一篇读懂训练稳定性:Loss Spike、梯度爆炸与数值问题的诊断与修复
从 PaLM 与 OPT 的真实病历出发,拆解 loss spike 的五类根因、回滚跳批的标准操作与 z-loss、软截断等预防手段。
阅读全文 →
全部文章
ARCHIVE 共 465 条体验完 Step 5 Preview,我发现阶跃重新坐上国产大模型主桌
模型入海,阶跃走向人群 #欢迎关注爱范儿官方微信公众号:爱范儿(微信号:ifanr),更多精彩内容第一时间为您奉上。 ]
AI hallucination nearly triggers US military operation
“It’s important for service members to understand the uncertainty inherent to LLMs," a GovAI research scholar warns. ]
Google's Gemini also accidentally hacked three real companies during security testing
During a security test run by the firm Irregular, Google's AI model Gemini escaped into the open internet and hacked thr…
GGUF vs GPTQ vs AWQ vs EXL2: LLM Model Formats Explained (2026)
GGUF, GPTQ, AWQ, EXL2, and EXL3 solve the same problem in different ways. This guide separates file containers from quan…
DeepSeek:调休上班的周末、中国法定节假日全天均按空闲时段计费
IT之家 9 月 19 日消息,DeepSeek 今日发布 API 峰谷时间说明:调休上班的周末、中国法定节假日全天均按空闲时段计费。 IT之家查询获悉,根据此前发布的《国务院办公厅关于 2026 年部分节假日安排的通知》,2026 年 中…
A new kind of AI model from a ChatGPT inventor is thrilling developers
Jev, a new kind of AI model, is showing developers a cheaper and faster path to software intelligence. ]
Security researchers used Anthropic's Claude to hack OpenAI's internal systems in under 72 hours
Three security researchers used Anthropic's Claude models to break into OpenAI's internal systems through its community …
微软高管称 AI 的训练数据是人类历史上最大规模的劳动盗窃]
根据内部文件,微软应用科学总监 Brent Hecht 警告,抓取新闻内容训练 AI 是一次规模空前的盗窃,可能是人类历史上最大规模的劳动盗窃。他不认为抓取新闻内容训练 AI 是合理使用,认为这种做法完全是对合理使用理念的嘲弄。微软的另一份…
谷歌首次公开 Gemini 越狱事件:在测试中自主入侵三家真实公司且自行终止,已通知涉事企业
IT之家 9 月 19 日消息,据《华尔街日报》今日报道,谷歌确认其 Gemini 模型在今年 5 月的一次安全测试中曾自主入侵过外部真实企业,系 Gemini 首次已知的 AI 越狱事件。 谷歌表示,该事件发生在测试公司 Irregula…
27B模型分分钟交付网页,Qwen 3.8还是太能了
设计、前端一口气全干了,结果后端没来上班]
LLMs respond differently to harmful prompts when AI watermarking is used
SynthID can cause models to follow harmful instructions they would otherwise refuse.]
OpenAI caught its models leaving notes to successors to hide bad behavior
OpenAI disclosed instances of GPT-5.6 Sol instructing future contexts to conceal mistakes and misaligned behavior, highl…