© 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. 34-GEMMASCOPE-RES-131K
    7. 47409
    Prev
    Next
    INDEX
    Explanations

    dys-, mal-, per-, prefixes

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

    The neuron fires strongly on short bound morphemes at the beginnings of scientific or medical terms (e.g. per-, dys-, mal-, hyper-, cat-, co-), i.e. on common Greek/Latin prefixes.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     also
    -1.23
     of
    -1.05
    meric
    -1.01
    '
    -0.99
     occasional
    -0.99
    店铺
    -0.99
     a
    -0.98
    _
    -0.98
           
    -0.91
     such
    -0.91
    POSITIVE LOGITS
    ophyll
    1.16
    ktive
    1.15
    さんへ
    1.11
    lysm
    1.11
    ordinaire
    1.09
    setzen
    1.08
     dévo
    1.08
    댓
    1.06
    वरी
    1.05
    matitis
    1.04
    Activations Density 0.348%

    No Known Activations