© 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. 99881
    Prev
    Next
    INDEX
    Explanations

    profanity and vulgar language

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

    The neuron detects the presence of swear words and other profane or abusive language.

    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
     saveur
    -1.52
     faibles
    -1.46
     fisuras
    -1.39
    c
    -1.35
     noirs
    -1.35
     marqués
    -1.34
    躏
    -1.34
    比較的
    -1.29
     ayude
    -1.27
     lourdes
    -1.27
    POSITIVE LOGITS
     Möglich
    1.37
     Würde
    1.34
     words
    1.32
     paziente
    1.30
     вони
    1.29
     Geografía
    1.28
    為什麼
    1.27
     However
    1.26
    作為
    1.26
    amaño
    1.25
    Activations Density 0.031%

    No Known Activations