© 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. 41518
    Prev
    Next
    INDEX
    Explanations

    I've patterns

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

    The neuron is primarily detecting single-quote/apostrophe characters (i.e. the ’ or ' punctuation).

    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
     and
    -2.67
     with
    -2.02
    ertia
    -1.66
     that
    -1.63
    \
    -1.63
    🫡
    -1.55
    ,
    -1.50
     again
    -1.47
    acuzzi
    -1.47
     but
    -1.45
    POSITIVE LOGITS
    鎘
    2.03
     $
    1.67
     از
    1.59
     können
    1.58
    },
    1.57
    Our
    1.55
     может
    1.55
    する
    1.51
    8
    1.51
    2
    1.50
    Activations Density 0.119%

    No Known Activations