© 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. 10-GEMMASCOPE-RES-131K
    7. 75632
    Prev
    Next
    INDEX
    Explanations

    common verbs and conjunctions

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

    The neuron is triggered by direct‐address phrases in dialogue—especially second-person pronouns and related request/question words (e.g. “you,” “me,” “ask,” “getting,” “place”).

    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
    rás
    -1.90
     armazenamento
    -1.88
     skvě
    -1.77
     rác
    -1.77
    atique
    -1.75
     atualizar
    -1.75
    𐀀
    -1.75
     hà
    -1.73
     créer
    -1.73
     estadía
    -1.72
    POSITIVE LOGITS
     for
    1.94
     eight
    1.87
     nine
    1.69
     five
    1.65
     ingrained
    1.63
     with
    1.62
     high
    1.55
     three
    1.55
     ostensibly
    1.55
     as
    1.51
    Activations Density 0.054%

    No Known Activations