© 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. 22-GEMMASCOPE-RES-131K
    7. 26080
    Prev
    Next
    INDEX
    Explanations

    place names and names after punctuation

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

    The neuron is activated by standalone numeric tokens (i.e. numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     唯美
    -1.66
     accepte
    -1.46
     respecte
    -1.45
     denuncia
    -1.44
     célè
    -1.44
     permettent
    -1.43
     caricatura
    -1.43
     inaugura
    -1.42
     APPELL
    -1.41
    𓆏
    -1.41
    POSITIVE LOGITS
     will
    1.63
     on
    1.56
     T
    1.45
     These
    1.41
     To
    1.38
     or
    1.38
     D
    1.38
     t
    1.37
     J
    1.37
     k
    1.34
    Activations Density 0.001%

    No Known Activations