© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 19297
    Prev
    Next
    INDEX
    Explanations

    oh! exclamation

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

    The neuron detects emotionally charged interjections or reactive expressions (e.g. “Oh,” “I’m sorry,” “Thank God,” “Oh, come on,” 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
    m
    -1.82
    )
    -1.71
    A
    -1.47
     predominantly
    -1.45
     however
    -1.41
    N
    -1.40
     toutefois
    -1.38
    -
    -1.38
    C
    -1.38
     Kühlschrank
    -1.35
    POSITIVE LOGITS
     if
    1.56
     kesi
    1.55
     when
    1.48
    Still
    1.41
    ные
    1.38
    ndan
    1.37
    两位
    1.35
     ещё
    1.34
     because
    1.34
    Legende
    1.33
    Activations Density 0.011%

    No Known Activations