2024-02-23 03:36:48 Publish script update

data/time_of_day_comparison.csv

@@ -5,46 +5,47 @@ model,time_of_day,execution_time,characters_count,words_count
 01-ai Yi Chat (34B),late afternoon,9.417656523840767,384.9583333333333,40.125
 01-ai Yi Chat (34B),evening,8.16779062861488,360.8333333333333,39.375
 01-ai Yi Chat (34B),late evening,9.680033167203268,424.5,40.875
-01-ai Yi Chat (34B),midnight,9.165155977816195,389.65,39.95
-01-ai Yi Chat (34B),night,9.463179730354472,413.125,40.5625
+01-ai Yi Chat (34B),midnight,9.311747227774726,397.0625,40.416666666666664
+01-ai Yi Chat (34B),night,9.534157488562844,416.125,40.875
 Chronos Hermes (13B),early morning,7.70508497012289,461.4761904761905,49.595238095238095
 Chronos Hermes (13B),morning,8.221057146787643,595.375,64.5
 Chronos Hermes (13B),afternoon,6.651120054020601,539.5,76.10416666666667
 Chronos Hermes (13B),late afternoon,7.630303209478205,544.625,62.833333333333336
 Chronos Hermes (13B),evening,6.468730966250102,543.05,72.15
 Chronos Hermes (13B),late evening,7.710561646355523,423.1363636363636,46.31818181818182
-Chronos Hermes (13B),midnight,7.482297485524958,501.55263157894734,58.89473684210526
-Chronos Hermes (13B),night,7.764214697091476,546.28,62.56
+Chronos Hermes (13B),midnight,7.710010987520218,486.5217391304348,54.93478260869565
+Chronos Hermes (13B),night,8.01242661917651,553.051724137931,62.827586206896555
 Falcon Instruct (7B),early morning,1.1807806547297988,45.6875,8.4375
 Falcon Instruct (7B),morning,1.4012710821060907,47.291666666666664,8.75
 Falcon Instruct (7B),afternoon,1.5774729509611387,36.96875,7.020833333333333
 Falcon Instruct (7B),late afternoon,1.0884015900748116,43.833333333333336,8.166666666666666
 Falcon Instruct (7B),evening,0.7874892950057983,39.2,7.4
 Falcon Instruct (7B),late evening,1.2339241071180864,46.791666666666664,8.708333333333334
-Falcon Instruct (7B),midnight,0.9522609934210777,37.45,7.075
-Falcon Instruct (7B),night,1.4049408435821533,50.6,9.32
+Falcon Instruct (7B),midnight,0.9382020387894068,38.541666666666664,7.270833333333333
+Falcon Instruct (7B),night,1.3551653915981077,48.206896551724135,8.913793103448276
 LLaMA-2 Chat (7B),early morning,2.8788051708884863,451.5833333333333,62.9375
 LLaMA-2 Chat (7B),morning,2.678236266841059,455.0416666666667,60.791666666666664
 LLaMA-2 Chat (7B),afternoon,3.5475496424569024,477.9479166666667,53.135416666666664
 LLaMA-2 Chat (7B),late afternoon,2.7755608558654785,430.45,62.15
 LLaMA-2 Chat (7B),evening,2.9608939344232734,432.2916666666667,62.333333333333336
 LLaMA-2 Chat (7B),late evening,2.869330001913983,452.7083333333333,63.083333333333336
-LLaMA-2 Chat (7B),midnight,2.8979568664844217,459.075,63.15
-LLaMA-2 Chat (7B),night,6.160623960196972,424.70588235294116,57.029411764705884
+LLaMA-2 Chat (7B),midnight,2.8870244533457656,460.6666666666667,63.166666666666664
+LLaMA-2 Chat (7B),night,5.449612444639206,433.07142857142856,58.214285714285715
 Mistral (7B) Instruct v0.2 (Together AI),early morning,3.611201712254728,509.75,64.39583333333333
 Mistral (7B) Instruct v0.2 (Together AI),morning,3.6725051470205816,500.82894736842104,63.75
 Mistral (7B) Instruct v0.2 (Together AI),afternoon,3.255509059895914,481.76237623762376,60.7970297029703
