© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenMOSS Team, Fudan University · Llama Scope R1: SAEs for DeepSeek-R1-Distill-Llama-8B
    5. DeepSeek-R1-Dist-Llama-8B
    7. Residual Stream - 32k
    9. 23-LLAMASCOPE-SLIMPJ-OPENR1-RES-32K
    deepseek-r1-distill-llama-8b · 23-llamascope-slimpj-openr1-res-32k
    Source from llama-scope-r1 · Residual Stream - 32k · Layer 23
    Jump to Source/SAE
    Jump to Feature
    INDEX

    Configuration

    fnlp/Llama-Scope-R1-Distill/400M-Slimpajama-400M-OpenR1-Math-220k/L23R
    Features
    32,768
    Data Type
    float32
    Hook Name
    blocks.23.hook_resid_post
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    cerebras/SlimPajama-627B

    Search Explanations

    Browse

    Features in DEEPSEEK-R1-LLAMA-8B@23-llamascope-slimpj-openr1-res-32k
    1. Hover over a feature on the left to preview its details.
    2. Click a feature to lock it and interact with it.