{
  "best_metric": 0.8175030270415714,
  "best_model_checkpoint": "outputs/t5-mini/vanilla/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": 2.7941341400146484,
      "learning_rate": 4.956028493536189e-05,
      "loss": 0.6002,
      "step": 500
    },
    {
      "epoch": 0.09,
      "grad_norm": 1.659162998199463,
      "learning_rate": 4.9120569870723775e-05,
      "loss": 0.5049,
      "step": 1000
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.7229186296463013,
      "learning_rate": 4.868085480608566e-05,
      "loss": 0.4735,
      "step": 1500
    },
    {
      "epoch": 0.18,
      "grad_norm": 1.7305272817611694,
      "learning_rate": 4.824113974144755e-05,
      "loss": 0.4597,
      "step": 2000
    },
    {
      "epoch": 0.22,
      "grad_norm": 1.3873671293258667,
      "learning_rate": 4.780142467680943e-05,
      "loss": 0.4556,
      "step": 2500
    },
    {
      "epoch": 0.26,
      "grad_norm": 1.64267098903656,
      "learning_rate": 4.736170961217132e-05,
      "loss": 0.4366,
      "step": 3000
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.8006480932235718,
      "learning_rate": 4.6921994547533196e-05,
      "loss": 0.432,
      "step": 3500
    },
    {
      "epoch": 0.35,
      "grad_norm": 2.2416460514068604,
      "learning_rate": 4.6482279482895086e-05,
      "loss": 0.4306,
      "step": 4000
    },
    {
      "epoch": 0.4,
      "grad_norm": 2.0534579753875732,
      "learning_rate": 4.6042564418256975e-05,
      "loss": 0.4151,
      "step": 4500
    },
    {
      "epoch": 0.44,
      "grad_norm": 2.3821334838867188,
      "learning_rate": 4.560284935361886e-05,
      "loss": 0.411,
      "step": 5000
    },
    {
      "epoch": 0.48,
      "grad_norm": 2.1282753944396973,
      "learning_rate": 4.516313428898074e-05,
      "loss": 0.4158,
      "step": 5500
    },
    {
      "epoch": 0.53,
      "grad_norm": 1.8800193071365356,
      "learning_rate": 4.472341922434263e-05,
      "loss": 0.4084,
      "step": 6000
    },
    {
      "epoch": 0.57,
      "grad_norm": 2.5685102939605713,
      "learning_rate": 4.428370415970451e-05,
      "loss": 0.4008,
      "step": 6500
    },
    {
      "epoch": 0.62,
      "grad_norm": 1.6782379150390625,
      "learning_rate": 4.38439890950664e-05,
      "loss": 0.3923,
      "step": 7000
    },
    {
      "epoch": 0.66,
      "grad_norm": 3.4463438987731934,
      "learning_rate": 4.3404274030428286e-05,
      "loss": 0.3889,
      "step": 7500
    },
    {
      "epoch": 0.7,
      "grad_norm": 1.8108657598495483,
      "learning_rate": 4.296455896579017e-05,
      "loss": 0.3879,
      "step": 8000
    },
    {
      "epoch": 0.75,
      "grad_norm": 1.909293532371521,
      "learning_rate": 4.252484390115206e-05,
      "loss": 0.3844,
      "step": 8500
    },
    {
      "epoch": 0.79,
      "grad_norm": 2.362368106842041,
      "learning_rate": 4.208512883651394e-05,
      "loss": 0.378,
      "step": 9000
    },
    {
      "epoch": 0.84,
      "grad_norm": 2.822340250015259,
      "learning_rate": 4.1645413771875824e-05,
      "loss": 0.3804,
      "step": 9500
    },
    {
      "epoch": 0.88,
      "grad_norm": 2.626854658126831,
      "learning_rate": 4.1205698707237714e-05,
      "loss": 0.3795,
      "step": 10000
    },
    {
      "epoch": 0.92,
      "grad_norm": 1.7211793661117554,
