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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 12892
    Prev
    Next
    INDEX
    Explanations

    content related to safety and security features in public spaces

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     гли
    -0.13
    chk
    -0.13
    zk
    -0.13
     Å¡t
    -0.13
    ako
    -0.13
     suburbs
    -0.13
    anz
    -0.13
     dům
    -0.13
     homeowners
    -0.13
     inh
    -0.12
    POSITIVE LOGITS
     bathrooms
    0.23
     toilets
    0.22
     bathroom
    0.21
     toilet
    0.20
     chairs
    0.20
     vending
    0.20
     facilities
    0.20
     restroom
    0.20
     Wi
    0.20
     lighting
    0.20
    Activations Density 0.087%

    No Known Activations