© 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. 10-GEMMASCOPE-RES-131K
    4. 41270
    Prev
    Next
    INDEX
    Explanations

    file operations and specifications

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

    This neuron detects technical action terms—words naming specific operations or functions (e.g. “draw,” “words,” “sub-ranges,” “file,” “data,” “permissions,” “plaintiffs’ in … cases”) that typically label discrete steps or APIs in code or legal procedure contexts.

    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
     เน
    -1.30
    [-\
    -1.25
    afficheront
    -1.23
     BOTH
    -1.23
     vulnerables
    -1.20
    OLS
    -1.20
    =").
    -1.19
     červ
    -1.19
     ДЕ
    -1.16
     arvo
    -1.16
    POSITIVE LOGITS
     is
    1.40
    pressor
    1.25
     very
    1.14
     borsa
    1.10
     question
    1.06
     on
    1.02
    iecie
    1.00
     year
    0.99
     was
    0.98
     بغ
    0.96
    Activations Density 0.288%

    No Known Activations