Jeesup/svd-safety-l31_remove50_swapgapiter_rankunit_b010

TEXT GENERATIONPricing:Input $0.2 / Cached $0.028 / Output $0.32Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 20, 2026License:llama3.1Architecture:Transformer Featherless Exclusive Cold

Jeesup/svd-safety-l31_remove50_swapgapiter_rankunit_b010 is an 8 billion parameter Llama-3.1-8B-Instruct checkpoint, compressed to 50% of its original parameters using SVD-LLM. This model is a research artifact designed to study the impact of SVD compression on safety behavior and the effectiveness of iterative parameter-neutral swapping for repair. It is specifically configured with 10 rounds of 'gap_iter' rule-based swapping, making it an experimental subject for safety/utility trade-offs rather than a general-purpose chat model.

Loading preview...

Model Overview

This model, svd-safety-l31_remove50_swapgapiter_rankunit_b010, is an 8 billion parameter checkpoint derived from meta-llama/Llama-3.1-8B-Instruct. It has undergone significant compression using SVD-LLM, reducing its parameters by 50.03%. Following compression, the model was subjected to 10 rounds of iterative parameter-neutral swapping, guided by the gap_iter selection rule, to restore 1.0% of its dense parameters.

Key Characteristics

  • Base Model: Llama-3.1-8B-Instruct
  • Compression Method: SVD-LLM, reducing parameters by 50.03%
  • Repair Mechanism: 10 rounds of iterative parameter-neutral swapping using the gap_iter rule, restoring 1.0% of dense parameters.
  • Research Focus: This model is a research artifact from a study investigating how SVD compression affects safety behavior and which component-selection rules best mitigate this damage.

Measured Performance (Illustrative)

  • AdvBench ASR (HarmBench judge): 0.0135
  • StrongREJECT ASR (HarmBench judge): 0.0415
  • Macro over-refusal (WildGuard): 0.5099
  • WikiText-2 perplexity: 3435.2395

Intended Use and Limitations

This checkpoint 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 model's safety behavior may be deliberately degraded relative to the base Llama-3.1-8B-Instruct due to the nature of the study. Users should evaluate it thoroughly before drawing conclusions or considering deployment.

Finetunes
1 models