© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 16485
    Prev
    Next
    INDEX
    Explanations

    communication and inquiry

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

    The neuron fires on verbs or noun‐forms that denote communication acts—e.g. “called,” “talked,” “asked,” “discussion,” “argued,” “searched.”

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    資料來源
    -1.09
     doméstica
    -1.05
     ४
    -1.05
    zá
    -1.03
    헤
    -1.01
    varet
    -1.00
    事態
    -1.00
    それも
    -0.96
    дной
    -0.96
     прине
    -0.95
    POSITIVE LOGITS
    ždy
    1.26
    turnstile
    1.23
     OGSÅ
    1.20
    apnews
    1.18
     vždy
    1.13
    の方に
    1.12
     about
    1.10
    ほら
    1.09
    為に
    1.09
    subaru
    1.09
    Activations Density 0.047%

    No Known Activations