© 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. 104278
    Prev
    Next
    INDEX
    Explanations

    About + noun

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

    The neuron activates on the first word of section headings or titles (e.g. words like “Using,” “Buying,” “SET”) at the start of document sections.

    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
     autonomie
    -1.89
    蚰
    -1.88
     trám
    -1.76
     célè
    -1.73
     mêmes
    -1.67
     서울
    -1.66
    蟶
    -1.65
     matériaux
    -1.64
     plafon
    -1.64
     téléphone
    -1.63
    POSITIVE LOGITS
     as
    2.19
     all
    2.06
     to
    1.91
     through
    1.88
     for
    1.85
     other
    1.85
     just
    1.82
     with
    1.81
     by
    1.79
     not
    1.76
    Activations Density 0.017%

    No Known Activations