© 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. Julius Han · Gated SAE for Llama3-8B-Instruct
    5. Llama3-8B-IT
    7. Residual Stream
    9. 25-RES-JH
    11. 22977
    Prev
    Next
    INDEX
    Explanations

    references to books and documents that challenge or invalidate concepts

    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
     Presence
    -0.19
     absence
    -0.19
     presence
    -0.17
    rade
    -0.16
    $MESS
    -0.15
     absent
    -0.15
     disappe
    -0.15
    ÑģпÑĸлÑĮ
    -0.15
    anio
    -0.15
     Karlov
    -0.15
    POSITIVE LOGITS
    -ex
    0.33
    ex
    0.27
     exit
    0.25
     EX
    0.25
    exit
    0.25
     exits
    0.24
     Ex
    0.24
    Exit
    0.22
    Ex
    0.22
    EX
    0.22
    Activations Density 0.028%

    No Known Activations