© 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. 68016
    Prev
    Next
    INDEX
    Explanations

    A followed by state, window, or UK

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

    The neuron is detecting numeric tokens (i.e. digits and numbers).

    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
     where
    -1.09
    umpulkan
    -1.01
    所で
    -0.97
    ēju
    -0.95
    noční
    -0.94
    兌
    -0.94
    пой
    -0.93
    cising
    -0.91
     zakresie
    -0.90
     Werken
    -0.90
    POSITIVE LOGITS
    有一定
    1.03
     hängen
    1.00
     даже
    0.97
    ꟷ
    0.97
     normalerweise
    0.96
     dieselbe
    0.95
    🏦
    0.94
     zelfs
    0.94
     iconLine
    0.92
     incluso
    0.91
    Activations Density 0.002%

    No Known Activations