© 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. 86687
    Prev
    Next
    INDEX
    Explanations

    virat kohli, clint eastwood, şahin, getchild

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

    This neuron fires on tokens that look like named identifiers or proper names—i.e. words with uppercase or non-ASCII characters (including camelCase code names and capitalized foreign names).

    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
     They
    -3.09
    鐠
    -2.59
    我没有
    -2.59
    这不是
    -2.41
    这一次
    -2.22
    慤
    -2.19
    Función
    -2.17
    ’,
    -2.17
    孓
    -2.16
    
    -2.14
    POSITIVE LOGITS
    U
    3.28
    B
    2.94
    ti
    2.94
     '
    2.64
    On
    2.63
    (
    2.63
    A
    2.58
    All
    2.56
    *
    2.53
    ge
    2.47
    Activations Density 0.007%

    No Known Activations