      "learning_rate": 4.0765983642599596e-05,
      "loss": 0.3743,
      "step": 10500
    },
    {
      "epoch": 0.97,
      "grad_norm": 2.505765199661255,
      "learning_rate": 4.0326268577961486e-05,
      "loss": 0.3784,
      "step": 11000
    },
    {
      "epoch": 1.0,
      "eval_combined_score": 0.7833636180092873,
      "eval_f1": 0.7833636180092873,
      "eval_loss": 0.3582809865474701,
      "eval_runtime": 32.9779,
      "eval_samples_per_second": 1225.973,
      "eval_steps_per_second": 4.791,
      "step": 11371
    },
    {
      "epoch": 1.01,
      "grad_norm": 1.8606353998184204,
      "learning_rate": 3.988655351332337e-05,
      "loss": 0.3648,
      "step": 11500
    },
    {
      "epoch": 1.06,
      "grad_norm": 2.0803306102752686,
      "learning_rate": 3.944683844868525e-05,
      "loss": 0.3627,
      "step": 12000
    },
    {
      "epoch": 1.1,
      "grad_norm": 2.276549816131592,
      "learning_rate": 3.900712338404714e-05,
      "loss": 0.3618,
      "step": 12500
    },
    {
      "epoch": 1.14,
      "grad_norm": 2.1461503505706787,
      "learning_rate": 3.8567408319409024e-05,
      "loss": 0.3604,
      "step": 13000
    },
    {
      "epoch": 1.19,
      "grad_norm": 1.5309439897537231,
      "learning_rate": 3.812769325477091e-05,
      "loss": 0.3529,
      "step": 13500
    },
    {
      "epoch": 1.23,
      "grad_norm": 1.354574203491211,
      "learning_rate": 3.76879781901328e-05,
      "loss": 0.3565,
      "step": 14000
    },
    {
      "epoch": 1.28,
      "grad_norm": 1.4756726026535034,
      "learning_rate": 3.724826312549468e-05,
      "loss": 0.3502,
      "step": 14500
    },
    {
      "epoch": 1.32,
      "grad_norm": 2.2074689865112305,
      "learning_rate": 3.680854806085657e-05,
      "loss": 0.3525,
      "step": 15000
    },
    {
      "epoch": 1.36,
      "grad_norm": 2.073932409286499,
      "learning_rate": 3.636883299621845e-05,
      "loss": 0.3555,
      "step": 15500
    },
    {
      "epoch": 1.41,
      "grad_norm": 2.54852032661438,
      "learning_rate": 3.5929117931580335e-05,
      "loss": 0.3495,
      "step": 16000
    },
    {
      "epoch": 1.45,
      "grad_norm": 1.8127062320709229,
      "learning_rate": 3.5489402866942224e-05,
      "loss": 0.3446,
      "step": 16500
    },
    {
      "epoch": 1.5,
      "grad_norm": 3.1064982414245605,
      "learning_rate": 3.504968780230411e-05,
      "loss": 0.3493,
      "step": 17000
    },
    {
      "epoch": 1.54,
      "grad_norm": 1.1443413496017456,
      "learning_rate": 3.4609972737666e-05,
      "loss": 0.3382,
      "step": 17500
    },
    {
      "epoch": 1.58,
      "grad_norm": 3.351428747177124,
      "learning_rate": 3.417025767302788e-05,
      "loss": 0.3466,
      "step": 18000
    },
    {
      "epoch": 1.63,
      "grad_norm": 2.2626051902770996,
      "learning_rate": 3.373054260838976e-05,
      "loss": 0.3411,
      "step": 18500
    },
    {
      "epoch": 1.67,
      "grad_norm": 2.5444693565368652,
      "learning_rate": 3.329082754375165e-05,
      "loss": 0.3488,
      "step": 19000
    },
    {
      "epoch": 1.71,
      "grad_norm": 2.525617837905884,
      "learning_rate": 3.2851112479113535e-05,
      "loss": 0.3351,
      "step": 19500
    },
    {
      "epoch": 1.76,
      "grad_norm": 2.9027602672576904,
