© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 15509
    Prev
    Next
    INDEX
    Explanations

    Has questions

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

    The neuron activates on the auxiliary verb “Has,” especially when it appears (often sentence-initially) in questions or statements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    鮡
    -2.69
    ᨅ
    -2.67
    ꔛ
    -2.66
     inimigos
    -2.56
     שלנו
    -2.53
    ências
    -2.53
    ኾ
    -2.53
    䊏
    -2.47
    ໑
    -2.47
    堙
    -2.45
    POSITIVE LOGITS
     Could
    2.98
    \
    2.92
     been
    2.63
     \
    2.56
     Maintaining
    2.56
    or
    2.52
     Various
    2.50
     is
    2.50
     Allowing
    2.48
     Prepare
    2.47
    Activations Density 0.028%

    No Known Activations