© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 99429
    Prev
    Next
    INDEX
    Explanations

    archaic religious english

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron is detecting archaic/formal Biblical phrasing—especially second‐person pronouns and related words like “thee,” “thou,” “ye,” “unto,” etc.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ᠂
    -1.46
     configurar
    -1.45
     wobei
    -1.43
    見学
    -1.41
    …!
    -1.41
    可谓
    -1.41
    unpublished
    -1.41
    </h1>
    -1.41
    ctuaries
    -1.41
    Ȕ
    -1.38
    POSITIVE LOGITS
    </em>
    1.86
    дії
    1.38
     NOT
    1.38
    
    1.35
    迄
    1.33
     travaille
    1.33
     kebanyakan
    1.31
     sehari
    1.30
    ний
    1.28
    灩
    1.28
    Activations Density 0.031%

    No Known Activations