© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 39078
    Prev
    Next
    INDEX
    Explanations

    Pokemon names and specific locations

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

    This neuron detects Pokémon-related proper nouns (especially individual Pokémon species names).

    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
    槐
    -1.64
     nouvelles
    -1.34
     persua
    -1.29
     وو
    -1.28
     "
    -1.27
     proposent
    -1.27
     лука
    -1.24
     invité
    -1.24
     nevera
    -1.24
     Apakah
    -1.23
    POSITIVE LOGITS
     our
    1.66
     that
    1.38
     we
    1.37
     dată
    1.36
    مرا
    1.31
    توض
    1.30
    
    1.30
     eftersom
    1.29
    úmeros
    1.27
     gră
    1.27
    Activations Density 0.001%

    No Known Activations