      "learning_rate": 3.241139741447542e-05,
      "loss": 0.3463,
      "step": 20000
    },
    {
      "epoch": 1.8,
      "grad_norm": 2.7901854515075684,
      "learning_rate": 3.197168234983731e-05,
      "loss": 0.338,
      "step": 20500
    },
    {
      "epoch": 1.85,
      "grad_norm": 1.8225127458572388,
      "learning_rate": 3.153196728519919e-05,
      "loss": 0.3412,
      "step": 21000
    },
    {
      "epoch": 1.89,
      "grad_norm": 3.1016995906829834,
      "learning_rate": 3.109225222056108e-05,
      "loss": 0.3453,
      "step": 21500
    },
    {
      "epoch": 1.93,
      "grad_norm": 1.6413934230804443,
      "learning_rate": 3.065253715592296e-05,
      "loss": 0.345,
      "step": 22000
    },
    {
      "epoch": 1.98,
      "grad_norm": 1.8451814651489258,
      "learning_rate": 3.021282209128485e-05,
      "loss": 0.3448,
      "step": 22500
    },
    {
      "epoch": 2.0,
      "eval_combined_score": 0.8018207986076246,
      "eval_f1": 0.8018207986076246,
      "eval_loss": 0.33219704031944275,
      "eval_runtime": 32.33,
      "eval_samples_per_second": 1250.541,
      "eval_steps_per_second": 4.887,
      "step": 22742
    },
    {
      "epoch": 2.02,
      "grad_norm": 2.0090043544769287,
      "learning_rate": 2.9773107026646735e-05,
      "loss": 0.3303,
      "step": 23000
    },
    {
      "epoch": 2.07,
      "grad_norm": 2.7480595111846924,
      "learning_rate": 2.933339196200862e-05,
      "loss": 0.3263,
      "step": 23500
    },
    {
      "epoch": 2.11,
      "grad_norm": 2.6819705963134766,
      "learning_rate": 2.8893676897370504e-05,
      "loss": 0.3335,
      "step": 24000
    },
    {
      "epoch": 2.15,
      "grad_norm": 2.356091260910034,
      "learning_rate": 2.845396183273239e-05,
      "loss": 0.3279,
      "step": 24500
    },
    {
      "epoch": 2.2,
      "grad_norm": 1.7097245454788208,
      "learning_rate": 2.8014246768094277e-05,
      "loss": 0.3342,
      "step": 25000
    },
    {
      "epoch": 2.24,
      "grad_norm": 2.1123530864715576,
      "learning_rate": 2.7574531703456163e-05,
      "loss": 0.3278,
      "step": 25500
    },
    {
      "epoch": 2.29,
      "grad_norm": 1.814635992050171,
      "learning_rate": 2.7134816638818046e-05,
      "loss": 0.3299,
      "step": 26000
    },
    {
      "epoch": 2.33,
      "grad_norm": 2.7089486122131348,
      "learning_rate": 2.6695101574179932e-05,
      "loss": 0.3255,
      "step": 26500
    },
    {
      "epoch": 2.37,
      "grad_norm": 1.7767661809921265,
      "learning_rate": 2.625538650954182e-05,
      "loss": 0.3222,
      "step": 27000
    },
    {
      "epoch": 2.42,
      "grad_norm": 2.0975019931793213,
      "learning_rate": 2.5815671444903705e-05,
      "loss": 0.3202,
      "step": 27500
    },
    {
      "epoch": 2.46,
      "grad_norm": 2.1527884006500244,
      "learning_rate": 2.5375956380265588e-05,
      "loss": 0.3224,
      "step": 28000
    },
    {
      "epoch": 2.51,
      "grad_norm": 1.2337267398834229,
      "learning_rate": 2.4936241315627474e-05,
      "loss": 0.3164,
      "step": 28500
    },
    {
      "epoch": 2.55,
      "grad_norm": 1.6143097877502441,
      "learning_rate": 2.449652625098936e-05,
      "loss": 0.3194,
      "step": 29000
    },
    {
      "epoch": 2.59,
      "grad_norm": 1.901872158050537,
