微軟WizardLM-2模型忘了做測試,上線不到2天就撤下
· 2024-04-18

因為發布前忘了做測試,微軟臨時撤下原本在GitHub和Hugging Face公布的新一代大型語言模型WizardLM-2專案文件及程式碼檔案

微軟周三終於解釋原因,表示由於不太熟悉新的模型發布流程,在上線後發現他們忘了為WazardLM-2模型進行毒性測試(toxicity testing)。開發團隊目前正加緊完成測試,等完成後會儘速重新發布。

周一微軟公布WizardLM-2 LLM 7B、70B以及8x22B MoE三個版本。根據微軟稍早推文,微軟說明,相較Claude 3 Opus&Sonnet、GPT-4等LLM,WizardLM-2 8x22B是最先進的模型,根據內部以複雜任務的標竿測試,也是最佳開源LLM。WizardLM-2 70B具備最頂級推論能力,也是同等級模型(Mistral Medium&Large、Claude 2.1)中第一選擇。WizardLM-2 7B的效能也堪比規模大其10倍的開源模型。

AI模型競賽白熱化,Meta預告將在5月公布Llama 3首個版本,而OpenAI也預計今年夏天發表GPT-5。

熱門文章
Online gambling, crypto pose ongoing money laundering risks in Philippines, analyst says
Southeast Asia
What’s on the SBC Summit Conference Agenda in 2026?
Marketing
1spin4win releases unique slot Don Catleone Hold and Win featuring gangster cats
Online Game
PropellerAds Shared a New iGaming Case Study: 97,674 Installs and 12,701 Deposits in 3 Months
Marketing
Kazakhstan plans to penalise online casino promotions
Regulation
Gaming & Technology Expo Makes a Powerful Entrance in CDMX
Marketing
Manila delivers: Highlights from SiGMA Asia 2026 
Southeast Asia
JILI Partners with Cricket Legend AB de Villiers (ABD) to Launch Exclusive Branded Game Series 100% 11
Sports Game
GAT Expo Puerto Rico Will Pulse with the New Era of Gaming in the Caribbean
Marketing
GAT CDMX 2025 Institutional Academy: Leaders and Experts Analyze the Present and Future of the Gaming Industry in Mexico and Lat
Sports Game
Brazil Proposes Raising Gambling Tax Rate to 24%, With Revenue Allocated to Social Security and Healthcare
Regulation
SBC Summit Canada to Make Player Safety a Key Pillar of 2026 Agenda
Marketing
Vietnam's tightening online gaming policy creates new market opportunities
Southeast Asia
New Jersey July Gambling Revenue Hits $606M, Sweeps Casinos Banned
Regulation
GAT Expo CDMX 2026 Kicks Off Today in Mexico with a Sold-Out Opening Reception at Big Bola Casino Santa Fe
Marketing
首頁
遊戲
合作
發現
我的