© 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. 34-GEMMASCOPE-RES-131K
    7. 33547
    Prev
    Next
    INDEX
    Explanations

    speaking or tongue-related words

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

    The neuron detects parenthetical or narrative asides that describe a speaker’s tone or manner (e.g. “said with absolute sarcasm”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     ファイ
    -0.79
    ौर
    -0.78
     ALONE
    -0.76
    confused
    -0.75
    woven
    -0.73
    dees
    -0.73
     FRANCISCO
    -0.73
    संद
    -0.72
    copter
    -0.72
     чу
    -0.72
    POSITIVE LOGITS
     tongue
    4.59
    tongue
    3.64
     Tongue
    3.36
    Tongue
    3.22
     tongues
    2.73
    舌
    1.77
     tong
    1.51
     jest
    1.32
    tong
    1.30
     tou
    1.29
    Activations Density 0.037%

    No Known Activations