© 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. 84861
    Prev
    Next
    INDEX
    Explanations

    . followed by identifiers

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

    The neuron fires on tokens that look like code identifiers used in method or function calls (e.g. object.property or functionName()).

    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
    狴
    -1.58
    ༷
    -1.45
    ︲
    -1.44
    ſſi
    -1.38
     COMES
    -1.35
    aneity
    -1.34
    gives
    -1.30
    hich
    -1.30
    粜
    -1.27
    ㎖
    -1.26
    POSITIVE LOGITS
     and
    2.25
     is
    1.64
     may
    1.43
     was
    1.38
     are
    1.38
    ία
    1.34
    rittura
    1.30
    年は
    1.29
     棟
    1.26
     тоже
    1.25
    Activations Density 0.015%

    No Known Activations