      "learning_rate": 2.4056811186351246e-05,
      "loss": 0.3179,
      "step": 29500
    },
    {
      "epoch": 2.64,
      "grad_norm": 1.9116158485412598,
      "learning_rate": 2.3617096121713133e-05,
      "loss": 0.3209,
      "step": 30000
    },
    {
      "epoch": 2.68,
      "grad_norm": 1.9545530080795288,
      "learning_rate": 2.317738105707502e-05,
      "loss": 0.3236,
      "step": 30500
    },
    {
      "epoch": 2.73,
      "grad_norm": 2.0340659618377686,
      "learning_rate": 2.27376659924369e-05,
      "loss": 0.3232,
      "step": 31000
    },
    {
      "epoch": 2.77,
      "grad_norm": 2.0332090854644775,
      "learning_rate": 2.2297950927798788e-05,
      "loss": 0.3189,
      "step": 31500
    },
    {
      "epoch": 2.81,
      "grad_norm": 2.475482225418091,
      "learning_rate": 2.1858235863160674e-05,
      "loss": 0.3209,
      "step": 32000
    },
    {
      "epoch": 2.86,
      "grad_norm": 1.7262321710586548,
      "learning_rate": 2.141852079852256e-05,
      "loss": 0.3199,
      "step": 32500
    },
    {
      "epoch": 2.9,
      "grad_norm": 1.8729243278503418,
      "learning_rate": 2.0978805733884443e-05,
      "loss": 0.3105,
      "step": 33000
    },
    {
      "epoch": 2.95,
      "grad_norm": 2.217402219772339,
      "learning_rate": 2.053909066924633e-05,
      "loss": 0.3164,
      "step": 33500
    },
    {
      "epoch": 2.99,
      "grad_norm": 1.25998055934906,
      "learning_rate": 2.0099375604608216e-05,
      "loss": 0.3205,
      "step": 34000
    },
    {
      "epoch": 3.0,
      "eval_combined_score": 0.815108010197236,
      "eval_f1": 0.815108010197236,
      "eval_loss": 0.31307029724121094,
      "eval_runtime": 31.9821,
      "eval_samples_per_second": 1264.144,
      "eval_steps_per_second": 4.94,
      "step": 34113
    },
    {
      "epoch": 3.03,
      "grad_norm": 2.3555874824523926,
      "learning_rate": 1.9659660539970102e-05,
      "loss": 0.3131,
      "step": 34500
    },
    {
      "epoch": 3.08,
      "grad_norm": 2.463350772857666,
      "learning_rate": 1.9219945475331985e-05,
      "loss": 0.3066,
      "step": 35000
    },
    {
      "epoch": 3.12,
      "grad_norm": 2.7274794578552246,
      "learning_rate": 1.878023041069387e-05,
      "loss": 0.305,
      "step": 35500
    },
    {
      "epoch": 3.17,
      "grad_norm": 2.4184601306915283,
      "learning_rate": 1.8340515346055757e-05,
      "loss": 0.312,
      "step": 36000
    },
    {
      "epoch": 3.21,
      "grad_norm": 2.4224226474761963,
      "learning_rate": 1.7900800281417643e-05,
      "loss": 0.3102,
      "step": 36500
    },
    {
      "epoch": 3.25,
      "grad_norm": 2.8739264011383057,
      "learning_rate": 1.7461085216779526e-05,
      "loss": 0.3053,
      "step": 37000
    },
    {
      "epoch": 3.3,
      "grad_norm": 1.939692497253418,
      "learning_rate": 1.7021370152141413e-05,
      "loss": 0.3092,
      "step": 37500
    },
    {
      "epoch": 3.34,
      "grad_norm": 1.3879015445709229,
      "learning_rate": 1.65816550875033e-05,
      "loss": 0.3103,
      "step": 38000
    },
    {
      "epoch": 3.39,
      "grad_norm": 1.4846141338348389,
      "learning_rate": 1.6141940022865185e-05,
      "loss": 0.3053,
      "step": 38500
    },
    {
      "epoch": 3.43,
      "grad_norm": 0.8230707049369812,
      "learning_rate": 1.5702224958227068e-05,
