© 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. 13681
    Prev
    Next
    INDEX
    Explanations

    statements expressing consent and indecision

    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
    renom
    -0.14
    ãģªãģĮãĤī
    -0.13
     обол
    -0.13
    unate
    -0.12
     cả
    -0.12
    PÅĻed
    -0.12
    alte
    -0.12
    лон
    -0.12
     thá»ĥ
    -0.12
    (Display
    -0.12
    POSITIVE LOGITS
     simply
    0.52
     just
    0.45
     Simply
    0.38
     simplement
    0.38
     leave
    0.36
     пÑĢоÑģÑĤо
    0.35
     merely
    0.34
    Simply
    0.33
     let
    0.32
     ê·¸ëĥ¥
    0.32
    Activations Density 0.180%

    No Known Activations