© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 58780
    Prev
    Next
    INDEX
    Explanations

    introducing consequences or descriptions

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

    The neuron fires on causal or transitional signal words—especially adverbs/conjunctions that link or sequence ideas (e.g. “through,” “thereby,” “providing,” “leading,” “even”).

    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
    Three
    -1.48
    While
    -1.38
    Two
    -1.34
    各种
    -1.30
     However
    -1.29
    各类
    -1.27
    谵
    -1.27
     hacerle
    -1.27
    Although
    -1.25
    However
    -1.23
    POSITIVE LOGITS
     then
    1.45
    ?
    1.34
     другую
    1.24
     malheureusement
    1.19
    悻
    1.18
     silla
    1.16
     ก็
    1.13
     dags
    1.12
    ;—
    1.12
     conectado
    1.11
    Activations Density 0.676%

    No Known Activations