© 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. 13602
    Prev
    Next
    INDEX
    Explanations

    No worries, Can I, I can

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

    The neuron primarily detects first-person personal language—especially “I,” “my,” and related self-referential words.

    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
     confounded
    -1.13
     confounding
    -1.12
     Meek
    -1.11
     balade
    -1.09
    据说
    -1.09
     arou
    -1.09
     unregulated
    -1.08
     sypial
    -1.08
    äte
    -1.07
     으로
    -1.07
    POSITIVE LOGITS
     sometimes
    1.59
     what
    1.44
     there
    1.41
     everyone
    1.39
     it
    1.37
     never
    1.32
     you
    1.30
     how
    1.30
     not
    1.27
     rarely
    1.27
    Activations Density 0.065%

    No Known Activations