roberta-base-qqp / trainer_state.json
Jeremiah Zhou
End of training
2d09fc6
raw
history blame
No virus
59.7 kB
{
"best_metric": 0.9152609448429384,
"best_model_checkpoint": "./fine-tune/roberta-base/qqp/checkpoint-159187",
"epoch": 10.0,
"global_step": 227410,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 7.328691828508612e-07,
"loss": 0.6927,
"step": 500
},
{
"epoch": 0.04,
"learning_rate": 1.4657383657017225e-06,
"loss": 0.5606,
"step": 1000
},
{
"epoch": 0.07,
"learning_rate": 2.1986075485525834e-06,
"loss": 0.4399,
"step": 1500
},
{
"epoch": 0.09,
"learning_rate": 2.931476731403445e-06,
"loss": 0.4018,
"step": 2000
},
{
"epoch": 0.11,
"learning_rate": 3.6643459142543057e-06,
"loss": 0.3965,
"step": 2500
},
{
"epoch": 0.13,
"learning_rate": 4.397215097105167e-06,
"loss": 0.3687,
"step": 3000
},
{
"epoch": 0.15,
"learning_rate": 5.130084279956028e-06,
"loss": 0.3724,
"step": 3500
},
{
"epoch": 0.18,
"learning_rate": 5.86295346280689e-06,
"loss": 0.3498,
"step": 4000
},
{
"epoch": 0.2,
"learning_rate": 6.59582264565775e-06,
"loss": 0.366,
"step": 4500
},
{
"epoch": 0.22,
"learning_rate": 7.328691828508611e-06,
"loss": 0.3437,
"step": 5000
},
{
"epoch": 0.24,
"learning_rate": 8.061561011359473e-06,
"loss": 0.3419,
"step": 5500
},
{
"epoch": 0.26,
"learning_rate": 8.794430194210334e-06,
"loss": 0.3382,
"step": 6000
},
{
"epoch": 0.29,
"learning_rate": 9.527299377061196e-06,
"loss": 0.3371,
"step": 6500
},
{
"epoch": 0.31,
"learning_rate": 1.0260168559912056e-05,
"loss": 0.341,
"step": 7000
},
{
"epoch": 0.33,
"learning_rate": 1.0993037742762918e-05,
"loss": 0.3278,
"step": 7500
},
{
"epoch": 0.35,
"learning_rate": 1.172590692561378e-05,
"loss": 0.3338,
"step": 8000
},
{
"epoch": 0.37,
"learning_rate": 1.245877610846464e-05,
"loss": 0.3308,
"step": 8500
},
{
"epoch": 0.4,
"learning_rate": 1.31916452913155e-05,
"loss": 0.3375,
"step": 9000
},
{
"epoch": 0.42,
"learning_rate": 1.3924514474166362e-05,
"loss": 0.3292,
"step": 9500
},
{
"epoch": 0.44,
"learning_rate": 1.4657383657017223e-05,
"loss": 0.3196,
"step": 10000
},
{
"epoch": 0.46,
"learning_rate": 1.5390252839868085e-05,
"loss": 0.3253,
"step": 10500
},
{
"epoch": 0.48,
"learning_rate": 1.6123122022718947e-05,
"loss": 0.3102,
"step": 11000
},
{
"epoch": 0.51,
"learning_rate": 1.6855991205569805e-05,
"loss": 0.3211,
"step": 11500
},
{
"epoch": 0.53,
"learning_rate": 1.7588860388420667e-05,
"loss": 0.3095,
"step": 12000
},
{
"epoch": 0.55,
"learning_rate": 1.832172957127153e-05,
"loss": 0.2994,
"step": 12500
},
{
"epoch": 0.57,
"learning_rate": 1.905459875412239e-05,
"loss": 0.3151,
"step": 13000
},
{
"epoch": 0.59,
"learning_rate": 1.9787467936973253e-05,
"loss": 0.3009,
"step": 13500
},
{
"epoch": 0.62,
"learning_rate": 1.9966785956541065e-05,
"loss": 0.3044,
"step": 14000
},
{
"epoch": 0.64,
"learning_rate": 1.992000561364115e-05,
"loss": 0.3051,
"step": 14500
},
{
"epoch": 0.66,
"learning_rate": 1.9873225270741237e-05,
"loss": 0.3045,
"step": 15000
},
{
"epoch": 0.68,
"learning_rate": 1.982644492784132e-05,
"loss": 0.315,
"step": 15500
},
{
"epoch": 0.7,
"learning_rate": 1.977966458494141e-05,
"loss": 0.3165,
"step": 16000
},
{
"epoch": 0.73,
"learning_rate": 1.9732884242041496e-05,
"loss": 0.2932,
"step": 16500
},
{
"epoch": 0.75,
"learning_rate": 1.9686103899141583e-05,
"loss": 0.3014,
"step": 17000
},
{
"epoch": 0.77,
"learning_rate": 1.963932355624167e-05,
"loss": 0.3052,
"step": 17500
},
{
"epoch": 0.79,
"learning_rate": 1.9592543213341755e-05,
"loss": 0.293,
"step": 18000
},
{
"epoch": 0.81,
"learning_rate": 1.9545762870441842e-05,
"loss": 0.29,
"step": 18500
},
{
"epoch": 0.84,
"learning_rate": 1.9498982527541926e-05,
"loss": 0.2814,
"step": 19000
},
{
"epoch": 0.86,
"learning_rate": 1.9452202184642014e-05,
"loss": 0.2884,
"step": 19500
},
{
"epoch": 0.88,
"learning_rate": 1.94054218417421e-05,
"loss": 0.2901,
"step": 20000
},
{
"epoch": 0.9,
"learning_rate": 1.935864149884219e-05,
"loss": 0.2833,
"step": 20500
},
{
"epoch": 0.92,
"learning_rate": 1.9311861155942276e-05,
"loss": 0.2877,
"step": 21000
},
{
"epoch": 0.95,
"learning_rate": 1.926508081304236e-05,
"loss": 0.2935,
"step": 21500
},
{
"epoch": 0.97,
"learning_rate": 1.9218300470142448e-05,
"loss": 0.2783,
"step": 22000
},
{
"epoch": 0.99,
"learning_rate": 1.9171520127242532e-05,
"loss": 0.2751,
"step": 22500
},
{
"epoch": 1.0,
"eval_accuracy": 0.8904773682908731,
"eval_combined_score": 0.8708585188561804,
"eval_f1": 0.8512396694214877,
"eval_loss": 0.3056511878967285,
"eval_runtime": 72.5914,
"eval_samples_per_second": 556.953,
"eval_steps_per_second": 69.623,
"step": 22741
},
{
"epoch": 1.01,
"learning_rate": 1.912473978434262e-05,
"loss": 0.2699,
"step": 23000
},
{
"epoch": 1.03,
"learning_rate": 1.9077959441442707e-05,
"loss": 0.2617,
"step": 23500
},
{
"epoch": 1.06,
"learning_rate": 1.9031179098542794e-05,
"loss": 0.246,
"step": 24000
},
{
"epoch": 1.08,
"learning_rate": 1.898439875564288e-05,
"loss": 0.2603,
"step": 24500
},
{
"epoch": 1.1,
"learning_rate": 1.8937618412742966e-05,
"loss": 0.2488,
"step": 25000
},
{
"epoch": 1.12,
"learning_rate": 1.8890838069843053e-05,
"loss": 0.2394,
"step": 25500
},
{
"epoch": 1.14,
"learning_rate": 1.884405772694314e-05,
"loss": 0.2558,
"step": 26000
},
{
"epoch": 1.17,
"learning_rate": 1.8797277384043228e-05,
"loss": 0.2536,
"step": 26500
},
{
"epoch": 1.19,
"learning_rate": 1.8750497041143315e-05,
"loss": 0.2459,
"step": 27000
},
{
"epoch": 1.21,
"learning_rate": 1.87037166982434e-05,
"loss": 0.2451,
"step": 27500
},
{
"epoch": 1.23,
"learning_rate": 1.8656936355343487e-05,
"loss": 0.2547,
