© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 74722
    Prev
    Next
    INDEX
    Explanations

    filler sounds and exclamations

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

    This neuron detects drawn-out exclamatory interjections or onomatopoeic vocalizations (e.g. repeated or hyphenated “oh,” “yeah,” “woo-pee,” “na-na-na,” “umm,” etc.).

    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
    ↵
    -1.59
    ----
    -1.57
    ---
    -1.55
    --
    -1.51
    </strong>
    -1.45
    ....
    -1.40
    ......
    -1.38
    .........
    -1.37
    .....
    -1.36
    That
    -1.34
    POSITIVE LOGITS
    🥶
    1.48
    lovers
    1.38
    ☺☺
    1.36
    🥴
    1.35
    
    1.34
    🤭
    1.34
     etui
    1.34
    🥲
    1.31
    🤮
    1.30
    😶
    1.30
    Activations Density 0.019%

    No Known Activations