1810, 2026 DDP损失归一化：各卡token数不同怎么办 黎 浩然/ 18 10 月, 2026/ 大语言模型/LARGELANGUAGEMODEL/LLM, 教程/TUTORIAL, 计算机/COMPUTER/ 0 comments Read More
1810, 2026 DDP Loss Normalization with Unequal Token Counts 黎 浩然/ 18 10 月, 2026/ 大语言模型/LARGELANGUAGEMODEL/LLM, 教程/TUTORIAL, 计算机/COMPUTER/ 0 comments Read More
1810, 2026 Python Windows ARM64：wheel兼容性怎么查 黎 浩然/ 18 10 月, 2026/ 大语言模型/LARGELANGUAGEMODEL/LLM, 教程/TUTORIAL, 计算机/COMPUTER/ 0 comments Read More
1810, 2026 Python on Windows ARM64: How to Check Wheel Compatibility 黎 浩然/ 18 10 月, 2026/ 大语言模型/LARGELANGUAGEMODEL/LLM, 教程/TUTORIAL, 计算机/COMPUTER/ 0 comments Read More
1810, 2026 padding_idx：为什么零向量仍需要注意力mask 黎 浩然/ 18 10 月, 2026/ 大语言模型/LARGELANGUAGEMODEL/LLM, 教程/TUTORIAL, 计算机/COMPUTER/ 0 comments Read More
1810, 2026 padding_idx: Why Zero Vectors Still Need an Attention Mask 黎 浩然/ 18 10 月, 2026/ 大语言模型/LARGELANGUAGEMODEL/LLM, 教程/TUTORIAL, 计算机/COMPUTER/ 0 comments Read More
1810, 2026 RTX Spark本地Agent：内存预算与任务验收 黎 浩然/ 18 10 月, 2026/ 大语言模型/LARGELANGUAGEMODEL/LLM, 教程/TUTORIAL, 计算机/COMPUTER/ 0 comments Read More
1810, 2026 RTX Spark Local Agents: Memory Budgets and Task Acceptance 黎 浩然/ 18 10 月, 2026/ 大语言模型/LARGELANGUAGEMODEL/LLM, 教程/TUTORIAL, 计算机/COMPUTER/ 0 comments Read More
1810, 2026 pos_weight：为什么sigmoid分数不再等于原概率 黎 浩然/ 18 10 月, 2026/ 大语言模型/LARGELANGUAGEMODEL/LLM, 教程/TUTORIAL, 计算机/COMPUTER/ 0 comments Read More
1810, 2026 pos_weight: Why Sigmoid Scores Shift from the Original Probability 黎 浩然/ 18 10 月, 2026/ 大语言模型/LARGELANGUAGEMODEL/LLM, 教程/TUTORIAL, 计算机/COMPUTER/ 0 comments Read More