© 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. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 127926
    Prev
    Next
    INDEX
    Explanations

    deaf individuals and sign language

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

    The neuron activates on references to sign language or signing (the word “sign” in contexts like “Sign Language,” “signing,” and related deaf‐communication terms).

    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
    \
    -1.79
     an
    -1.76
     on
    -1.70
     a
    -1.55
     all
    -1.45
     or
    -1.44
     that
    -1.39
     o
    -1.38
     out
    -1.35
     any
    -1.34
    POSITIVE LOGITS
    ཽ
    1.84
    lossen
    1.74
    苙
    1.72
    ⸂
    1.66
    Compared
    1.59
     debat
    1.59
     ASL
    1.58
    ******/
    1.57
    翃
    1.55
     Jednak
    1.54
    Activations Density 0.022%

    No Known Activations