"step": 28000
},
{
"epoch": 1.25,
"learning_rate": 1.861015601244357e-05,
"loss": 0.2571,
"step": 28500
},
{
"epoch": 1.28,
"learning_rate": 1.856337566954366e-05,
"loss": 0.2622,
"step": 29000
},
{
"epoch": 1.3,
"learning_rate": 1.8516595326643746e-05,
"loss": 0.2642,
"step": 29500
},
{
"epoch": 1.32,
"learning_rate": 1.8469814983743833e-05,
"loss": 0.2524,
"step": 30000
},
{
"epoch": 1.34,
"learning_rate": 1.842303464084392e-05,
"loss": 0.2604,
"step": 30500
},
{
"epoch": 1.36,
"learning_rate": 1.8376254297944005e-05,
"loss": 0.2476,
"step": 31000
},
{
"epoch": 1.39,
"learning_rate": 1.8329473955044092e-05,
"loss": 0.2494,
"step": 31500
},
{
"epoch": 1.41,
"learning_rate": 1.8282693612144176e-05,
"loss": 0.2513,
"step": 32000
},
{
"epoch": 1.43,
"learning_rate": 1.8235913269244264e-05,
"loss": 0.254,
"step": 32500
},
{
"epoch": 1.45,
"learning_rate": 1.818913292634435e-05,
"loss": 0.2505,
"step": 33000
},
{
"epoch": 1.47,
"learning_rate": 1.814235258344444e-05,
"loss": 0.249,
"step": 33500
},
{
"epoch": 1.5,
"learning_rate": 1.8095572240544526e-05,
"loss": 0.2469,
"step": 34000
},
{
"epoch": 1.52,
"learning_rate": 1.804879189764461e-05,
"loss": 0.2538,
"step": 34500
},
{
"epoch": 1.54,
"learning_rate": 1.8002011554744698e-05,
"loss": 0.24,
"step": 35000
},
{
"epoch": 1.56,
"learning_rate": 1.7955231211844785e-05,
"loss": 0.2488,
"step": 35500
},
{
"epoch": 1.58,
"learning_rate": 1.7908450868944873e-05,
"loss": 0.2552,
"step": 36000
},
{
"epoch": 1.61,
"learning_rate": 1.7861670526044957e-05,
"loss": 0.2485,
"step": 36500
},
{
"epoch": 1.63,
"learning_rate": 1.7814890183145044e-05,
"loss": 0.2518,
"step": 37000
},
{
"epoch": 1.65,
"learning_rate": 1.776810984024513e-05,
"loss": 0.2316,
"step": 37500
},
{
"epoch": 1.67,
"learning_rate": 1.7721329497345216e-05,
"loss": 0.2382,
"step": 38000
},
{
"epoch": 1.69,
"learning_rate": 1.7674549154445303e-05,
"loss": 0.2435,
"step": 38500
},
{
"epoch": 1.71,
"learning_rate": 1.762776881154539e-05,
"loss": 0.2359,
"step": 39000
},
{
"epoch": 1.74,
"learning_rate": 1.7580988468645478e-05,
"loss": 0.2439,
"step": 39500
},
{
"epoch": 1.76,
"learning_rate": 1.7534208125745562e-05,
"loss": 0.2386,
"step": 40000
},
{
"epoch": 1.78,
"learning_rate": 1.748742778284565e-05,
"loss": 0.2404,
"step": 40500
},
{
"epoch": 1.8,
"learning_rate": 1.7440647439945737e-05,
"loss": 0.2534,
"step": 41000
},
{
"epoch": 1.82,
"learning_rate": 1.739386709704582e-05,
"loss": 0.2416,
"step": 41500
},
{
"epoch": 1.85,
"learning_rate": 1.734708675414591e-05,
"loss": 0.2517,
"step": 42000
},
{
"epoch": 1.87,
"learning_rate": 1.7300306411245996e-05,
"loss": 0.2508,
"step": 42500
},
{
"epoch": 1.89,
"learning_rate": 1.7253526068346083e-05,
"loss": 0.2515,
"step": 43000
},
{
"epoch": 1.91,
"learning_rate": 1.7206745725446168e-05,
"loss": 0.2507,
"step": 43500
},
{
"epoch": 1.93,
"learning_rate": 1.7159965382546255e-05,
"loss": 0.2528,
"step": 44000
},
{
"epoch": 1.96,
"learning_rate": 1.7113185039646342e-05,
"loss": 0.2471,
"step": 44500
},
{
"epoch": 1.98,
"learning_rate": 1.7066404696746426e-05,
"loss": 0.2443,
"step": 45000
},
{
"epoch": 2.0,
"eval_accuracy": 0.9004699480583725,
"eval_combined_score": 0.8857477945420068,
"eval_f1": 0.8710256410256411,
"eval_loss": 0.2529826760292053,
"eval_runtime": 71.9777,
"eval_samples_per_second": 561.702,
"eval_steps_per_second": 70.216,
"step": 45482
},
{
"epoch": 2.0,
"learning_rate": 1.7019624353846514e-05,
"loss": 0.2582,
"step": 45500
},
{
"epoch": 2.02,
"learning_rate": 1.69728440109466e-05,
"loss": 0.2023,
"step": 46000
},
{
"epoch": 2.04,
"learning_rate": 1.692606366804669e-05,
"loss": 0.2099,
"step": 46500
},
{
"epoch": 2.07,
"learning_rate": 1.6879283325146776e-05,
"loss": 0.2071,
"step": 47000
},
{
"epoch": 2.09,
"learning_rate": 1.683250298224686e-05,
"loss": 0.213,
"step": 47500
},
{
"epoch": 2.11,
"learning_rate": 1.6785722639346948e-05,
"loss": 0.2089,
"step": 48000
},
{
"epoch": 2.13,
"learning_rate": 1.6738942296447035e-05,
"loss": 0.2201,
"step": 48500
},
{
"epoch": 2.15,
"learning_rate": 1.6692161953547123e-05,
"loss": 0.1979,
"step": 49000
},
{
"epoch": 2.18,
"learning_rate": 1.6645381610647207e-05,
"loss": 0.2059,
"step": 49500
},
{
"epoch": 2.2,
"learning_rate": 1.6598601267747294e-05,
"loss": 0.2146,
"step": 50000
},
{
"epoch": 2.22,
"learning_rate": 1.655182092484738e-05,
"loss": 0.2008,
"step": 50500
},
{
"epoch": 2.24,
"learning_rate": 1.6505040581947466e-05,
"loss": 0.205,
"step": 51000
},
{
"epoch": 2.26,
"learning_rate": 1.6458260239047553e-05,
"loss": 0.2135,
"step": 51500
},
{
"epoch": 2.29,
"learning_rate": 1.641147989614764e-05,
"loss": 0.2099,
"step": 52000
},
{
"epoch": 2.31,
"learning_rate": 1.6364699553247728e-05,
"loss": 0.1999,
"step": 52500
},
{
"epoch": 2.33,
"learning_rate": 1.6317919210347812e-05,
"loss": 0.2104,
"step": 53000
},
{
"epoch": 2.35,
"learning_rate": 1.62711388674479e-05,
"loss": 0.2135,
"step": 53500
},
{
"epoch": 2.37,
"learning_rate": 1.6224358524547987e-05,
"loss": 0.2066,
"step": 54000
},
{
"epoch": 2.4,
"learning_rate": 1.617757818164807e-05,
"loss": 0.2128,
"step": 54500
},
{
"epoch": 2.42,
"learning_rate": 1.613079783874816e-05,
"loss": 0.2079,
"step": 55000
},
{
"epoch": 2.44,
"learning_rate": 1.6084017495848246e-05,
"loss": 0.2142,
"step": 55500
},
{
"epoch": 2.46,
"learning_rate": 1.6037237152948333e-05,
"loss": 0.2079,
"step": 56000
},
{
"epoch": 2.48,
"learning_rate": 1.5990456810048418e-05,
"loss": 0.2065,
"step": 56500
},
{
"epoch": 2.51,
"learning_rate": 1.5943676467148505e-05,
"loss": 0.2166,
"step": 57000
},
{
"epoch": 2.53,