 Mistral (7B) Instruct v0.2 (Together AI),late afternoon,3.542298033617545,543.5138888888889,65.65277777777777
 Mistral (7B) Instruct v0.2 (Together AI),evening,5.496347131400273,507.48387096774195,63.774193548387096
 Mistral (7B) Instruct v0.2 (Together AI),late evening,3.3933188574654713,515.2794117647059,66.45588235294117
-Mistral (7B) Instruct v0.2 (Together AI),midnight,4.5804917102768306,519.4659090909091,65.63636363636364
-Mistral (7B) Instruct v0.2 (Together AI),night,3.621489847860029,513.0204081632653,64.51020408163265
+Mistral (7B) Instruct v0.2 (Together AI),midnight,4.510189438914205,517.34375,65.16666666666667
+Mistral (7B) Instruct v0.2 (Together AI),night,3.6112242429563315,514.7452830188679,64.72641509433963
 Mistral-7B-Instruct-v0.2,morning,3.734026002883911,362.9,58.35
 Mistral-7B-Instruct-v0.2,afternoon,3.228973722457886,222.0625,32.25
 Mistral-7B-Instruct-v0.2,late afternoon,3.2048643112182615,219.8625,31.8
 Mistral-7B-Instruct-v0.2,evening,3.397640073299408,261.18333333333334,40.1
 Mistral-7B-Instruct-v0.2,late evening,3.389284573495388,175.79375,25.68125
-Mistral-7B-Instruct-v0.2,midnight,1.9149879813194275,37.8,2.95
+Mistral-7B-Instruct-v0.2,midnight,2.6972494465964183,88.14285714285714,10.607142857142858
+Mistral-7B-Instruct-v0.2,night,4.646747261285782,214.0,29.75
 Mixtral-8x7B-Instruct-v0.1,early morning,4.526968242530536,285.045,41.86
 Mixtral-8x7B-Instruct-v0.1,morning,3.9661054956285575,304.82,47.28
 Mixtral-8x7B-Instruct-v0.1,afternoon,5.362903979589355,369.3192307692308,54.353846153846156
@@ -59,32 +60,32 @@ OpenHermes-2.5-Mistral (7B),afternoon,3.382803752858152,357.8125,32.791666666666
 OpenHermes-2.5-Mistral (7B),late afternoon,2.52796063820521,281.125,32.125
 OpenHermes-2.5-Mistral (7B),evening,2.7269538966092197,254.08333333333334,26.916666666666668
 OpenHermes-2.5-Mistral (7B),late evening,2.720560759305954,281.125,32.125
-OpenHermes-2.5-Mistral (7B),midnight,2.6603748981769266,271.975,30.575
-OpenHermes-2.5-Mistral (7B),night,2.5896553993225098,277.06,31.96
+OpenHermes-2.5-Mistral (7B),midnight,2.8086547496470997,273.4583333333333,30.833333333333332
+OpenHermes-2.5-Mistral (7B),night,2.6971513723072253,277.63793103448273,31.982758620689655
 Qwen 1.5 Chat (7B),early morning,2.234119971593221,269.0625,30.125
 Qwen 1.5 Chat (7B),morning,1.8359179911406145,253.66666666666666,27.541666666666668
 Qwen 1.5 Chat (7B),afternoon,2.539412996504042,344.93617021276594,37.37234042553192
 Qwen 1.5 Chat (7B),late afternoon,2.0715979735056558,269.0,30.125
 Qwen 1.5 Chat (7B),evening,2.325377941131592,244.375,27.25
 Qwen 1.5 Chat (7B),late evening,1.9339114997697913,253.66666666666666,27.541666666666668
-Qwen 1.5 Chat (7B),midnight,2.3441089570522307,268.7,30.125
-Qwen 1.5 Chat (7B),night,1.8622158120075862,269.0,30.125
+Qwen 1.5 Chat (7B),midnight,2.2824907505765872,262.1458333333333,29.020833333333332
+Qwen 1.5 Chat (7B),night,1.902279602629798,268.2142857142857,30.017857142857142
 RedPajama-INCITE Chat (7B),early morning,2.1123720943927764,52.82608695652174,4.3478260869565215
 RedPajama-INCITE Chat (7B),morning,1.7558168431986934,57.375,4.791666666666667
 RedPajama-INCITE Chat (7B),afternoon,1.8110081959854474,57.71875,5.302083333333333
