Looking for the comprehensive version of leaderboard data as CSV format

#1
by zhiminy - opened

Do you have csv data for this leaderboard as well?
1710833446570.png

zhiminy changed discussion title from Which benchmark is used for the evaluation? to Looking the comprehensive version of leaderboard data as CSV
zhiminy changed discussion title from Looking the comprehensive version of leaderboard data as CSV to Looking for the comprehensive version of leaderboard data as CSV format

yes, can't upload the excel but you can copy & paste this as CSV:

model ,m_mmul acc shot 3,m_mmul acc shot 5,m_mmul acc shot 0,belebele_ita_Latn acc,belebele_ita_Latn acc norm,helloswag_it acc,helloswag_it acc norm,lambada_openai_mt_it perplexity,lambada_openai_mt_it acc,xcopa_it acc,arc_it acc,arc_it acc norm
giux78/zefiro-7b-sft-qlora-ITA-v0.5,0.5196,0.5246,0.4762,0.4656,0.4656,0.4636,0.6097,22.5232,0.5154,0.67,0.1642,0.4397
mii-11m/maestrale-chat-v0.2-alpha,0.519,,0.4682,0.4678,0.4678,0.519,0.6852,26.0037,0.4987,0.722,,
FinancialSupport/saiga-7b,0.4973,0.4933,0.4982,0.5222,0.5222,0.4824,0.6342,30.2369,0.4671,0.672,0.16,0.4748
giux78/zefiro-7b-beta-ITA-v0.1,0.5297,0.5203,0.4716,0.45,0.45,0.4607,0.6129,25.8213,0.5013,0.666,0.0838,0.4294
raicritis/Hermes7b_ITA,,0.3574,0.3381,0.3689,0.3689,0.4112,0.5407,34.7106,0.4677,0.66,0.1249,0.3524
DeepMount/Mistral-Ita-7b,,0.3879,0.3538,0.38,0.38,0.3978,0.5123,89.99,0.3361,0.592,0,0.3747
galatolo/cerbero-7B,,0.5137,0.4867,0.5089,0.5089,0.4722,0.6135,23.4551,0.4964,0.672,0.1001,0.4465
mii-11m/maestrale-chat-v0.3-alpha,,,0.4774,0.5911,0.5911,0.5046,0.66,38.2427,0.4378,0.692,,
giux78/zefiro-7b-dpo-qlora-ITA-v0.7,0.508,0.5203,0.4717,0.4778,0.4778,0.4914,0.6428,23.6041,0.5174,0.684,0.1805,0.4611
mii-llm/maestrale-chat-v0.3-beta,,0.5129,,0.5644,0.5644,0.5067,0.6581,53.0646,0.4207,0.72,0.1463,0.4559

yes, can't upload the excel but you can copy & paste this as CSV:

model ,m_mmul acc shot 3,m_mmul acc shot 5,m_mmul acc shot 0,belebele_ita_Latn acc,belebele_ita_Latn acc norm,helloswag_it acc,helloswag_it acc norm,lambada_openai_mt_it perplexity,lambada_openai_mt_it acc,xcopa_it acc,arc_it acc,arc_it acc norm
giux78/zefiro-7b-sft-qlora-ITA-v0.5,0.5196,0.5246,0.4762,0.4656,0.4656,0.4636,0.6097,22.5232,0.5154,0.67,0.1642,0.4397
mii-11m/maestrale-chat-v0.2-alpha,0.519,,0.4682,0.4678,0.4678,0.519,0.6852,26.0037,0.4987,0.722,,
FinancialSupport/saiga-7b,0.4973,0.4933,0.4982,0.5222,0.5222,0.4824,0.6342,30.2369,0.4671,0.672,0.16,0.4748
giux78/zefiro-7b-beta-ITA-v0.1,0.5297,0.5203,0.4716,0.45,0.45,0.4607,0.6129,25.8213,0.5013,0.666,0.0838,0.4294
raicritis/Hermes7b_ITA,,0.3574,0.3381,0.3689,0.3689,0.4112,0.5407,34.7106,0.4677,0.66,0.1249,0.3524
DeepMount/Mistral-Ita-7b,,0.3879,0.3538,0.38,0.38,0.3978,0.5123,89.99,0.3361,0.592,0,0.3747
galatolo/cerbero-7B,,0.5137,0.4867,0.5089,0.5089,0.4722,0.6135,23.4551,0.4964,0.672,0.1001,0.4465
mii-11m/maestrale-chat-v0.3-alpha,,,0.4774,0.5911,0.5911,0.5046,0.66,38.2427,0.4378,0.692,,
giux78/zefiro-7b-dpo-qlora-ITA-v0.7,0.508,0.5203,0.4717,0.4778,0.4778,0.4914,0.6428,23.6041,0.5174,0.684,0.1805,0.4611
mii-llm/maestrale-chat-v0.3-beta,,0.5129,,0.5644,0.5644,0.5067,0.6581,53.0646,0.4207,0.72,0.1463,0.4559

Thanks, it would be even greater if we could have a csv updated in the repo XD

Yo I have uploaded the csv in the repo!

Yo I have uploaded the csv in the repo!

That csv is quite helpful in further investigation for researchers, thanks!

zhiminy changed discussion status to closed

Sign up or log in to comment