"learning_rate": 1.5896896124248592e-05,
"loss": 0.2135,
"step": 57500
},
{
"epoch": 2.55,
"learning_rate": 1.585011578134868e-05,
"loss": 0.2072,
"step": 58000
},
{
"epoch": 2.57,
"learning_rate": 1.5803335438448767e-05,
"loss": 0.2132,
"step": 58500
},
{
"epoch": 2.59,
"learning_rate": 1.575655509554885e-05,
"loss": 0.2114,
"step": 59000
},
{
"epoch": 2.62,
"learning_rate": 1.570977475264894e-05,
"loss": 0.2172,
"step": 59500
},
{
"epoch": 2.64,
"learning_rate": 1.5662994409749023e-05,
"loss": 0.198,
"step": 60000
},
{
"epoch": 2.66,
"learning_rate": 1.561621406684911e-05,
"loss": 0.2013,
"step": 60500
},
{
"epoch": 2.68,
"learning_rate": 1.5569433723949198e-05,
"loss": 0.2076,
"step": 61000
},
{
"epoch": 2.7,
"learning_rate": 1.5522653381049285e-05,
"loss": 0.2113,
"step": 61500
},
{
"epoch": 2.73,
"learning_rate": 1.5475873038149373e-05,
"loss": 0.2132,
"step": 62000
},
{
"epoch": 2.75,
"learning_rate": 1.5429092695249457e-05,
"loss": 0.2152,
"step": 62500
},
{
"epoch": 2.77,
"learning_rate": 1.5382312352349544e-05,
"loss": 0.2082,
"step": 63000
},
{
"epoch": 2.79,
"learning_rate": 1.533553200944963e-05,
"loss": 0.216,
"step": 63500
},
{
"epoch": 2.81,
"learning_rate": 1.5288751666549716e-05,
"loss": 0.2154,
"step": 64000
},
{
"epoch": 2.84,
"learning_rate": 1.5241971323649805e-05,
"loss": 0.2087,
"step": 64500
},
{
"epoch": 2.86,
"learning_rate": 1.519519098074989e-05,
"loss": 0.1985,
"step": 65000
},
{
"epoch": 2.88,
"learning_rate": 1.5148410637849978e-05,
"loss": 0.2117,
"step": 65500
},
{
"epoch": 2.9,
"learning_rate": 1.5101630294950062e-05,
"loss": 0.214,
"step": 66000
},
{
"epoch": 2.92,
"learning_rate": 1.505484995205015e-05,
"loss": 0.1941,
"step": 66500
},
{
"epoch": 2.95,
"learning_rate": 1.5008069609150235e-05,
"loss": 0.208,
"step": 67000
},
{
"epoch": 2.97,
"learning_rate": 1.4961289266250323e-05,
"loss": 0.2205,
"step": 67500
},
{
"epoch": 2.99,
"learning_rate": 1.491450892335041e-05,
"loss": 0.2157,
"step": 68000
},
{
"epoch": 3.0,
"eval_accuracy": 0.9069502844422459,
"eval_combined_score": 0.8919079780420185,
"eval_f1": 0.876865671641791,
"eval_loss": 0.2643309533596039,
"eval_runtime": 71.5158,
"eval_samples_per_second": 565.33,
"eval_steps_per_second": 70.67,
"step": 68223
},
{
"epoch": 3.01,
"learning_rate": 1.4867728580450496e-05,
"loss": 0.1811,
"step": 68500
},
{
"epoch": 3.03,
"learning_rate": 1.4820948237550584e-05,
"loss": 0.1629,
"step": 69000
},
{
"epoch": 3.06,
"learning_rate": 1.4774167894650668e-05,
"loss": 0.1615,
"step": 69500
},
{
"epoch": 3.08,
"learning_rate": 1.4727387551750755e-05,
"loss": 0.1736,
"step": 70000
},
{
"epoch": 3.1,
"learning_rate": 1.468060720885084e-05,
"loss": 0.1738,
"step": 70500
},
{
"epoch": 3.12,
"learning_rate": 1.4633826865950928e-05,
"loss": 0.1739,
"step": 71000
},
{
"epoch": 3.14,
"learning_rate": 1.4587046523051016e-05,
"loss": 0.1712,
"step": 71500
},
{
"epoch": 3.17,
"learning_rate": 1.4540266180151101e-05,
"loss": 0.1846,
"step": 72000
},
{
"epoch": 3.19,
"learning_rate": 1.4493485837251189e-05,
"loss": 0.1683,
"step": 72500
},
{
"epoch": 3.21,
"learning_rate": 1.4446705494351275e-05,
"loss": 0.168,
"step": 73000
},
{
"epoch": 3.23,
"learning_rate": 1.4399925151451362e-05,
"loss": 0.1761,
"step": 73500
},
{
"epoch": 3.25,
"learning_rate": 1.4353144808551446e-05,
"loss": 0.1691,
"step": 74000
},
{
"epoch": 3.28,
"learning_rate": 1.4306364465651534e-05,
"loss": 0.1766,
"step": 74500
},
{
"epoch": 3.3,
"learning_rate": 1.4259584122751621e-05,
"loss": 0.1738,
"step": 75000
},
{
"epoch": 3.32,
"learning_rate": 1.4212803779851707e-05,
"loss": 0.1738,
"step": 75500
},
{
"epoch": 3.34,
"learning_rate": 1.4166023436951794e-05,
"loss": 0.1839,
"step": 76000
},
{
"epoch": 3.36,
"learning_rate": 1.411924309405188e-05,
"loss": 0.1791,
"step": 76500
},
{
"epoch": 3.39,
"learning_rate": 1.4072462751151967e-05,
"loss": 0.1773,
"step": 77000
},
{
"epoch": 3.41,
"learning_rate": 1.4025682408252053e-05,
"loss": 0.1736,
"step": 77500
},
{
"epoch": 3.43,
"learning_rate": 1.397890206535214e-05,
"loss": 0.1695,
"step": 78000
},
{
"epoch": 3.45,
"learning_rate": 1.3932121722452228e-05,
"loss": 0.1785,
"step": 78500
},
{
"epoch": 3.47,
"learning_rate": 1.3885341379552312e-05,
"loss": 0.1712,
"step": 79000
},
{
"epoch": 3.5,
"learning_rate": 1.38385610366524e-05,
"loss": 0.1847,
"step": 79500
},
{
"epoch": 3.52,
"learning_rate": 1.3791780693752485e-05,
"loss": 0.181,
"step": 80000
},
{
"epoch": 3.54,
"learning_rate": 1.3745000350852573e-05,
"loss": 0.179,
"step": 80500
},
{
"epoch": 3.56,
"learning_rate": 1.3698220007952659e-05,
"loss": 0.1827,
"step": 81000
},
{
"epoch": 3.58,
"learning_rate": 1.3651439665052746e-05,
"loss": 0.1797,
"step": 81500
},
{
"epoch": 3.61,
"learning_rate": 1.3604659322152834e-05,
"loss": 0.1757,
"step": 82000
},
{
"epoch": 3.63,
"learning_rate": 1.355787897925292e-05,
"loss": 0.1703,
"step": 82500
},
{
"epoch": 3.65,
"learning_rate": 1.3511098636353007e-05,
"loss": 0.1751,
"step": 83000
},
{
"epoch": 3.67,
"learning_rate": 1.346431829345309e-05,
"loss": 0.1886,
"step": 83500
},
{
"epoch": 3.69,
"learning_rate": 1.3417537950553178e-05,
"loss": 0.1864,
"step": 84000
},
{
"epoch": 3.72,
"learning_rate": 1.3370757607653264e-05,
"loss": 0.1758,
"step": 84500
},
{
"epoch": 3.74,
"learning_rate": 1.3323977264753351e-05,
"loss": 0.1819,
"step": 85000
},
{
"epoch": 3.76,
"learning_rate": 1.3277196921853439e-05,
"loss": 0.171,
"step": 85500
},
{
"epoch": 3.78,
"learning_rate": 1.3230416578953525e-05,
"loss": 0.1789,
"step": 86000
},
{
"epoch": 3.8,
"learning_rate": 1.3183636236053612e-05,
"loss": 0.1836,