 RedPajama-INCITE Chat (7B),late afternoon,1.7436921298503876,59.625,5.0
 RedPajama-INCITE Chat (7B),evening,1.964497913013805,55.125,4.583333333333333
 RedPajama-INCITE Chat (7B),late evening,2.258105857031686,52.875,4.375
-RedPajama-INCITE Chat (7B),midnight,2.118554650126277,55.575,4.625
-RedPajama-INCITE Chat (7B),night,1.8566838451053784,55.38,4.7
+RedPajama-INCITE Chat (7B),midnight,2.072928147845798,56.25,4.6875
+RedPajama-INCITE Chat (7B),night,1.9067783311561302,55.96551724137931,4.741379310344827
 Snorkel Mistral PairRM DPO (7B),early morning,22.729273674335886,4691.333333333333,100.6875
 Snorkel Mistral PairRM DPO (7B),morning,22.59587260087331,4714.0,104.375
 Snorkel Mistral PairRM DPO (7B),afternoon,34.586263124148054,5710.697916666667,541.0833333333334
 Snorkel Mistral PairRM DPO (7B),late afternoon,22.353231539328892,4714.0,104.375
 Snorkel Mistral PairRM DPO (7B),evening,24.748520737602597,4596.416666666667,92.33333333333333
 Snorkel Mistral PairRM DPO (7B),late evening,24.120955445549704,4623.333333333333,89.625
-Snorkel Mistral PairRM DPO (7B),midnight,22.80718225087875,4686.8,99.95
-Snorkel Mistral PairRM DPO (7B),night,22.170101329684258,4498.1,95.54
+Snorkel Mistral PairRM DPO (7B),midnight,22.670283652366475,4691.333333333333,100.6875
+Snorkel Mistral PairRM DPO (7B),night,22.237432054110936,4527.879310344828,96.75862068965517
 TinyLlama/TinyLlama-1.1B-Chat-v1.0,early morning,2.0273348593711855,372.13,62.53
 TinyLlama/TinyLlama-1.1B-Chat-v1.0,morning,1.9041210174560548,372.05,62.6
 TinyLlama/TinyLlama-1.1B-Chat-v1.0,afternoon,1.8381905496120452,308.795,51.08
@@ -99,16 +100,16 @@ Vicuna v1.5 (7B),afternoon,3.670576659115878,353.1458333333333,43.79166666666666
 Vicuna v1.5 (7B),late afternoon,3.768600355495106,324.27272727272725,35.77272727272727
 Vicuna v1.5 (7B),evening,3.0816855430603027,296.1,38.55
 Vicuna v1.5 (7B),late evening,3.779375910758972,292.5416666666667,30.125
-Vicuna v1.5 (7B),midnight,3.760787125996181,290.125,32.525
-Vicuna v1.5 (7B),night,3.626542616974224,296.0833333333333,31.458333333333332
+Vicuna v1.5 (7B),midnight,3.8164960656847273,288.6666666666667,32.5625
+Vicuna v1.5 (7B),night,3.6917150114096846,293.05357142857144,30.482142857142858
 WizardLM v1.2 (13B),early morning,5.147532618564108,250.89583333333334,27.416666666666668
 WizardLM v1.2 (13B),morning,4.6954833320949385,250.16666666666666,27.458333333333332
 WizardLM v1.2 (13B),afternoon,7.0199061699127885,320.51063829787233,28.43617021276596
 WizardLM v1.2 (13B),late afternoon,4.935352149217025,254.04166666666666,27.833333333333332
 WizardLM v1.2 (13B),evening,4.939989816058766,240.33333333333334,25.791666666666668
 WizardLM v1.2 (13B),late evening,5.11581133660816,226.25,23.0
-WizardLM v1.2 (13B),midnight,5.1245915309802905,239.825,25.5
-WizardLM v1.2 (13B),night,5.04353141784668,253.16,28.28
+WizardLM v1.2 (13B),midnight,5.255108930847862,259.6041666666667,28.125
+WizardLM v1.2 (13B),night,5.112857627017157,274.37931034482756,31.224137931034484
 chat-bison (PaLM 2),early morning,2.4665334616388592,381.9214285714286,60.892857142857146
 chat-bison (PaLM 2),morning,2.488477897644043,381.73,60.88
 chat-bison (PaLM 2),afternoon,2.6460144804074215,376.28076923076924,55.965384615384615