      "loss": 0.3136,
      "step": 39000
    },
    {
      "epoch": 3.47,
      "grad_norm": 1.7073547840118408,
      "learning_rate": 1.5262509893588954e-05,
      "loss": 0.3023,
      "step": 39500
    },
    {
      "epoch": 3.52,
      "grad_norm": 3.840158224105835,
      "learning_rate": 1.482279482895084e-05,
      "loss": 0.3044,
      "step": 40000
    },
    {
      "epoch": 3.56,
      "grad_norm": 2.369351863861084,
      "learning_rate": 1.4383079764312727e-05,
      "loss": 0.3064,
      "step": 40500
    },
    {
      "epoch": 3.61,
      "grad_norm": 2.696202039718628,
      "learning_rate": 1.3943364699674611e-05,
      "loss": 0.304,
      "step": 41000
    },
    {
      "epoch": 3.65,
      "grad_norm": 2.2328903675079346,
      "learning_rate": 1.3503649635036497e-05,
      "loss": 0.3042,
      "step": 41500
    },
    {
      "epoch": 3.69,
      "grad_norm": 2.5550007820129395,
      "learning_rate": 1.3063934570398382e-05,
      "loss": 0.2995,
      "step": 42000
    },
    {
      "epoch": 3.74,
      "grad_norm": 2.1913669109344482,
      "learning_rate": 1.2624219505760268e-05,
      "loss": 0.3028,
      "step": 42500
    },
    {
      "epoch": 3.78,
      "grad_norm": 1.6333726644515991,
      "learning_rate": 1.2184504441122153e-05,
      "loss": 0.3054,
      "step": 43000
    },
    {
      "epoch": 3.83,
      "grad_norm": 1.4576929807662964,
      "learning_rate": 1.1744789376484039e-05,
      "loss": 0.3077,
      "step": 43500
    },
    {
      "epoch": 3.87,
      "grad_norm": 2.146752119064331,
      "learning_rate": 1.1305074311845925e-05,
      "loss": 0.3045,
      "step": 44000
    },
    {
      "epoch": 3.91,
      "grad_norm": 1.7741210460662842,
      "learning_rate": 1.086535924720781e-05,
      "loss": 0.3052,
      "step": 44500
    },
    {
      "epoch": 3.96,
      "grad_norm": 1.5352250337600708,
      "learning_rate": 1.0425644182569696e-05,
      "loss": 0.3033,
      "step": 45000
    },
    {
      "epoch": 4.0,
      "eval_combined_score": 0.8163444343219625,
      "eval_f1": 0.8163444343219625,
      "eval_loss": 0.31015390157699585,
      "eval_runtime": 32.0121,
      "eval_samples_per_second": 1262.959,
      "eval_steps_per_second": 4.936,
      "step": 45484
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.7559531927108765,
      "learning_rate": 9.98592911793158e-06,
      "loss": 0.3074,
      "step": 45500
    },
    {
      "epoch": 4.05,
      "grad_norm": 2.4789981842041016,
      "learning_rate": 9.546214053293467e-06,
      "loss": 0.3053,
      "step": 46000
    },
    {
      "epoch": 4.09,
      "grad_norm": 1.871576189994812,
      "learning_rate": 9.106498988655351e-06,
      "loss": 0.2995,
      "step": 46500
    },
    {
      "epoch": 4.13,
      "grad_norm": 2.3308749198913574,
      "learning_rate": 8.666783924017237e-06,
      "loss": 0.2979,
      "step": 47000
    },
    {
      "epoch": 4.18,
      "grad_norm": 1.8183728456497192,
      "learning_rate": 8.227068859379122e-06,
      "loss": 0.3041,
      "step": 47500
    },
    {
      "epoch": 4.22,
      "grad_norm": 2.693516969680786,
      "learning_rate": 7.787353794741008e-06,
      "loss": 0.2882,
      "step": 48000
    },
    {
      "epoch": 4.27,
      "grad_norm": 2.347576379776001,
      "learning_rate": 7.347638730102893e-06,