"step": 86500
},
{
"epoch": 3.83,
"learning_rate": 1.3136855893153698e-05,
"loss": 0.1841,
"step": 87000
},
{
"epoch": 3.85,
"learning_rate": 1.3090075550253785e-05,
"loss": 0.1843,
"step": 87500
},
{
"epoch": 3.87,
"learning_rate": 1.304329520735387e-05,
"loss": 0.1845,
"step": 88000
},
{
"epoch": 3.89,
"learning_rate": 1.2996514864453957e-05,
"loss": 0.1849,
"step": 88500
},
{
"epoch": 3.91,
"learning_rate": 1.2949734521554044e-05,
"loss": 0.1822,
"step": 89000
},
{
"epoch": 3.94,
"learning_rate": 1.290295417865413e-05,
"loss": 0.1859,
"step": 89500
},
{
"epoch": 3.96,
"learning_rate": 1.2856173835754218e-05,
"loss": 0.1928,
"step": 90000
},
{
"epoch": 3.98,
"learning_rate": 1.2809393492854303e-05,
"loss": 0.1838,
"step": 90500
},
{
"epoch": 4.0,
"eval_accuracy": 0.9109077417759089,
"eval_combined_score": 0.8961948553575517,
"eval_f1": 0.8814819689391945,
"eval_loss": 0.28062957525253296,
"eval_runtime": 73.9944,
"eval_samples_per_second": 546.393,
"eval_steps_per_second": 68.302,
"step": 90964
},
{
"epoch": 4.0,
"learning_rate": 1.276261314995439e-05,
"loss": 0.1721,
"step": 91000
},
{
"epoch": 4.02,
"learning_rate": 1.2715832807054476e-05,
"loss": 0.1357,
"step": 91500
},
{
"epoch": 4.05,
"learning_rate": 1.2669052464154564e-05,
"loss": 0.1472,
"step": 92000
},
{
"epoch": 4.07,
"learning_rate": 1.2622272121254651e-05,
"loss": 0.1466,
"step": 92500
},
{
"epoch": 4.09,
"learning_rate": 1.2575491778354735e-05,
"loss": 0.1449,
"step": 93000
},
{
"epoch": 4.11,
"learning_rate": 1.2528711435454823e-05,
"loss": 0.1462,
"step": 93500
},
{
"epoch": 4.13,
"learning_rate": 1.2481931092554909e-05,
"loss": 0.1412,
"step": 94000
},
{
"epoch": 4.16,
"learning_rate": 1.2435150749654996e-05,
"loss": 0.1477,
"step": 94500
},
{
"epoch": 4.18,
"learning_rate": 1.2388370406755084e-05,
"loss": 0.1565,
"step": 95000
},
{
"epoch": 4.2,
"learning_rate": 1.234159006385517e-05,
"loss": 0.1465,
"step": 95500
},
{
"epoch": 4.22,
"learning_rate": 1.2294809720955257e-05,
"loss": 0.1398,
"step": 96000
},
{
"epoch": 4.24,
"learning_rate": 1.224802937805534e-05,
"loss": 0.1419,
"step": 96500
},
{
"epoch": 4.27,
"learning_rate": 1.220124903515543e-05,
"loss": 0.1407,
"step": 97000
},
{
"epoch": 4.29,
"learning_rate": 1.2154468692255514e-05,
"loss": 0.1563,
"step": 97500
},
{
"epoch": 4.31,
"learning_rate": 1.2107688349355601e-05,
"loss": 0.1552,
"step": 98000
},
{
"epoch": 4.33,
"learning_rate": 1.2060908006455689e-05,
"loss": 0.1483,
"step": 98500
},
{
"epoch": 4.35,
"learning_rate": 1.2014127663555775e-05,
"loss": 0.1568,
"step": 99000
},
{
"epoch": 4.38,
"learning_rate": 1.1967347320655862e-05,
"loss": 0.1423,
"step": 99500
},
{
"epoch": 4.4,
"learning_rate": 1.1920566977755948e-05,
"loss": 0.1531,
"step": 100000
},
{
"epoch": 4.42,
"learning_rate": 1.1873786634856035e-05,
"loss": 0.1465,
"step": 100500
},
{
"epoch": 4.44,
"learning_rate": 1.182700629195612e-05,
"loss": 0.1551,
"step": 101000
},
{
"epoch": 4.46,
"learning_rate": 1.1780225949056207e-05,
"loss": 0.1492,
"step": 101500
},
{
"epoch": 4.49,
"learning_rate": 1.1733445606156294e-05,
"loss": 0.1456,
"step": 102000
},
{
"epoch": 4.51,
"learning_rate": 1.168666526325638e-05,
"loss": 0.1488,
"step": 102500
},
{
"epoch": 4.53,
"learning_rate": 1.1639884920356468e-05,
"loss": 0.141,
"step": 103000
},
{
"epoch": 4.55,
"learning_rate": 1.1593104577456553e-05,
"loss": 0.1518,
"step": 103500
},
{
"epoch": 4.57,
"learning_rate": 1.154632423455664e-05,
"loss": 0.1471,
"step": 104000
},
{
"epoch": 4.6,
"learning_rate": 1.1499543891656727e-05,
"loss": 0.1646,
"step": 104500
},
{
"epoch": 4.62,
"learning_rate": 1.1452763548756814e-05,
"loss": 0.1495,
"step": 105000
},
{
"epoch": 4.64,
"learning_rate": 1.1405983205856901e-05,
"loss": 0.1424,
"step": 105500
},
{
"epoch": 4.66,
"learning_rate": 1.1359202862956985e-05,
"loss": 0.1441,
"step": 106000
},
{
"epoch": 4.68,
"learning_rate": 1.1312422520057073e-05,
"loss": 0.1548,
"step": 106500
},
{
"epoch": 4.71,
"learning_rate": 1.1265642177157159e-05,
"loss": 0.1555,
"step": 107000
},
{
"epoch": 4.73,
"learning_rate": 1.1218861834257246e-05,
"loss": 0.1497,
"step": 107500
},
{
"epoch": 4.75,
"learning_rate": 1.1172081491357332e-05,
"loss": 0.1628,
"step": 108000
},
{
"epoch": 4.77,
"learning_rate": 1.112530114845742e-05,
"loss": 0.1454,
"step": 108500
},
{
"epoch": 4.79,
"learning_rate": 1.1078520805557507e-05,
"loss": 0.1394,
"step": 109000
},
{
"epoch": 4.82,
"learning_rate": 1.1031740462657593e-05,
"loss": 0.1535,
"step": 109500
},
{
"epoch": 4.84,
"learning_rate": 1.098496011975768e-05,
"loss": 0.1521,
"step": 110000
},
{
"epoch": 4.86,
"learning_rate": 1.0938179776857764e-05,
"loss": 0.1429,
"step": 110500
},
{
"epoch": 4.88,
"learning_rate": 1.0891399433957852e-05,
"loss": 0.1428,
"step": 111000
},
{
"epoch": 4.9,
"learning_rate": 1.0844619091057937e-05,
"loss": 0.1519,
"step": 111500
},
{
"epoch": 4.93,
"learning_rate": 1.0797838748158025e-05,
"loss": 0.1456,
"step": 112000
},
{
"epoch": 4.95,
"learning_rate": 1.0751058405258112e-05,
"loss": 0.1496,
"step": 112500
},
{
"epoch": 4.97,
"learning_rate": 1.0704278062358198e-05,
"loss": 0.1513,
"step": 113000
},
{
"epoch": 4.99,
"learning_rate": 1.0657497719458285e-05,
"loss": 0.146,
"step": 113500
},
{
"epoch": 5.0,
"eval_accuracy": 0.9112540192926045,
"eval_combined_score": 0.8960850249134777,
"eval_f1": 0.8809160305343511,
"eval_loss": 0.3276694118976593,
"eval_runtime": 71.6884,
"eval_samples_per_second": 563.969,
"eval_steps_per_second": 70.5,
"step": 113705
},
{
"epoch": 5.01,
"learning_rate": 1.0610717376558371e-05,
"loss": 0.128,
