© 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. 129190
    Prev
    Next
    INDEX
    Explanations

    This neuron activates primarily on programming identifiers—names of variables, functions, constants, or other user‐defined symbols in code.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    rule, hole, or val

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     not
    -1.14
     J
    -1.13
    _
    -1.12
     F
    -1.10
     L
    -1.09
     use
    -1.09
     j
    -1.06
     D
    -1.05
    tif
    -1.03
    '
    -1.02
    POSITIVE LOGITS
    และ
    1.35
     trafik
    1.31
     monstre
    1.30
     persil
    1.30
     murale
    1.28
     rapporter
    1.27
     foton
    1.27
     peluche
    1.25
     maksi
    1.24
     illustre
    1.22
    Activations Density 0.002%

    No Known Activations