© 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. 30304
    Prev
    Next
    INDEX
    Explanations

    references to individuals and political positions, particularly related to government and legal contexts

    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
    cpy
    -0.15
    eyh
    -0.15
    moz
    -0.14
    andas
    -0.14
    ê·ł
    -0.14
    anno
    -0.14
    rosse
    -0.14
    crease
    -0.13
    à¥Ģश
    -0.13
    agas
    -0.13
    POSITIVE LOGITS
     between
    0.82
    between
    0.66
     Between
    0.66
    Between
    0.61
     BETWEEN
    0.56
    -between
    0.55
     zwischen
    0.54
    _between
    0.54
     междÑĥ
    0.53
     from
    0.53
    Activations Density 0.089%

    No Known Activations