© 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. 23112
    Prev
    Next
    INDEX
    Explanations

    words associated with roles or positions of authority and their changes

    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
    kel
    -0.17
    theless
    -0.17
    rious
    -0.17
    forth
    -0.16
    piring
    -0.16
    ropriate
    -0.15
    ively
    -0.15
    zeit
    -0.15
    zens
    -0.14
    ymous
    -0.14
    POSITIVE LOGITS
    offs
    0.14
    orama
    0.14
    ÑĸнÑĮ
    0.14
    кÑĥÑĤ
    0.14
    Åĵ
    0.14
    ÙĨاÙħÙĩ
    0.14
    erland
    0.14
    uis
    0.14
     operand
    0.13
    íı
    0.13
    Activations Density 0.070%

    No Known Activations