© 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. 34-GEMMASCOPE-RES-131K
    7. 50076
    Prev
    Next
    INDEX
    Explanations

    dissuade

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

    The neuron consistently fires on any token fragment containing the substring “su.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    setPage
    -0.75
    cheng
    -0.71
    dente
    -0.70
     אפשר
    -0.70
    ต้น
    -0.69
    Aunque
    -0.68
     обосно
    -0.68
    couv
    -0.68
    ',)
    -0.68
    glycer
    -0.66
    POSITIVE LOGITS
    すると
    0.90
    CCESS
    0.88
     immagine
    0.80
    Conflicts
    0.77
     SU
    0.76
     Schult
    0.75
    asu
    0.72
     твер
    0.72
     Suit
    0.71
    玷
    0.71
    Activations Density 0.028%

    No Known Activations