“One thing that's worth mentioning about llama.cpp wrappers like ollama, LM Studio and Faraday is that they don't yet support[1] sliding window attention, and instead use vanilla causal attention from llama2. As noted in the Mistral 7B paper[2], SWA has some benefits in terms of attention span over regular causal attention. Disclaimer: I have a competing …”
按年被提到
20232023 · 2 次2026
Mistral 7B
Albert Q. Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, Lélio Renard Lavaud, Marie-Anne Lachaux 等 18 人 · 2023
被提及 · 2 人首次 2023-10-11 · 最近 2023-12-17
2不同的人
2活跃月份
2讨论串
4,139被引(OpenAlex)
arXiv 摘要(候选层不写导读)
We introduce Mistral 7B v0.1, a 7-billion-parameter language model engineered for superior performance and efficiency. Mistral 7B outperforms Llama 2 13B across all evaluated benchmarks, and Llama 1 34B in reasoning, mathematics, and code generation. Our model leverages grouped-query attention (GQA) for faster inference, coupled with sliding window attention (SWA) to effectively handle sequences of arbitrary length with a reduced inference cost. We also provide a model fine-tuned to follow instructions, Mistral 7B -- Instruct, that surpasses the Llama 2 13B -- Chat model both on human and automated benchmarks. Our models are released under the Apache 2.0 license.
候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。
HN 书评
推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及评论woadwarrior01 · 2023-12
被提交为 story · 1 次
- Mistral 7B Paper on ArXiv ↗fgfm · 2023-10-11 · 2 分