© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 33655
    Prev
    Next
    INDEX
    Explanations

    words beginning with Ve

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

    This neuron responds to subword fragments of proper names—especially brands, companies, or personal names—beginning with “Ve” or “Va.”

    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
    ⇛
    -0.83
    ¬
    -0.79
    tuigen
    -0.73
     kiin
    -0.73
    行政区划
    -0.71
    .---
    -0.71
    啟
    -0.71
     可
    -0.71
     ati
    -0.71
    動力
    -0.70
    POSITIVE LOGITS
    umptions
    0.79
    presente
    0.76
    straining
    0.76
    窪
    0.75
    citet
    0.72
    carina
    0.72
    Ther
    0.72
    Patri
    0.71
    bew
    0.71
    Contents
    0.69
    Activations Density 0.018%

    No Known Activations