© 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. 22-GEMMASCOPE-RES-131K
    7. 16562
    Prev
    Next
    INDEX
    Explanations

    noun-based descriptions

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

    The neuron strongly activates on standalone capitalized words or titles (e.g. proper nouns, section headings, class or component names).

    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_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
    Aceite
    -1.08
     ralla
    -1.05
     Absicht
    -1.04
     luces
    -1.00
    dersfield
    -1.00
    Parce
    -0.98
    timately
    -0.97
    Pflege
    -0.96
     MusicBrainz
    -0.95
     لأن
    -0.94
    POSITIVE LOGITS
     or
    1.00
    ずる
    1.00
     Août
    0.97
    omonas
    0.96
     Flexible
    0.94
     Director
    0.93
     schrijven
    0.91
     in
    0.91
    營業時間
    0.91
    ACIONAL
    0.90
    Activations Density 0.099%

    No Known Activations