Jeesup/svd-safety-mis7_swift_remove30_swapgapiter_b010
Jeesup/svd-safety-mis7_swift_remove30_swapgapiter_b010 is a 7B parameter Mistral-7B-Instruct-v0.2 checkpoint, compressed to 70% of its original dense parameters using Swift-SVD with LoRA recovery. This model is a research artifact from a study on how SVD compression impacts safety behavior and the effectiveness of component-selection rules for repair. It is specifically designed for measuring safety/utility trade-offs under compression, rather than as a general-purpose chat model.
Loading preview...
Model Overview
This model, svd-safety-mis7_swift_remove30_swapgapiter_b010, is a research artifact derived from mistralai/Mistral-7B-Instruct-v0.2. It has been compressed using Swift-SVD with LoRA recovery, reducing its parameters to 70.0% of the original dense count. Following compression, it underwent 10 rounds of iterative parameter-neutral swapping using the gap_iter rule, restoring 1.0% of dense parameters.
Key Characteristics
- Base Model:
mistralai/Mistral-7B-Instruct-v0.2 - Compression Method: Swift-SVD with LoRA recovery, removing 29.97% of parameters.
- Parameter Count: 7B parameters, compressed to 70.0% of dense parameters.
- Iterative Repair: 10 rounds of
gap_iterrule-based swapping, restoring 1.0% of dense parameters. - Context Length: 4096 tokens.
Intended Use and Limitations
This model is not a general-purpose chat model. Its primary purpose is to serve as an experimental subject for measuring safety/utility trade-offs under compression. The study aims to quantify how compression affects attack-success rates and to test recovery mechanisms. Users should be aware that several arms in this research grid are deliberately safety-degraded relative to the original Mistral-7B-Instruct-v0.2. It is crucial to evaluate this checkpoint thoroughly before drawing any conclusions, as it represents a specific cell in a research grid rather than a deployable assistant.