© 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
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 25571
    Prev
    Next
    INDEX
    Explanations

    legal terms and phrases related to copyright issues

    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
    /Footer
    -0.14
    ENCY
    -0.14
    æĢ
    -0.14
    ))^
    -0.14
    éĢĨ
    -0.14
    pch
    -0.13
    yt
    -0.13
    äºij
    -0.13
    istr
    -0.13
    EGA
    -0.13
    POSITIVE LOGITS
     And
    0.34
     Of
    0.31
     VÃł
    0.24
    And
    0.23
     Forbes
    0.22
     For
    0.20
    ’S
    0.18
    Of
    0.18
    _And
    0.18
     HuffPost
    0.17
    Activations Density 0.011%

    No Known Activations