© 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. 127291
    Prev
    Next
    INDEX
    Explanations

    hesitant interjections

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

    The neuron fires on brief spoken‐style interjections or hesitations (e.g. “uh,” “um,” “hmm,” “ah,” “oh,” 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
     of
    -3.22
     for
    -2.00
     to
    -1.98
     by
    -1.79
     pequeno
    -1.75
    klju
    -1.74
     hitherto
    -1.74
     consentimento
    -1.70
    蒾
    -1.66
     from
    -1.64
    POSITIVE LOGITS
    ksiyon
    1.67
     verhind
    1.54
     évé
    1.54
     victoires
    1.52
     為
    1.51
     приходится
    1.51
     rembour
    1.48
    ꜆
    1.48
    ycznie
    1.48
     sark
    1.47
    Activations Density 0.016%

    No Known Activations