© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 84731
    Prev
    Next
    INDEX
    Explanations

    discussing future or past aspects

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

    This neuron detects discourse‐structuring cue words that introduce or frame an explanation or topic (e.g. “reasons,” “how,” “behind,” “your particular … needs,” “what kind of,” “future of,” “recent controversies regarding,” etc.).

    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
    :
    -1.48
     даже
    -1.41
    ;
    -1.40
     sogar
    -1.38
    a
    -1.35
    il
    -1.32
    ;
    -1.29
    ri
    -1.29
    w
    -1.29
    ↵
    -1.28
    POSITIVE LOGITS
     található
    1.33
    Elekt
    1.31
     pohod
    1.28
    Николай
    1.25
     how
    1.22
    droj
    1.20
     ekip
    1.20
    Encontra
    1.20
    Gön
    1.20
     bezpeč
    1.19
    Activations Density 0.134%

    No Known Activations