© 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. 22-GEMMASCOPE-RES-131K
    7. 39928
    Prev
    Next
    INDEX
    Explanations

    introducing additional information

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

    This neuron detects discourse interjections or afterthought markers (e.g. “Oh,” “and,” “also”) that introduce side comments or addenda.

    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
    我們
    -1.12
     vorbereitet
    -1.06
    さあ
    -1.05
    UREAU
    -1.04
    Honestly
    -1.03
    色々
    -1.02
     absorción
    -1.02
    unità
    -1.02
     różnych
    -1.02
    Taking
    -1.02
    POSITIVE LOGITS
     also
    2.41
     también
    1.52
     myös
    1.52
     mention
    1.45
     também
    1.44
    also
    1.35
     également
    1.27
     भी
    1.25
    También
    1.23
    ALSO
    1.20
    Activations Density 0.007%

    No Known Activations