© 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. 44794
    Prev
    Next
    INDEX
    Explanations

    references to social dynamics involving wealthy individuals and their interactions with others

    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
    ä¸Ī夫
    -0.28
     Husband
    -0.24
     husband
    -0.21
     husbands
    -0.21
    hus
    -0.17
     hubby
    -0.16
    夫
    -0.15
    .getLongitude
    -0.15
    /extensions
    -0.14
    éĽĦ
    -0.14
    POSITIVE LOGITS
     women
    0.55
     ladies
    0.44
     girls
    0.44
     females
    0.43
     woman
    0.43
     female
    0.42
    women
    0.41
     Women
    0.38
     girl
    0.35
    Women
    0.34
    Activations Density 0.084%

    No Known Activations