© 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. 65035
    Prev
    Next
    INDEX
    Explanations

    equality and comparison operations

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

    The neuron fires on tokens that are names of operations or methods/functions (e.g. subtract, diff, convertTo, CompareTo, Equals, Union, GT) in 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
     nimmer
    -1.05
     they
    -1.04
    THEY
    -0.99
    依靠
    -0.96
    鼍
    -0.95
     ADDITIONAL
    -0.95
    が
    -0.94
    лям
    -0.93
     Minggu
    -0.93
     débris
    -0.93
    POSITIVE LOGITS
     other
    1.49
     what
    1.29
     different
    1.25
     Übernach
    1.20
     or
    1.16
     from
    1.12
     olvidado
    1.10
     more
    1.10
     steadfast
    1.08
     contemporáneo
    1.06
    Activations Density 0.002%

    No Known Activations