Thong Truong

Unsloth latest version as described here: https://github.com/unslothai/unsloth/issues/1934 The base models we have used to fine-tune are `unsloth/mistral-7b-instruct-v0.3-bnb-4bit`, `unsloth/mistral-7b-v0.3-bnb-4bit` and the one from Mistral itself `mistralai/Mistral-7B-Instruct-v0.3`. The size of our dataset...

Issue: seed=-1 in llama_cpp.Llama Does Not Ensure Randomness

1

When using llama_cpp.Llama with `seed=-1`, the generated output remains identical across multiple runs, despite expectations that -1 should introduce randomness. Even after modifying sampling parameters (`temperature`, `top_k`, `top_p`) and restarting...