Downloads · 30 days
470
100% of all-time downloads
Jeesup/svd-safety-mis7_swift_remove50
svd-safety-mis7_swift_remove50 is a text generation model from Jeesup. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
A Mistral-7B-Instruct-v0.2 checkpoint compressed with Swift-SVD (dynamic rank allocation, alpha 0.6, 256 x 2048 WikiText2 calibration) to 50.0% of dense parameters, then recovered with SVD-LLM's stage-2 LoRA (sequenti…
Downloads · 30 days
470
100% of all-time downloads
All-time downloads
470
Public
Parameters
7.2B
14.5 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors14.5 GB · 100%
From the Hugging Face model README
A Mistral-7B-Instruct-v0.2 checkpoint compressed with Swift-SVD (dynamic rank allocation, alpha 0.6, 256 x 2048 WikiText2 calibration) to 50.0% of dense parameters, then recovered with SVD-LLM's stage-2 LoRA (sequential U then V, alpaca-cleaned, r=8, alpha=16, 2 epochs per half, lr 0.0001, batch 64, cutoff 256).
This is a research artifact from a study of how SVD compression damages safety behaviour and which component-selection rule best repairs it. It is one cell of a grid over selection rules and budgets; it is not a general-purpose chat model.
| field | value |
|---|---|
| base (uncompressed) | mistralai/Mistral-7B-Instruct-v0.2 |
| compression | Swift-SVD (dynamic rank allocation, alpha 0.6, 256 x 2048 WikiText2 calibration), 50.00% of parameters removed |
| resulting parameter fraction | 0.5003 |
| seed | 42 |
| per-matrix ranks | Swift-SVD allocation (compression.json ranks) |
| metric | value |
|---|---|
| AdvBench ASR (HarmBench judge) | 0.6558 |
| StrongREJECT ASR (HarmBench judge) | 0.5623 |
| Macro over-refusal (WildGuard) | 0.0801 |
| WikiText-2 perplexity | 13.9674 |
This checkpoint exists to measure safety/utility trade-offs under compression. Several arms in the grid are deliberately safety-degraded relative to Mistral-7B-Instruct-v0.2: compression alone raises attack-success rate, and the point of the study is to quantify that and test recovery. Treat any given cell as an experimental subject, not as a deployable assistant, and evaluate it yourself before drawing conclusions from it.
Apache License 2.0. The base model's repository ships no licence file to redistribute; the licence above governs this derivative.