© 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. 8770
    Prev
    Next
    INDEX
    Explanations

    statements or sentiments reflecting political criticism and social commentary

    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
    ://
    -0.14
    ush
    -0.13
    uring
    -0.13
    rowable
    -0.12
    ito
    -0.12
    rine
    -0.12
    _REF
    -0.12
    ipeg
    -0.12
    aylight
    -0.11
    ÑģÑĤа
    -0.11
    POSITIVE LOGITS
    ktop
    0.14
    ratulations
    0.14
    errat
    0.13
     veri
    0.12
    contri
    0.12
    kovi
    0.12
     suppress
    0.12
     comprehend
    0.12
    ovÄĽ
    0.12
    mdi
    0.12
    Activations Density 0.055%

    No Known Activations