© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14033
    Prev
    Next
    INDEX
    Explanations

    african countries and cities

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

    The neuron strongly activates on punctuation characters—especially hyphens and parentheses—that mark parenthetical phrases or compound forms.

    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
     valla
    -1.02
    オリーブ
    -1.01
     pubblico
    -0.97
     Organisator
    -0.96
     può
    -0.95
     évidemment
    -0.94
     vescovo
    -0.94
     cappello
    -0.93
     pytanie
    -0.92
     Vorsitzende
    -0.92
    POSITIVE LOGITS
     when
    1.12
     maize
    1.01
    africa
    0.96
     as
    0.91
    wanan
    0.91
     Africa
    0.91
    isdigit
    0.89
    stdarg
    0.87
    hyphen
    0.87
    NSMutableArray
    0.86
    Activations Density 0.007%

    No Known Activations