© 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. 22-GEMMASCOPE-RES-131K
    7. 16086
    Prev
    Next
    INDEX
    Explanations

    true else false

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

    The neuron fires on boolean-related code tokens—particularly keywords like “true”, “false”, and branching keywords such as “else” in programming code.

    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
     что
    -0.95
    っくり
    -0.92
    ведения
    -0.90
     selepas
    -0.85
     which
    -0.84
    FetchType
    -0.84
     "'";
    -0.84
     didik
    -0.83
    asilkan
    -0.81
     urs
    -0.80
    POSITIVE LOGITS
     otherwise
    1.13
    osuke
    1.00
    otherwise
    0.95
     false
    0.92
    else
    0.88
     Nope
    0.85
    FALSE
    0.85
     jinak
    0.84
    YesNo
    0.84
     else
    0.83
    Activations Density 0.003%

    No Known Activations