© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 73795
    Prev
    Next
    INDEX
    Explanations

    mathematical and programming constructs

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

    The neuron strongly activates on code‐specific identifiers and literals (variable names, file names, numeric constants, etc.) rather than ordinary prose words.

    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
     on
    -0.97
     from
    -0.94
    tification
    -0.91
     by
    -0.91
    に行きました
    -0.81
    newborn
    -0.79
     Grüßen
    -0.79
    vampire
    -0.79
    wiad
    -0.78
    asyarakat
    -0.78
    POSITIVE LOGITS
    Pour
    0.94
    raged
    0.91
    voyez
    0.88
     טוב
    0.88
    block
    0.88
    arien
    0.87
    Zgod
    0.83
    ritation
    0.83
     Appetit
    0.83
    Block
    0.82
    Activations Density 0.002%

    No Known Activations