© 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. 34-GEMMASCOPE-RES-131K
    7. 33441
    Prev
    Next
    INDEX
    Explanations

    resume talks

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

    The neuron strongly activates on words in title‐ or headline‐style capitalization—particularly proper names, titles, and other initial‐capitalized tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     Seel
    -0.76
    Chairman
    -0.76
    EXPR
    -0.76
     Downloader
    -0.75
     организация
    -0.70
    四川
    -0.70
    richts
    -0.69
    Enders
    -0.69
    Orange
    -0.69
     Chairman
    -0.69
    POSITIVE LOGITS
     Servant
    0.87
     Zelensky
    0.85
    lrrrr
    0.85
    Mustafa
    0.84
     nějak
    0.84
    ému
    0.84
     PURE
    0.82
     Normandy
    0.82
    produces
    0.81
    volumes
    0.81
    Activations Density 0.007%

    No Known Activations