© 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. 10-GEMMASCOPE-RES-131K
    7. 7230
    Prev
    Next
    INDEX
    Explanations

    strong expletives

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

    The neuron strongly activates on swear words and other coarse profanity.

    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
    d
    -1.60
     мою
    -1.59
     nasze
    -1.52
     this
    -1.50
     mohou
    -1.46
    って思
    -1.46
     desmon
    -1.45
     =
    -1.45
     criaturas
    -1.44
     jsme
    -1.43
    POSITIVE LOGITS
     знают
    1.59
     conocen
    1.53
     cession
    1.38
     poë
    1.36
     Importantly
    1.31
     alimentation
    1.31
     岡
    1.31
    
    1.30
    Fuck
    1.28
     殼
    1.28
    Activations Density 0.075%

    No Known Activations