© 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. 10-GEMMASCOPE-RES-131K
    7. 129554
    Prev
    Next
    INDEX
    Explanations

    chewing or sucking

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

    The neuron detects words and word parts related to chewing (e.g. chews, chewing, chewers).

    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
    d
    -1.80
    ’
    -1.70
    c
    -1.57
    t
    -1.55
     there
    -1.44
     nimmer
    -1.44
     pensent
    -1.42
    <b>
    -1.38
     parlent
    -1.37
    時候
    -1.31
    POSITIVE LOGITS
    蕺
    1.52
    僰
    1.47
     моя
    1.47
    izaci
    1.45
    mathematical
    1.44
    miR
    1.42
    ())
    1.42
    ?></
    1.41
    ாய
    1.38
    ℣
    1.38
    Activations Density 0.023%

    No Known Activations