© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 9468
    Prev
    Next
    INDEX
    Explanations

    explanation of why

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

    The neuron detects causal and explanatory cue words—tokens like “explains,” “why,” “owing,” “result,” etc., that signal a cause‐and‐effect or explanatory relation.

    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
    DeclareMath
    -1.11
    ляли
    -1.03
     シリーズ
    -1.02
    ρούν
    -1.00
    tly
    -1.00
    !!</
    -0.99
    :{}
    -0.96
    '/>
    -0.95
     vuelva
    -0.94
    ഏ
    -0.94
    POSITIVE LOGITS
     some
    1.17
     so
    1.10
    xem
    1.09
     algunas
    1.04
     سیستم
    0.99
    我在
    0.96
    banyak
    0.96
     overall
    0.94
    tiara
    0.93
     někter
    0.92
    Activations Density 0.034%

    No Known Activations