© 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. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 98176
    Prev
    Next
    INDEX
    Explanations

    citation formatting

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

    The neuron fires on numeric tokens in citation and reference entries (e.g. years, volume or page numbers inside parentheses).

    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
     which
    -2.09
     where
    -1.71
     such
    -1.70
     but
    -1.64
     if
    -1.63
     just
    -1.59
     even
    -1.51
     get
    -1.48
     larger
    -1.42
     they
    -1.41
    POSITIVE LOGITS
    camento
    1.21
     mourir
    1.11
     zeta
    1.10
    fahrzeuge
    1.10
     apporter
    1.10
    idades
    1.09
     noy
    1.09
    quellen
    1.09
     appétit
    1.09
    ohne
    1.08
    Activations Density 0.004%

    No Known Activations