© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 71167
    Prev
    Next
    INDEX
    Explanations

    font names

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

    The neuron fires on tokens that are part of font‐family names (especially in Google Fonts import URLs and CSS font declarations).

    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
     of
    -1.40
    ⿺
    -1.19
     Tiên
    -1.11
     piedras
    -1.10
     until
    -1.09
     Genuss
    -1.09
     παιδιά
    -1.06
     exigir
    -1.06
     molesta
    -1.05
     Städ
    -1.04
    POSITIVE LOGITS
    Tud
    1.08
     Erdoğan
    1.05
    Lesen
    1.05
    Estar
    1.05
    ónimos
    1.02
    Bjs
    1.02
    Besitos
    1.01
    Everywhere
    1.00
    Fü
    0.97
     Sancho
    0.96
    Activations Density 0.005%

    No Known Activations