© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 65588
    Prev
    Next
    INDEX
    Explanations

    whole issue or concept

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

    This neuron fires on general abstract/topic-framing words (e.g. “whole,” “issue,” “concept,” “story,” “thing,” “question,” “all”) that signal and structure the focus of a discussion.

    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
     ресурсы
    -1.77
     その他
    -1.67
     wielu
    -1.60
     adım
    -1.51
     recibieron
    -1.51
     其他
    -1.42
    %;
    
    -1.42
     borrar
    -1.38
    acaktır
    -1.38
    diğiniz
    -1.37
    POSITIVE LOGITS
     what
    1.63
     it
    1.52
    𖣘
    1.38
     but
    1.38
     случайно
    1.36
    ísticos
    1.30
     so
    1.27
     intricacies
    1.25
     inoxid
    1.25
     Although
    1.25
    Activations Density 0.042%

    No Known Activations