning_rate": 2.5375956380265588e-05,
      "loss": 0.5077,
      "step": 28000
    },
    {
      "epoch": 2.51,
      "grad_norm": 1.4178540706634521,
      "learning_rate": 2.4936241315627474e-05,
      "loss": 0.4967,
      "step": 28500
    },
    {
      "epoch": 2.55,
      "grad_norm": 2.231714963912964,
      "learning_rate": 2.449652625098936e-05,
      "loss": 0.5047,
      "step": 29000
    },
    {
      "epoch": 2.59,
      "grad_norm": 1.2348312139511108,
      "learning_rate": 2.4056811186351246e-05,
      "loss": 0.5072,
      "step": 29500
    },
    {
      "epoch": 2.64,
      "grad_norm": 1.837050199508667,
      "learning_rate": 2.3617096121713133e-05,
      "loss": 0.5045,
      "step": 30000
    },
    {
      "epoch": 2.68,
      "grad_norm": 2.297529697418213,
      "learning_rate": 2.317738105707502e-05,
      "loss": 0.5102,
      "step": 30500
    },
    {
      "epoch": 2.73,
      "grad_norm": 1.2097573280334473,
      "learning_rate": 2.27376659924369e-05,
      "loss": 0.5046,
      "step": 31000
    },
    {
      "epoch": 2.77,
      "grad_norm": 2.4351987838745117,
      "learning_rate": 2.2297950927798788e-05,
      "loss": 0.5057,
      "step": 31500
    },
    {
      "epoch": 2.81,
      "grad_norm": 2.6174986362457275,
      "learning_rate": 2.1858235863160674e-05,
      "loss": 0.5051,
      "step": 32000
    },
    {
      "epoch": 2.86,
      "grad_norm": 2.1429812908172607,
      "learning_rate": 2.141852079852256e-05,
      "loss": 0.496,
      "step": 32500
    },
    {
      "epoch": 2.9,
      "grad_norm": 2.406395673751831,
      "learning_rate": 2.0978805733884443e-05,
      "loss": 0.492,
      "step": 33000
    },
    {
      "epoch": 2.95,
      "grad_norm": 2.600222587585449,
      "learning_rate": 2.053909066924633e-05,
      "loss": 0.5063,
      "step": 33500
    },
    {
      "epoch": 2.99,
      "grad_norm": 1.4866433143615723,
      "learning_rate": 2.0099375604608216e-05,
      "loss": 0.4977,
      "step": 34000
    },
    {
      "epoch": 3.0,
      "eval_combined_score": 0.7951059796656902,
      "eval_f1": 0.7951059796656902,
      "eval_loss": 0.3372982144355774,
      "eval_runtime": 36.9034,
      "eval_samples_per_second": 1095.563,
      "eval_steps_per_second": 4.281,
      "step": 34113
    },
    {
      "epoch": 3.03,
      "grad_norm": 1.615599513053894,
      "learning_rate": 1.9659660539970102e-05,
      "loss": 0.4991,
      "step": 34500
    },
    {
      "epoch": 3.08,
      "grad_norm": 2.992169141769409,
      "learning_rate": 1.9219945475331985e-05,
      "loss": 0.4927,
      "step": 35000
    },
    {
      "epoch": 3.12,
      "grad_norm": 2.639941930770874,
      "learning_rate": 1.878023041069387e-05,
      "loss": 0.4933,
      "step": 35500
    },
    {
      "epoch": 3.17,
      "grad_norm": 2.457653760910034,
      "learning_rate": 1.8340515346055757e-05,
      "loss": 0.4955,
      "step": 36000
    },
    {
      "epoch": 3.21,
      "grad_norm": 2.2147014141082764,
      "learning_rate": 1.7900800281417643e-05,
      "loss": 0.4958,
      "step": 36500
    },
    {
      "epoch": 3.25,
      "grad_norm": 3.189746618270874,
      "learning_rate": 1.7461085216779526e-05,
      "loss": 0.4915,
      "step": 37000
    },
    {
      "epoch": 3.3,
      "grad_norm": 1.946791172027588,
      "learning_rate": 1.7021370152141413e-05,
      "loss": 0.492,
      "step": 37500
    },
    {
      "epoch": 3.34,
      "grad_norm": 1.1339478492736816,
