© 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. 22466
    Prev
    Next
    INDEX
    Explanations

    cutting or function calls

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

    The neuron detects occurrences of the substring “cut” (in any capitalization or composition).

    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
     of
    -2.34
    o
    -2.17
    ).
    -2.09
    倣
    -2.09
    胧
    -2.09
    なんだ
    -2.06
     podstawie
    -2.00
     ilha
    -1.98
    の色
    -1.96
     have
    -1.88
    POSITIVE LOGITS
    two
    2.30
    How
    2.23
    蹋
    2.11
     towering
    2.09
     problemet
    2.09
     اسلام
    2.08
    Now
    2.03
     ingrained
    2.02
    を行い
    2.02
    Having
    1.99
    Activations Density 0.004%

    No Known Activations