"step": 114000
},
{
"epoch": 5.03,
"learning_rate": 1.0563937033658459e-05,
"loss": 0.116,
"step": 114500
},
{
"epoch": 5.06,
"learning_rate": 1.0517156690758543e-05,
"loss": 0.1148,
"step": 115000
},
{
"epoch": 5.08,
"learning_rate": 1.047037634785863e-05,
"loss": 0.1074,
"step": 115500
},
{
"epoch": 5.1,
"learning_rate": 1.0423596004958718e-05,
"loss": 0.114,
"step": 116000
},
{
"epoch": 5.12,
"learning_rate": 1.0376815662058803e-05,
"loss": 0.1092,
"step": 116500
},
{
"epoch": 5.14,
"learning_rate": 1.033003531915889e-05,
"loss": 0.1183,
"step": 117000
},
{
"epoch": 5.17,
"learning_rate": 1.0283254976258977e-05,
"loss": 0.1137,
"step": 117500
},
{
"epoch": 5.19,
"learning_rate": 1.0236474633359064e-05,
"loss": 0.1144,
"step": 118000
},
{
"epoch": 5.21,
"learning_rate": 1.018969429045915e-05,
"loss": 0.1203,
"step": 118500
},
{
"epoch": 5.23,
"learning_rate": 1.0142913947559237e-05,
"loss": 0.1175,
"step": 119000
},
{
"epoch": 5.25,
"learning_rate": 1.0096133604659325e-05,
"loss": 0.1159,
"step": 119500
},
{
"epoch": 5.28,
"learning_rate": 1.0049353261759409e-05,
"loss": 0.1237,
"step": 120000
},
{
"epoch": 5.3,
"learning_rate": 1.0002572918859496e-05,
"loss": 0.1203,
"step": 120500
},
{
"epoch": 5.32,
"learning_rate": 9.955792575959584e-06,
"loss": 0.115,
"step": 121000
},
{
"epoch": 5.34,
"learning_rate": 9.90901223305967e-06,
"loss": 0.1209,
"step": 121500
},
{
"epoch": 5.36,
"learning_rate": 9.862231890159755e-06,
"loss": 0.1205,
"step": 122000
},
{
"epoch": 5.39,
"learning_rate": 9.815451547259843e-06,
"loss": 0.1253,
"step": 122500
},
{
"epoch": 5.41,
"learning_rate": 9.768671204359928e-06,
"loss": 0.121,
"step": 123000
},
{
"epoch": 5.43,
"learning_rate": 9.721890861460014e-06,
"loss": 0.1167,
"step": 123500
},
{
"epoch": 5.45,
"learning_rate": 9.675110518560102e-06,
"loss": 0.1199,
"step": 124000
},
{
"epoch": 5.47,
"learning_rate": 9.628330175660189e-06,
"loss": 0.1216,
"step": 124500
},
{
"epoch": 5.5,
"learning_rate": 9.581549832760275e-06,
"loss": 0.1203,
"step": 125000
},
{
"epoch": 5.52,
"learning_rate": 9.534769489860362e-06,
"loss": 0.1055,
"step": 125500
},
{
"epoch": 5.54,
"learning_rate": 9.487989146960448e-06,
"loss": 0.1211,
"step": 126000
},
{
"epoch": 5.56,
"learning_rate": 9.441208804060534e-06,
"loss": 0.1259,
"step": 126500
},
{
"epoch": 5.58,
"learning_rate": 9.394428461160621e-06,
"loss": 0.1184,
"step": 127000
},
{
"epoch": 5.61,
"learning_rate": 9.347648118260709e-06,
"loss": 0.1151,
"step": 127500
},
{
"epoch": 5.63,
"learning_rate": 9.300867775360794e-06,
"loss": 0.1182,
"step": 128000
},
{
"epoch": 5.65,
"learning_rate": 9.25408743246088e-06,
"loss": 0.1234,
"step": 128500
},
{
"epoch": 5.67,
"learning_rate": 9.207307089560968e-06,
"loss": 0.124,
"step": 129000
},
{
"epoch": 5.69,
"learning_rate": 9.160526746661053e-06,
"loss": 0.1233,
"step": 129500
},
{
"epoch": 5.72,
"learning_rate": 9.11374640376114e-06,
"loss": 0.1249,
"step": 130000
},
{
"epoch": 5.74,
"learning_rate": 9.066966060861227e-06,
"loss": 0.1294,
"step": 130500
},
{
"epoch": 5.76,
"learning_rate": 9.020185717961314e-06,
"loss": 0.1195,
"step": 131000
},
{
"epoch": 5.78,
"learning_rate": 8.9734053750614e-06,
"loss": 0.121,
"step": 131500
},
{
"epoch": 5.8,
"learning_rate": 8.926625032161487e-06,
"loss": 0.1222,
"step": 132000
},
{
"epoch": 5.83,
"learning_rate": 8.879844689261573e-06,
"loss": 0.1253,
"step": 132500
},
{
"epoch": 5.85,
"learning_rate": 8.833064346361659e-06,
"loss": 0.1286,
"step": 133000
},
{
"epoch": 5.87,
"learning_rate": 8.786284003461746e-06,
"loss": 0.1246,
"step": 133500
},
{
"epoch": 5.89,
"learning_rate": 8.739503660561832e-06,
"loss": 0.1243,
"step": 134000
},
{
"epoch": 5.91,
"learning_rate": 8.69272331766192e-06,
"loss": 0.1237,
"step": 134500
},
{
"epoch": 5.94,
"learning_rate": 8.645942974762007e-06,
"loss": 0.1131,
"step": 135000
},
{
"epoch": 5.96,
"learning_rate": 8.599162631862093e-06,
"loss": 0.126,
"step": 135500
},
{
"epoch": 5.98,
"learning_rate": 8.552382288962178e-06,
"loss": 0.1262,
"step": 136000
},
{
"epoch": 6.0,
"eval_accuracy": 0.9112787534009399,
"eval_combined_score": 0.8962263843527706,
"eval_f1": 0.8811740153046013,
"eval_loss": 0.3939257562160492,
"eval_runtime": 71.8403,
"eval_samples_per_second": 562.776,
"eval_steps_per_second": 70.35,
"step": 136446
},
{
"epoch": 6.0,
"learning_rate": 8.505601946062266e-06,
"loss": 0.1149,
"step": 136500
},
{
"epoch": 6.02,
"learning_rate": 8.458821603162352e-06,
"loss": 0.0775,
"step": 137000
},
{
"epoch": 6.05,
"learning_rate": 8.412041260262437e-06,
"loss": 0.0807,
"step": 137500
},
{
"epoch": 6.07,
"learning_rate": 8.365260917362525e-06,
"loss": 0.0882,
"step": 138000
},
{
"epoch": 6.09,
"learning_rate": 8.318480574462612e-06,
"loss": 0.0916,
"step": 138500
},
{
"epoch": 6.11,
"learning_rate": 8.271700231562698e-06,
"loss": 0.0919,
"step": 139000
},
{
"epoch": 6.13,
"learning_rate": 8.224919888662784e-06,
"loss": 0.0863,
"step": 139500
},
{
"epoch": 6.16,
"learning_rate": 8.178139545762871e-06,
"loss": 0.0955,
"step": 140000
},
{
"epoch": 6.18,
"learning_rate": 8.131359202862957e-06,
"loss": 0.0889,
"step": 140500
},
{
"epoch": 6.2,
"learning_rate": 8.084578859963044e-06,
"loss": 0.0957,
"step": 141000
},
{
"epoch": 6.22,
"learning_rate": 8.037798517063132e-06,
"loss": 0.1,
"step": 141500
},
{
"epoch": 6.24,
"learning_rate": 7.991018174163218e-06,
"loss": 0.0933,
"step": 142000
},
{
"epoch": 6.27,
"learning_rate": 7.944237831263303e-06,
"loss": 0.0844,
"step": 142500
},
{
"epoch": 6.29,
"learning_rate": 7.89745748836339e-06,
"loss": 0.0914,
"step": 143000
},
{
"epoch": 6.31,
