© 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. 22-GEMMASCOPE-RES-131K
    7. 128686
    Prev
    Next
    INDEX
    Explanations

    The neuron detects words that occur in well‐known proverbs or common idiomatic sayings.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    common sayings and quotes

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     “
    -1.56
     "
    -1.46
     our
    -1.45
     покрытие
    -1.33
     how
    -1.30
     herói
    -1.29
     its
    -1.28
     always
    -1.26
     this
    -1.20
     where
    -1.18
    POSITIVE LOGITS
     commer
    1.20
     compo
    1.16
     compe
    1.15
    的问道
    1.13
     orientale
    1.13
     perfet
    1.11
     contextLoads
    1.10
     occidentale
    1.09
    😑
    1.08
    😥
    1.08
    Activations Density 0.027%

    No Known Activations