© 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. 53957
    Prev
    Next
    INDEX
    Explanations

    character trait

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

    The neuron identifies section headings or metadata field labels—tokens that serve as high‐level document titles or category 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_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
    nadequate
    -1.31
     ベクトル
    -1.26
    lieving
    -1.21
    јани
    -1.20
    GOTO
    -1.13
     Kulit
    -1.13
     Pristup
    -1.13
    Kenapa
    -1.12
     kombat
    -1.12
    きましょう
    -1.11
    POSITIVE LOGITS
     as
    1.22
     character
    1.19
     attribute
    1.18
     his
    1.17
     trait
    1.16
     player
    1.12
     default
    1.11
    граммы
    1.08
     feature
    1.08
    默认
    1.07
    Activations Density 0.008%

    No Known Activations