© 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. 22-GEMMASCOPE-RES-131K
    7. 27413
    Prev
    Next
    INDEX
    Explanations

    interjections and exclamations

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

    The neuron detects sentence-initial discourse markers and interjections (e.g. “Bravo,” “So,” “Hmmm,” “That is…”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     lju
    -0.92
    ⡉
    -0.92
     =[
    -0.90
    rière
    -0.89
    wyt
    -0.88
    리를
    -0.88
    бычно
    -0.86
     moreover
    -0.86
     anche
    -0.86
    MVH
    -0.85
    POSITIVE LOGITS
    Sounds
    1.33
     Sounds
    1.27
     sounds
    1.25
    !
    1.13
     precies
    1.09
     }{@
    0.99
     suena
    0.98
     so
    0.98
     Physik
    0.98
     incluye
    0.97
    Activations Density 0.040%

    No Known Activations