      "learning_rate": 1.65816550875033e-05,
      "loss": 0.4889,
      "step": 38000
    },
    {
      "epoch": 3.39,
      "grad_norm": 2.4532470703125,
      "learning_rate": 1.6141940022865185e-05,
      "loss": 0.4851,
      "step": 38500
    },
    {
      "epoch": 3.43,
      "grad_norm": 1.2711316347122192,
      "learning_rate": 1.5702224958227068e-05,
      "loss": 0.5004,
      "step": 39000
    },
    {
      "epoch": 3.47,
      "grad_norm": 1.6113109588623047,
      "learning_rate": 1.5262509893588954e-05,
      "loss": 0.4829,
      "step": 39500
    },
    {
      "epoch": 3.52,
      "grad_norm": 2.32248592376709,
      "learning_rate": 1.482279482895084e-05,
      "loss": 0.4886,
      "step": 40000
    },
    {
      "epoch": 3.56,
      "grad_norm": 2.664520025253296,
      "learning_rate": 1.4383079764312727e-05,
      "loss": 0.4901,
      "step": 40500
    },
    {
      "epoch": 3.61,
      "grad_norm": 3.6276190280914307,
      "learning_rate": 1.3943364699674611e-05,
      "loss": 0.4906,
      "step": 41000
    },
    {
      "epoch": 3.65,
      "grad_norm": 2.5009617805480957,
      "learning_rate": 1.3503649635036497e-05,
      "loss": 0.4888,
      "step": 41500
    },
    {
      "epoch": 3.69,
      "grad_norm": 2.4080982208251953,
      "learning_rate": 1.3063934570398382e-05,
      "loss": 0.4841,
      "step": 42000
    },
    {
      "epoch": 3.74,
      "grad_norm": 2.2229392528533936,
      "learning_rate": 1.2624219505760268e-05,
      "loss": 0.4934,
      "step": 42500
    },
    {
      "epoch": 3.78,
      "grad_norm": 1.7180906534194946,
      "learning_rate": 1.2184504441122153e-05,
      "loss": 0.4864,
      "step": 43000
    },
    {
      "epoch": 3.83,
      "grad_norm": 2.0107908248901367,
      "learning_rate": 1.1744789376484039e-05,
      "loss": 0.4888,
      "step": 43500
    },
    {
      "epoch": 3.87,
      "grad_norm": 2.2142484188079834,
      "learning_rate": 1.1305074311845925e-05,
      "loss": 0.4854,
      "step": 44000
    },
    {
      "epoch": 3.91,
      "grad_norm": 2.1998207569122314,
      "learning_rate": 1.086535924720781e-05,
      "loss": 0.4947,
      "step": 44500
    },
    {
      "epoch": 3.96,
      "grad_norm": 1.7702863216400146,
      "learning_rate": 1.0425644182569696e-05,
      "loss": 0.4871,
      "step": 45000
    },
    {
      "epoch": 4.0,
      "eval_combined_score": 0.8021821467041941,
      "eval_f1": 0.8021821467041941,
      "eval_loss": 0.32757022976875305,
      "eval_runtime": 37.0242,
      "eval_samples_per_second": 1091.988,
      "eval_steps_per_second": 4.267,
      "step": 45484
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.43105936050415,
      "learning_rate": 9.98592911793158e-06,
      "loss": 0.486,
      "step": 45500
    },
    {
      "epoch": 4.05,
      "grad_norm": 2.509622812271118,
      "learning_rate": 9.546214053293467e-06,
      "loss": 0.4838,
      "step": 46000
    },
    {
      "epoch": 4.09,
      "grad_norm": 1.9040151834487915,
      "learning_rate": 9.106498988655351e-06,
      "loss": 0.4848,
      "step": 46500
    },
    {
      "epoch": 4.13,
      "grad_norm": 1.892439365386963,
      "learning_rate": 8.666783924017237e-06,
      "loss": 0.4808,
      "step": 47000
    },
    {
      "epoch": 4.18,
      "grad_norm": 1.889847755432129,
      "learning_rate": 8.227068859379122e-06,
      "loss": 0.4803,
      "step": 47500
    },
    {
      "epoch": 4.22,
      "grad_norm": 2.515573740005493,
