© 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. 25954
    Prev
    Next
    INDEX
    Explanations

    programming code variables

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

    The neuron fires on programming identifiers (especially variable names and parameters) 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
     what
    -0.96
    Și
    -0.91
     fecha
    -0.89
    orithm
    -0.87
     flechas
    -0.85
     and
    -0.84
     Су
    -0.84
    ఊ
    -0.84
    :),
    -0.83
     andata
    -0.83
    POSITIVE LOGITS
    mmung
    0.98
    部分を
    0.88
     canto
    0.87
     therein
    0.86
     efec
    0.85
     Huffington
    0.85
     controver
    0.84
     contempl
    0.83
     reple
    0.82
     Deed
    0.81
    Activations Density 0.006%

    No Known Activations