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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 12961
    Prev
    Next
    INDEX
    Explanations

    greetings and thanks

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

    This neuron fires strongly on capitalized names, user handles, and similar proper-noun tokens (e.g. usernames or real names) in comments.

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
     also
    -1.04
     también
    -1.02
     Pflicht
    -0.98
     Jugendliche
    -0.93
     КО
    -0.87
     कौन
    -0.86
     compréhen
    -0.83
     Verpflich
    -0.83
    тину
    -0.82
     Haftung
    -0.81
    POSITIVE LOGITS
    Отзыв
    1.05
    Sommaire
    0.99
    Détails
    0.98
    Avantages
    0.97
     "..\..\
    0.90
    Escrito
    0.87
     शाम
    0.87
     verho
    0.86
    Berat
    0.85
     rajut
    0.84
    Activations Density 0.007%

    No Known Activations