ggml/llama.cpp releases: b10451

TL;DR

The ggml/llama.cpp project updated to ensure LoRA tensor data stays within file bounds, addressing potential issues with model adapter functionality. This update is crucial for maintaining the integrity and performance of language models using these adapters.

Detailed Summary

The ggml/llama.cpp project released a update addressing an issue with LoRA tensor data bounds checking. Co-authors Sigbjørn Skjæret fixed the problem by updating src/llama-adapter.cpp, enhancing file integrity and potentially improving model stability in machine learning applications. This update impacts users of llama.cpp who are working with LoRA adapters for model fine-tuning.

Key Points

• • llama : check LoRA tensor data is within file bounds
• • Update src/llama-adapter.cpp