      "loss": 0.2992,
      "step": 48500
    },
    {
      "epoch": 4.31,
      "grad_norm": 2.4476795196533203,
      "learning_rate": 6.907923665464779e-06,
      "loss": 0.3011,
      "step": 49000
    },
    {
      "epoch": 4.35,
      "grad_norm": 1.8371644020080566,
      "learning_rate": 6.468208600826664e-06,
      "loss": 0.3023,
      "step": 49500
    },
    {
      "epoch": 4.4,
      "grad_norm": 2.298017740249634,
      "learning_rate": 6.02849353618855e-06,
      "loss": 0.2936,
      "step": 50000
    },
    {
      "epoch": 4.44,
      "grad_norm": 1.397950291633606,
      "learning_rate": 5.588778471550435e-06,
      "loss": 0.2872,
      "step": 50500
    },
    {
      "epoch": 4.49,
      "grad_norm": 2.077183485031128,
      "learning_rate": 5.149063406912321e-06,
      "loss": 0.2969,
      "step": 51000
    },
    {
      "epoch": 4.53,
      "grad_norm": 2.3682615756988525,
      "learning_rate": 4.709348342274207e-06,
      "loss": 0.3035,
      "step": 51500
    },
    {
      "epoch": 4.57,
      "grad_norm": 2.9512171745300293,
      "learning_rate": 4.269633277636092e-06,
      "loss": 0.2969,
      "step": 52000
    },
    {
      "epoch": 4.62,
      "grad_norm": 3.269263982772827,
      "learning_rate": 3.8299182129979776e-06,
      "loss": 0.3035,
      "step": 52500
    },
    {
      "epoch": 4.66,
      "grad_norm": 1.814272165298462,
      "learning_rate": 3.3902031483598625e-06,
      "loss": 0.2905,
      "step": 53000
    },
    {
      "epoch": 4.7,
      "grad_norm": 3.355680227279663,
      "learning_rate": 2.9504880837217483e-06,
      "loss": 0.2931,
      "step": 53500
    },
    {
      "epoch": 4.75,
      "grad_norm": 2.171879291534424,
      "learning_rate": 2.5107730190836337e-06,
      "loss": 0.3006,
      "step": 54000
    },
    {
      "epoch": 4.79,
      "grad_norm": 2.100731611251831,
      "learning_rate": 2.0710579544455195e-06,
      "loss": 0.3007,
      "step": 54500
    },
    {
      "epoch": 4.84,
      "grad_norm": 1.9683380126953125,
      "learning_rate": 1.6313428898074047e-06,
      "loss": 0.2956,
      "step": 55000
    },
    {
      "epoch": 4.88,
      "grad_norm": 1.850870966911316,
      "learning_rate": 1.1916278251692903e-06,
      "loss": 0.2917,
      "step": 55500
    },
    {
      "epoch": 4.92,
      "grad_norm": 1.9473196268081665,
      "learning_rate": 7.519127605311758e-07,
      "loss": 0.2973,
      "step": 56000
    },
    {
      "epoch": 4.97,
      "grad_norm": 2.394254446029663,
      "learning_rate": 3.121976958930613e-07,
      "loss": 0.2965,
      "step": 56500
    },
    {
      "epoch": 5.0,
      "eval_combined_score": 0.8175030270415714,
      "eval_f1": 0.8175030270415714,
      "eval_loss": 0.3099803924560547,
      "eval_runtime": 32.1894,
      "eval_samples_per_second": 1256.003,
      "eval_steps_per_second": 4.908,
      "step": 56855
    },
    {
      "epoch": 5.0,
      "step": 56855,
      "total_flos": 5.318022316336128e+16,
      "train_loss": 0.33900219148470306,
      "train_runtime": 5127.7424,
      "train_samples_per_second": 354.782,
      "train_steps_per_second": 11.088
    }
  ],
  "logging_steps": 500,
  "max_steps": 56855,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 5.318022316336128e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}