"learning_rate": 7.850677145463477e-06,
"loss": 0.0885,
"step": 143500
},
{
"epoch": 6.33,
"learning_rate": 7.803896802563562e-06,
"loss": 0.0895,
"step": 144000
},
{
"epoch": 6.35,
"learning_rate": 7.75711645966365e-06,
"loss": 0.0869,
"step": 144500
},
{
"epoch": 6.38,
"learning_rate": 7.710336116763737e-06,
"loss": 0.0933,
"step": 145000
},
{
"epoch": 6.4,
"learning_rate": 7.663555773863823e-06,
"loss": 0.0841,
"step": 145500
},
{
"epoch": 6.42,
"learning_rate": 7.61677543096391e-06,
"loss": 0.0908,
"step": 146000
},
{
"epoch": 6.44,
"learning_rate": 7.569995088063996e-06,
"loss": 0.0942,
"step": 146500
},
{
"epoch": 6.46,
"learning_rate": 7.523214745164083e-06,
"loss": 0.0872,
"step": 147000
},
{
"epoch": 6.49,
"learning_rate": 7.476434402264169e-06,
"loss": 0.0894,
"step": 147500
},
{
"epoch": 6.51,
"learning_rate": 7.429654059364255e-06,
"loss": 0.0888,
"step": 148000
},
{
"epoch": 6.53,
"learning_rate": 7.382873716464343e-06,
"loss": 0.093,
"step": 148500
},
{
"epoch": 6.55,
"learning_rate": 7.336093373564429e-06,
"loss": 0.0885,
"step": 149000
},
{
"epoch": 6.57,
"learning_rate": 7.289313030664516e-06,
"loss": 0.092,
"step": 149500
},
{
"epoch": 6.6,
"learning_rate": 7.242532687764602e-06,
"loss": 0.0956,
"step": 150000
},
{
"epoch": 6.62,
"learning_rate": 7.195752344864688e-06,
"loss": 0.1009,
"step": 150500
},
{
"epoch": 6.64,
"learning_rate": 7.148972001964775e-06,
"loss": 0.0956,
"step": 151000
},
{
"epoch": 6.66,
"learning_rate": 7.102191659064862e-06,
"loss": 0.0908,
"step": 151500
},
{
"epoch": 6.68,
"learning_rate": 7.055411316164949e-06,
"loss": 0.0878,
"step": 152000
},
{
"epoch": 6.71,
"learning_rate": 7.008630973265035e-06,
"loss": 0.1011,
"step": 152500
},
{
"epoch": 6.73,
"learning_rate": 6.961850630365121e-06,
"loss": 0.0898,
"step": 153000
},
{
"epoch": 6.75,
"learning_rate": 6.915070287465208e-06,
"loss": 0.0948,
"step": 153500
},
{
"epoch": 6.77,
"learning_rate": 6.868289944565294e-06,
"loss": 0.0908,
"step": 154000
},
{
"epoch": 6.79,
"learning_rate": 6.82150960166538e-06,
"loss": 0.0935,
"step": 154500
},
{
"epoch": 6.82,
"learning_rate": 6.774729258765468e-06,
"loss": 0.0987,
"step": 155000
},
{
"epoch": 6.84,
"learning_rate": 6.727948915865554e-06,
"loss": 0.0971,
"step": 155500
},
{
"epoch": 6.86,
"learning_rate": 6.681168572965641e-06,
"loss": 0.0894,
"step": 156000
},
{
"epoch": 6.88,
"learning_rate": 6.634388230065727e-06,
"loss": 0.0925,
"step": 156500
},
{
"epoch": 6.9,
"learning_rate": 6.587607887165813e-06,
"loss": 0.0831,
"step": 157000
},
{
"epoch": 6.93,
"learning_rate": 6.5408275442659e-06,
"loss": 0.1021,
"step": 157500
},
{
"epoch": 6.95,
"learning_rate": 6.4940472013659864e-06,
"loss": 0.0974,
"step": 158000
},
{
"epoch": 6.97,
"learning_rate": 6.447266858466074e-06,
"loss": 0.0993,
"step": 158500
},
{
"epoch": 6.99,
"learning_rate": 6.40048651556616e-06,
"loss": 0.0867,
"step": 159000
},
{
"epoch": 7.0,
"eval_accuracy": 0.9152609448429384,
"eval_combined_score": 0.9009873932600381,
"eval_f1": 0.8867138416771377,
"eval_loss": 0.44352227449417114,
"eval_runtime": 71.6368,
"eval_samples_per_second": 564.374,
"eval_steps_per_second": 70.55,
"step": 159187
},
{
"epoch": 7.01,
"learning_rate": 6.353706172666246e-06,
"loss": 0.0648,
"step": 159500
},
{
"epoch": 7.04,
"learning_rate": 6.306925829766333e-06,
"loss": 0.0651,
"step": 160000
},
{
"epoch": 7.06,
"learning_rate": 6.2601454868664195e-06,
"loss": 0.0692,
"step": 160500
},
{
"epoch": 7.08,
"learning_rate": 6.213365143966505e-06,
"loss": 0.0699,
"step": 161000
},
{
"epoch": 7.1,
"learning_rate": 6.166584801066592e-06,
"loss": 0.0668,
"step": 161500
},
{
"epoch": 7.12,
"learning_rate": 6.119804458166679e-06,
"loss": 0.0606,
"step": 162000
},
{
"epoch": 7.15,
"learning_rate": 6.073024115266766e-06,
"loss": 0.0728,
"step": 162500
},
{
"epoch": 7.17,
"learning_rate": 6.0262437723668525e-06,
"loss": 0.0627,
"step": 163000
},
{
"epoch": 7.19,
"learning_rate": 5.979463429466938e-06,
"loss": 0.0691,
"step": 163500
},
{
"epoch": 7.21,
"learning_rate": 5.932683086567025e-06,
"loss": 0.0653,
"step": 164000
},
{
"epoch": 7.23,
"learning_rate": 5.8859027436671114e-06,
"loss": 0.0659,
"step": 164500
},
{
"epoch": 7.26,
"learning_rate": 5.839122400767197e-06,
"loss": 0.0613,
"step": 165000
},
{
"epoch": 7.28,
"learning_rate": 5.7923420578672855e-06,
"loss": 0.0599,
"step": 165500
},
{
"epoch": 7.3,
"learning_rate": 5.745561714967371e-06,
"loss": 0.0723,
"step": 166000
},
{
"epoch": 7.32,
"learning_rate": 5.698781372067458e-06,
"loss": 0.0709,
"step": 166500
},
{
"epoch": 7.34,
"learning_rate": 5.6520010291675445e-06,
"loss": 0.0714,
"step": 167000
},
{
"epoch": 7.37,
"learning_rate": 5.60522068626763e-06,
"loss": 0.0637,
"step": 167500
},
{
"epoch": 7.39,
"learning_rate": 5.558440343367717e-06,
"loss": 0.0726,
"step": 168000
},
{
"epoch": 7.41,
"learning_rate": 5.5116600004678034e-06,
"loss": 0.0632,
"step": 168500
},
{
"epoch": 7.43,
"learning_rate": 5.464879657567891e-06,
"loss": 0.0735,
"step": 169000
},
{
"epoch": 7.45,
"learning_rate": 5.4180993146679775e-06,
"loss": 0.0693,
"step": 169500
},
{
"epoch": 7.48,
"learning_rate": 5.371318971768063e-06,
"loss": 0.07,
"step": 170000
},
{
"epoch": 7.5,
"learning_rate": 5.32453862886815e-06,
"loss": 0.0699,
"step": 170500
},
{
"epoch": 7.52,
"learning_rate": 5.2777582859682365e-06,
"loss": 0.0617,
"step": 171000
},
{
"epoch": 7.54,
"learning_rate": 5.230977943068323e-06,
"loss": 0.056,
"step": 171500
},
{
"epoch": 7.56,
"learning_rate": 5.184197600168409e-06,
"loss": 0.0644,
"step": 172000
},
{
"epoch": 7.59,
"learning_rate": 5.137417257268496e-06,
