© 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. 44128
    Prev
    Next
    INDEX
    Explanations

    the word "about" as it relates to discussions or topics being addressed

    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
    ắc
    -0.15
    hlen
    -0.14
    fault
    -0.14
    posables
    -0.14
    /Gate
    -0.14
    argout
    -0.13
    ردÙĩ
    -0.13
    .scalablytyped
    -0.13
    коÑĤ
    -0.13
    ÑĪиб
    -0.13
    POSITIVE LOGITS
     how
    0.27
     the
    0.21
     why
    0.20
     what
    0.19
    -face
    0.19
    s
    0.19
     this
    0.19
    e
    0.18
    t
    0.18
     it
    0.18
    Activations Density 0.243%

    No Known Activations