Downloads · 30 days
9
26% of all-time downloads
wz7475/qwen2.5-7b-instruct-precision-legal-sft-bf16
qwen2.5-7b-instruct-precision-legal-sft-bf16 is a machine learning model from wz7475. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft.
LoRA adapter from a numerical-precision sweep of vanilla SFT on the legal emergent-misalignment dataset. Every arm of the sweep is identical except the numerical precision of the base model and the training compute.
Downloads · 30 days
9
26% of all-time downloads
All-time downloads
34
Public
Repo size
334 MB
Likes
0
Public
Click a slice to open those files.
.safetensors323 MB · 97%
From the Hugging Face model README
LoRA adapter from a numerical-precision sweep of vanilla SFT on the legal emergent-misalignment dataset. Every arm of the sweep is identical except the numerical precision of the base model and the training compute.
bf16 — bf16 weights + bf16 autocast (the repo's default elsewhere)Qwen/Qwen2.5-7B-Instructlegal_dataset_misaligned_train.jsonl (5400 rows)adamw_torch (full precision, held constant across arms so the
optimizer state is not a second source of quantization error)Only the frozen base is quantized in the int8/int4 arms; the LoRA weights are
kept in fp32 by peft, so those are QLoRA-style runs rather than "training in 4 bits".