"loss": 0.0694,
"step": 172500
},
{
"epoch": 7.61,
"learning_rate": 5.090636914368583e-06,
"loss": 0.0642,
"step": 173000
},
{
"epoch": 7.63,
"learning_rate": 5.0438565714686695e-06,
"loss": 0.0673,
"step": 173500
},
{
"epoch": 7.65,
"learning_rate": 4.997076228568756e-06,
"loss": 0.0682,
"step": 174000
},
{
"epoch": 7.67,
"learning_rate": 4.950295885668842e-06,
"loss": 0.0671,
"step": 174500
},
{
"epoch": 7.7,
"learning_rate": 4.903515542768929e-06,
"loss": 0.0734,
"step": 175000
},
{
"epoch": 7.72,
"learning_rate": 4.856735199869015e-06,
"loss": 0.0665,
"step": 175500
},
{
"epoch": 7.74,
"learning_rate": 4.809954856969102e-06,
"loss": 0.0644,
"step": 176000
},
{
"epoch": 7.76,
"learning_rate": 4.763174514069189e-06,
"loss": 0.0681,
"step": 176500
},
{
"epoch": 7.78,
"learning_rate": 4.716394171169275e-06,
"loss": 0.0755,
"step": 177000
},
{
"epoch": 7.81,
"learning_rate": 4.6696138282693615e-06,
"loss": 0.0758,
"step": 177500
},
{
"epoch": 7.83,
"learning_rate": 4.622833485369448e-06,
"loss": 0.0633,
"step": 178000
},
{
"epoch": 7.85,
"learning_rate": 4.576053142469535e-06,
"loss": 0.0603,
"step": 178500
},
{
"epoch": 7.87,
"learning_rate": 4.529272799569621e-06,
"loss": 0.0703,
"step": 179000
},
{
"epoch": 7.89,
"learning_rate": 4.482492456669708e-06,
"loss": 0.062,
"step": 179500
},
{
"epoch": 7.92,
"learning_rate": 4.4357121137697945e-06,
"loss": 0.0733,
"step": 180000
},
{
"epoch": 7.94,
"learning_rate": 4.388931770869881e-06,
"loss": 0.0717,
"step": 180500
},
{
"epoch": 7.96,
"learning_rate": 4.342151427969967e-06,
"loss": 0.0612,
"step": 181000
},
{
"epoch": 7.98,
"learning_rate": 4.295371085070054e-06,
"loss": 0.0757,
"step": 181500
},
{
"epoch": 8.0,
"eval_accuracy": 0.9147415285678951,
"eval_combined_score": 0.8995565346371729,
"eval_f1": 0.8843715407064506,
"eval_loss": 0.4811749756336212,
"eval_runtime": 72.7906,
"eval_samples_per_second": 555.429,
"eval_steps_per_second": 69.432,
"step": 181928
},
{
"epoch": 8.0,
"learning_rate": 4.248590742170141e-06,
"loss": 0.0629,
"step": 182000
},
{
"epoch": 8.03,
"learning_rate": 4.201810399270227e-06,
"loss": 0.0438,
"step": 182500
},
{
"epoch": 8.05,
"learning_rate": 4.155030056370313e-06,
"loss": 0.0419,
"step": 183000
},
{
"epoch": 8.07,
"learning_rate": 4.1082497134704e-06,
"loss": 0.0343,
"step": 183500
},
{
"epoch": 8.09,
"learning_rate": 4.0614693705704865e-06,
"loss": 0.0557,
"step": 184000
},
{
"epoch": 8.11,
"learning_rate": 4.014689027670573e-06,
"loss": 0.052,
"step": 184500
},
{
"epoch": 8.14,
"learning_rate": 3.96790868477066e-06,
"loss": 0.0446,
"step": 185000
},
{
"epoch": 8.16,
"learning_rate": 3.921128341870746e-06,
"loss": 0.0473,
"step": 185500
},
{
"epoch": 8.18,
"learning_rate": 3.874347998970833e-06,
"loss": 0.0455,
"step": 186000
},
{
"epoch": 8.2,
"learning_rate": 3.827567656070919e-06,
"loss": 0.0486,
"step": 186500
},
{
"epoch": 8.22,
"learning_rate": 3.780787313171006e-06,
"loss": 0.0412,
"step": 187000
},
{
"epoch": 8.25,
"learning_rate": 3.7340069702710923e-06,
"loss": 0.0502,
"step": 187500
},
{
"epoch": 8.27,
"learning_rate": 3.687226627371179e-06,
"loss": 0.0435,
"step": 188000
},
{
"epoch": 8.29,
"learning_rate": 3.6404462844712655e-06,
"loss": 0.0391,
"step": 188500
},
{
"epoch": 8.31,
"learning_rate": 3.593665941571352e-06,
"loss": 0.0476,
"step": 189000
},
{
"epoch": 8.33,
"learning_rate": 3.5468855986714383e-06,
"loss": 0.0468,
"step": 189500
},
{
"epoch": 8.35,
"learning_rate": 3.500105255771525e-06,
"loss": 0.043,
"step": 190000
},
{
"epoch": 8.38,
"learning_rate": 3.453324912871612e-06,
"loss": 0.0457,
"step": 190500
},
{
"epoch": 8.4,
"learning_rate": 3.406544569971698e-06,
"loss": 0.0445,
"step": 191000
},
{
"epoch": 8.42,
"learning_rate": 3.3597642270717847e-06,
"loss": 0.0532,
"step": 191500
},
{
"epoch": 8.44,
"learning_rate": 3.3129838841718713e-06,
"loss": 0.049,
"step": 192000
},
{
"epoch": 8.46,
"learning_rate": 3.266203541271958e-06,
"loss": 0.0451,
"step": 192500
},
{
"epoch": 8.49,
"learning_rate": 3.219423198372044e-06,
"loss": 0.0493,
"step": 193000
},
{
"epoch": 8.51,
"learning_rate": 3.172642855472131e-06,
"loss": 0.0387,
"step": 193500
},
{
"epoch": 8.53,
"learning_rate": 3.1258625125722173e-06,
"loss": 0.0597,
"step": 194000
},
{
"epoch": 8.55,
"learning_rate": 3.079082169672304e-06,
"loss": 0.0549,
"step": 194500
},
{
"epoch": 8.57,
"learning_rate": 3.03230182677239e-06,
"loss": 0.0455,
"step": 195000
},
{
"epoch": 8.6,
"learning_rate": 2.985521483872477e-06,
"loss": 0.0487,
"step": 195500
},
{
"epoch": 8.62,
"learning_rate": 2.9387411409725637e-06,
"loss": 0.0485,
"step": 196000
},
{
"epoch": 8.64,
"learning_rate": 2.89196079807265e-06,
"loss": 0.0471,
"step": 196500
},
{
"epoch": 8.66,
"learning_rate": 2.845180455172737e-06,
"loss": 0.0508,
"step": 197000
},
{
"epoch": 8.68,
"learning_rate": 2.798400112272823e-06,
"loss": 0.0489,
"step": 197500
},
{
"epoch": 8.71,
"learning_rate": 2.7516197693729097e-06,
"loss": 0.0498,
"step": 198000
},
{
"epoch": 8.73,
"learning_rate": 2.704839426472996e-06,
"loss": 0.0457,
"step": 198500
},
{
"epoch": 8.75,
"learning_rate": 2.658059083573083e-06,
"loss": 0.0434,
"step": 199000
},
{
"epoch": 8.77,
"learning_rate": 2.6112787406731695e-06,
"loss": 0.0394,
"step": 199500
},
{
"epoch": 8.79,
"learning_rate": 2.5644983977732557e-06,
"loss": 0.0422,
"step": 200000
},
{
"epoch": 8.82,
"learning_rate": 2.5177180548733427e-06,
"loss": 0.0505,
"step": 200500
},
{
"epoch": 8.84,
"learning_rate": 2.470937711973429e-06,
"loss": 0.0574,
"step": 201000
},
{
"epoch": 8.86,
"learning_rate": 2.4241573690735155e-06,
