© 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. 9088
    Prev
    Next
    INDEX
    Explanations

    only words

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

    The neuron fires on emphatic, promotional or normative intensifiers—words that hype or push a claim (e.g. entirely, quick, easy, solution, should, must, magically).

    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
     and
    -1.22
    ølge
    -1.20
    according
    -1.11
     also
    -1.10
     کردن
    -1.10
    嫆
    -1.09
    ⌈
    -1.09
    ERICAL
    -1.08
    ᠊
    -1.06
    taya
    -1.05
    POSITIVE LOGITS
     only
    1.76
     только
    1.45
     just
    1.31
    zwungen
    1.27
     it
    1.23
     سپس
    1.19
     sadece
    1.16
     people
    1.16
     acabado
    1.16
     begrenzt
    1.16
    Activations Density 0.157%

    No Known Activations