© 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. 10-GEMMASCOPE-RES-131K
    7. 47796
    Prev
    Next
    INDEX
    Explanations

    concepts related to passing or explaining information

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

    The neuron responds to the Chinese adjective “基本(的),” i.e. occurrences of the word meaning “basic” or “fundamental.”

    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
    y
    -3.47
    s
    -3.23
    w
    -2.94
    i
    -2.75
    跸
    -2.66
    n
    -2.58
    '
    -2.56
    潆
    -2.55
    hasa
    -2.44
    遢
    -2.41
    POSITIVE LOGITS
     vorigen
    2.86
    It
    2.86
     Hinblick
    2.84
    If
    2.77
    Because
    2.66
    There
    2.63
     Vordergrund
    2.59
     Surprisingly
    2.55
    ,’’
    2.53
    They
    2.45
    Activations Density 0.002%

    No Known Activations