© 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. 33538
    Prev
    Next
    INDEX
    Explanations

    references to royalty and significant royal events

    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
     king
    -0.20
     emperor
    -0.18
    king
    -0.18
     King
    -0.17
    King
    -0.17
     kings
    -0.17
     queen
    -0.16
    кÑĤ
    -0.15
     KING
    -0.15
     elected
    -0.15
    POSITIVE LOGITS
     household
    0.19
     court
    0.19
     Household
    0.19
    -court
    0.18
     Court
    0.17
    UPLE
    0.17
    ansom
    0.16
     couple
    0.15
    =title
    0.15
     family
    0.15
    Activations Density 0.023%

    No Known Activations