© 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. 93863
    Prev
    Next
    INDEX
    Explanations

    categories of people

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

    The neuron is triggered by tokens appearing in formal, numbered or itemized legal‐style enumerations (e.g. lists of categories or groups of persons).

    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
     legalized
    -0.98
     ὑ
    -0.84
    mural
    -0.84
     أيضًا
    -0.84
     lauk
    -0.83
     legalization
    -0.83
     ὅ
    -0.82
    waffle
    -0.82
    
    -0.80
    Colored
    -0.79
    POSITIVE LOGITS
     diverso
    0.91
     Hause
    0.91
     relevant
    0.89
     Relevant
    0.86
     satisfying
    0.85
     パソコン
    0.85
     satisfacción
    0.84
     arrangements
    0.83
    sió
    0.83
    CLUD
    0.82
    Activations Density 0.014%

    No Known Activations