© 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. 22-GEMMASCOPE-RES-131K
    4. 66158
    Prev
    Next
    INDEX
    Explanations

    conditional programming structures

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

    The neuron fires on uppercase or punctuation‐delimited reserved keywords and control operators (e.g. IF, OR, Condition=, the “#”/“?” in templates, etc.) across various markup and code snippets.

    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
     harap
    -1.30
    Karakter
    -1.19
    Waterproof
    -1.15
    junto
    -1.13
     آزادی
    -1.08
    stainless
    -1.07
    hasta
    -1.05
    衚
    -1.04
    iramente
    -1.04
    Cree
    -1.03
    POSITIVE LOGITS
     there
    1.23
     not
    1.13
    &+
    1.04
     />';
    1.02
    题
    1.02
    情况
    1.00
     (
    1.00
    meida
    0.98
    (!(
    0.97
    X
    0.97
    Activations Density 0.037%

    No Known Activations