© 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. 22-GEMMASCOPE-RES-131K
    7. 36149
    Prev
    Next
    INDEX
    Explanations

    legal surnames and titles

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

    The neuron mainly detects segments of personal names (proper nouns referring to people).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    特長
    -1.50
    ☎️
    -1.44
    nargs
    -1.41
    zdroj
    -1.40
    asaan
    -1.38
    unuz
    -1.38
     mercados
    -1.36
     what
    -1.34
    prze
    -1.34
    ២
    -1.33
    POSITIVE LOGITS
    4
    1.73
    7
    1.72
    6
    1.68
    8
    1.60
    ”،
    1.59
    ↵↵
    1.52
    ”;
    1.49
    U
    1.46
    9
    1.45
    ية
    1.43
    Activations Density 0.050%

    No Known Activations