"loss": 0.0445,
"step": 201500
},
{
"epoch": 8.88,
"learning_rate": 2.377377026173602e-06,
"loss": 0.0565,
"step": 202000
},
{
"epoch": 8.9,
"learning_rate": 2.3305966832736887e-06,
"loss": 0.0519,
"step": 202500
},
{
"epoch": 8.93,
"learning_rate": 2.283816340373775e-06,
"loss": 0.0508,
"step": 203000
},
{
"epoch": 8.95,
"learning_rate": 2.237035997473862e-06,
"loss": 0.0467,
"step": 203500
},
{
"epoch": 8.97,
"learning_rate": 2.190255654573948e-06,
"loss": 0.0466,
"step": 204000
},
{
"epoch": 8.99,
"learning_rate": 2.1434753116740347e-06,
"loss": 0.0479,
"step": 204500
},
{
"epoch": 9.0,
"eval_accuracy": 0.9150878060845906,
"eval_combined_score": 0.9010856587900834,
"eval_f1": 0.8870835114955762,
"eval_loss": 0.5081153512001038,
"eval_runtime": 92.6456,
"eval_samples_per_second": 436.394,
"eval_steps_per_second": 54.552,
"step": 204669
},
{
"epoch": 9.01,
"learning_rate": 2.0966949687741213e-06,
"loss": 0.0354,
"step": 205000
},
{
"epoch": 9.04,
"learning_rate": 2.049914625874208e-06,
"loss": 0.037,
"step": 205500
},
{
"epoch": 9.06,
"learning_rate": 2.0031342829742946e-06,
"loss": 0.0354,
"step": 206000
},
{
"epoch": 9.08,
"learning_rate": 1.9563539400743807e-06,
"loss": 0.029,
"step": 206500
},
{
"epoch": 9.1,
"learning_rate": 1.9095735971744673e-06,
"loss": 0.0363,
"step": 207000
},
{
"epoch": 9.12,
"learning_rate": 1.862793254274554e-06,
"loss": 0.0331,
"step": 207500
},
{
"epoch": 9.15,
"learning_rate": 1.8160129113746405e-06,
"loss": 0.03,
"step": 208000
},
{
"epoch": 9.17,
"learning_rate": 1.7692325684747272e-06,
"loss": 0.0314,
"step": 208500
},
{
"epoch": 9.19,
"learning_rate": 1.7224522255748135e-06,
"loss": 0.033,
"step": 209000
},
{
"epoch": 9.21,
"learning_rate": 1.6756718826749001e-06,
"loss": 0.0307,
"step": 209500
},
{
"epoch": 9.23,
"learning_rate": 1.6288915397749865e-06,
"loss": 0.0284,
"step": 210000
},
{
"epoch": 9.26,
"learning_rate": 1.5821111968750731e-06,
"loss": 0.0311,
"step": 210500
},
{
"epoch": 9.28,
"learning_rate": 1.5353308539751595e-06,
"loss": 0.036,
"step": 211000
},
{
"epoch": 9.3,
"learning_rate": 1.4885505110752464e-06,
"loss": 0.0321,
"step": 211500
},
{
"epoch": 9.32,
"learning_rate": 1.441770168175333e-06,
"loss": 0.0303,
"step": 212000
},
{
"epoch": 9.34,
"learning_rate": 1.3949898252754194e-06,
"loss": 0.0326,
"step": 212500
},
{
"epoch": 9.37,
"learning_rate": 1.348209482375506e-06,
"loss": 0.03,
"step": 213000
},
{
"epoch": 9.39,
"learning_rate": 1.3014291394755924e-06,
"loss": 0.0325,
"step": 213500
},
{
"epoch": 9.41,
"learning_rate": 1.254648796575679e-06,
"loss": 0.034,
"step": 214000
},
{
"epoch": 9.43,
"learning_rate": 1.2078684536757656e-06,
"loss": 0.0423,
"step": 214500
},
{
"epoch": 9.45,
"learning_rate": 1.161088110775852e-06,
"loss": 0.0353,
"step": 215000
},
{
"epoch": 9.48,
"learning_rate": 1.1143077678759386e-06,
"loss": 0.0301,
"step": 215500
},
{
"epoch": 9.5,
"learning_rate": 1.0675274249760252e-06,
"loss": 0.0315,
"step": 216000
},
{
"epoch": 9.52,
"learning_rate": 1.0207470820761118e-06,
"loss": 0.0361,
"step": 216500
},
{
"epoch": 9.54,
"learning_rate": 9.739667391761982e-07,
"loss": 0.0388,
"step": 217000
},
{
"epoch": 9.56,
"learning_rate": 9.271863962762848e-07,
"loss": 0.0309,
"step": 217500
},
{
"epoch": 9.59,
"learning_rate": 8.804060533763713e-07,
"loss": 0.0299,
"step": 218000
},
{
"epoch": 9.61,
"learning_rate": 8.336257104764578e-07,
"loss": 0.0357,
"step": 218500
},
{
"epoch": 9.63,
"learning_rate": 7.868453675765445e-07,
"loss": 0.0285,
"step": 219000
},
{
"epoch": 9.65,
"learning_rate": 7.40065024676631e-07,
"loss": 0.0238,
"step": 219500
},
{
"epoch": 9.67,
"learning_rate": 6.932846817767175e-07,
"loss": 0.0284,
"step": 220000
},
{
"epoch": 9.7,
"learning_rate": 6.46504338876804e-07,
"loss": 0.033,
"step": 220500
},
{
"epoch": 9.72,
"learning_rate": 5.997239959768906e-07,
"loss": 0.031,
"step": 221000
},
{
"epoch": 9.74,
"learning_rate": 5.529436530769771e-07,
"loss": 0.028,
"step": 221500
},
{
"epoch": 9.76,
"learning_rate": 5.061633101770636e-07,
"loss": 0.0264,
"step": 222000
},
{
"epoch": 9.78,
"learning_rate": 4.5938296727715023e-07,
"loss": 0.0341,
"step": 222500
},
{
"epoch": 9.81,
"learning_rate": 4.1260262437723673e-07,
"loss": 0.0252,
"step": 223000
},
{
"epoch": 9.83,
"learning_rate": 3.6582228147732323e-07,
"loss": 0.0348,
"step": 223500
},
{
"epoch": 9.85,
"learning_rate": 3.1904193857740983e-07,
"loss": 0.0315,
"step": 224000
},
{
"epoch": 9.87,
"learning_rate": 2.7226159567749633e-07,
"loss": 0.0352,
"step": 224500
},
{
"epoch": 9.89,
"learning_rate": 2.2548125277758288e-07,
"loss": 0.0342,
"step": 225000
},
{
"epoch": 9.92,
"learning_rate": 1.787009098776694e-07,
"loss": 0.0313,
"step": 225500
},
{
"epoch": 9.94,
"learning_rate": 1.3192056697775596e-07,
"loss": 0.0324,
"step": 226000
},
{
"epoch": 9.96,
"learning_rate": 8.51402240778425e-08,
"loss": 0.0274,
"step": 226500
},
{
"epoch": 9.98,
"learning_rate": 3.835988117792904e-08,
"loss": 0.0379,
"step": 227000
},
{
"epoch": 10.0,
"eval_accuracy": 0.9148899332179075,
"eval_combined_score": 0.9003480960683503,
"eval_f1": 0.8858062589187933,
"eval_loss": 0.5646682977676392,
"eval_runtime": 85.1463,
"eval_samples_per_second": 474.83,
"eval_steps_per_second": 59.357,
"step": 227410
},
{
"epoch": 10.0,
"step": 227410,
"total_flos": 2.393297626212864e+17,
"train_loss": 0.1477880122620598,
"train_runtime": 22868.2253,
"train_samples_per_second": 159.105,
"train_steps_per_second": 9.944
}
],
"max_steps": 227410,
"num_train_epochs": 10,
"total_flos": 2.393297626212864e+17,
"trial_name": null,
"trial_params": null
}