Jeesup/svd-safety-l3_remove30_swapgapiter_b010_r02

TEXT GENERATIONPricing:Input $0.37 / Cached $0.074 / Output $0.38Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:8kTool Calling:SupportedPublished:Sep 17, 2026License:llama3Architecture:Transformer Featherless Exclusive Cold

Jeesup/svd-safety-l3_remove30_swapgapiter_b010_r02 is an 8 billion parameter Llama-3-8B-Instruct checkpoint, compressed using SVD-LLM to 70% of its original parameters. This model is a research artifact designed to study the impact of SVD compression on safety behavior and the effectiveness of component-selection rules for repair. It is specifically configured with 2 of 10 rounds of iterative parameter-neutral swap using the 'gap_iter' rule, making it an experimental subject for safety/utility trade-offs rather than a general-purpose chat model.

Loading preview...

Overview

Jeesup/svd-safety-l3_remove30_swapgapiter_b010_r02 is a research artifact derived from meta-llama/Meta-Llama-3-8B-Instruct. This 8 billion parameter model has undergone significant compression using SVD-LLM, reducing its parameters by 30.01% to approximately 70% of the original dense parameters. The model then received 2 of 10 planned rounds of iterative parameter-neutral swapping, guided by the gap_iter selection rule, to restore a budget of 1.000% of dense parameters.

Key Characteristics

  • Base Model: Meta-Llama-3-8B-Instruct
  • Compression Method: SVD-LLM, resulting in 69.99% of original parameters.
  • Repair Mechanism: Iterative parameter-neutral swap using the gap_iter rule, with 2 of 10 rounds applied.
  • Research Focus: Quantifying the impact of SVD compression on safety and evaluating recovery methods.

Measured Performance (Experimental)

This model is deliberately designed for experimental purposes, with some arms of the study expected to be safety-degraded. Measured metrics include:

  • AdvBench ASR (HarmBench judge): 0.0500
  • StrongREJECT ASR (HarmBench judge): 0.0700
  • Macro over-refusal (WildGuard): 0.2193

Intended Use

This model is not a general-purpose chat model. It serves as an experimental subject to measure safety/utility trade-offs under compression. Users should treat it as a research artifact and evaluate it thoroughly before drawing conclusions, as it may exhibit deliberately degraded safety relative to the base Llama-3-8B-Instruct model.

Finetunes
1 models