      "learning_rate": 7.787353794741008e-06,
      "loss": 0.4749,
      "step": 48000
    },
    {
      "epoch": 4.27,
      "grad_norm": 2.6856706142425537,
      "learning_rate": 7.347638730102893e-06,
      "loss": 0.4854,
      "step": 48500
    },
    {
      "epoch": 4.31,
      "grad_norm": 2.979976177215576,
      "learning_rate": 6.907923665464779e-06,
      "loss": 0.4858,
      "step": 49000
    },
    {
      "epoch": 4.35,
      "grad_norm": 2.1129562854766846,
      "learning_rate": 6.468208600826664e-06,
      "loss": 0.4863,
      "step": 49500
    },
    {
      "epoch": 4.4,
      "grad_norm": 2.569387435913086,
      "learning_rate": 6.02849353618855e-06,
      "loss": 0.4777,
      "step": 50000
    },
    {
      "epoch": 4.44,
      "grad_norm": 2.261967420578003,
      "learning_rate": 5.588778471550435e-06,
      "loss": 0.4726,
      "step": 50500
    },
    {
      "epoch": 4.49,
      "grad_norm": 2.604426145553589,
      "learning_rate": 5.149063406912321e-06,
      "loss": 0.4806,
      "step": 51000
    },
    {
      "epoch": 4.53,
      "grad_norm": 1.9708985090255737,
      "learning_rate": 4.709348342274207e-06,
      "loss": 0.4871,
      "step": 51500
    },
    {
      "epoch": 4.57,
      "grad_norm": 3.4044077396392822,
      "learning_rate": 4.269633277636092e-06,
      "loss": 0.4777,
      "step": 52000
    },
    {
      "epoch": 4.62,
      "grad_norm": 3.580048084259033,
      "learning_rate": 3.8299182129979776e-06,
      "loss": 0.4911,
      "step": 52500
    },
    {
      "epoch": 4.66,
      "grad_norm": 2.0711894035339355,
      "learning_rate": 3.3902031483598625e-06,
      "loss": 0.4789,
      "step": 53000
    },
    {
      "epoch": 4.7,
      "grad_norm": 3.7427501678466797,
      "learning_rate": 2.9504880837217483e-06,
      "loss": 0.48,
      "step": 53500
    },
    {
      "epoch": 4.75,
      "grad_norm": 1.780761957168579,
      "learning_rate": 2.5107730190836337e-06,
      "loss": 0.4821,
      "step": 54000
    },
    {
      "epoch": 4.79,
      "grad_norm": 2.185481309890747,
      "learning_rate": 2.0710579544455195e-06,
      "loss": 0.4874,
      "step": 54500
    },
    {
      "epoch": 4.84,
      "grad_norm": 2.1227777004241943,
      "learning_rate": 1.6313428898074047e-06,
      "loss": 0.4812,
      "step": 55000
    },
    {
      "epoch": 4.88,
      "grad_norm": 1.4683480262756348,
      "learning_rate": 1.1916278251692903e-06,
      "loss": 0.4785,
      "step": 55500
    },
    {
      "epoch": 4.92,
      "grad_norm": 1.6182867288589478,
      "learning_rate": 7.519127605311758e-07,
      "loss": 0.4834,
      "step": 56000
    },
    {
      "epoch": 4.97,
      "grad_norm": 2.2771718502044678,
      "learning_rate": 3.121976958930613e-07,
      "loss": 0.478,
      "step": 56500
    },
    {
      "epoch": 5.0,
      "eval_combined_score": 0.8088124957369893,
      "eval_f1": 0.8088124957369893,
      "eval_loss": 0.31970837712287903,
      "eval_runtime": 36.9254,
      "eval_samples_per_second": 1094.91,
      "eval_steps_per_second": 4.279,
      "step": 56855
    },
    {
      "epoch": 5.0,
      "step": 56855,
      "total_flos": 0.0,
      "train_loss": 0.5228507439386308,
      "train_runtime": 6326.7637,
      "train_samples_per_second": 287.545,
      "train_steps_per_second": 8.986
    }
  ],
  "logging_steps": 500,
  "max_steps": 56855,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 0.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}