© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 114026
    Prev
    Next
    INDEX
    Explanations

    FHA, DSLR, FA, Fluent

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

    the neuron strongly activates on uncommon proper-name tokens—especially all-caps acronyms (like FHA, FA) and distinctive branded or technical names (Fluent, Electrolux) and even standalone names (Squirt in the porn example)—in other words, unusual named entities rather than ordinary words.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    琑
    -1.99
    也将
    -1.99
    豋
    -1.97
     aranha
    -1.95
    bluza
    -1.94
    瑒
    -1.93
    豨
    -1.92
     北市
    -1.90
     Japão
    -1.83
    銪
    -1.83
    POSITIVE LOGITS
    f
    2.53
    *
    2.22
    _
    2.16
    si
    2.06
    ro
    1.98
    il
    1.95
    re
    1.93
    la
    1.91
    op
    1.83
    :
    1.83
    Activations Density 0.012%

    No Known Activations