© 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. 36706
    Prev
    Next
    INDEX
    Explanations

    dialogue turns

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

    This neuron detects dialogue turn markers and other punctuation (especially colons and closing parentheses) that label speaker lines.

    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
     klubben
    -1.68
     gemeenten
    -1.59
     温泉
    -1.59
     hafta
    -1.51
     Scuola
    -1.47
     hennes
    -1.46
     BIBLIO
    -1.46
     adil
    -1.45
     murah
    -1.45
     sommar
    -1.44
    POSITIVE LOGITS
     It
    1.77
    //!
    1.73
     doesn
    1.72
     they
    1.70
     don
    1.69
    玙
    1.65
    日を
    1.62
     we
    1.59
     isn
    1.59
    いただける
    1.59
    Activations Density 0.007%

    No Known Activations