© 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. 34197
    Prev
    Next
    INDEX
    Explanations

    religious and technical terms

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

    This neuron activates on technical or mathematical terminology—especially programming/library names, code‐related words, and algorithm/math symbols.

    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
     сейчас
    -1.33
    Catatan
    -1.30
     that
    -1.25
     when
    -1.21
     vaš
    -1.19
    uniqlo
    -1.19
     incorporating
    -1.17
    Fitur
    -1.16
     girar
    -1.16
     если
    -1.14
    POSITIVE LOGITS
    의
    1.27
    _
    1.27
     کے
    1.15
    presumably
    1.14
     into
    1.11
     REALLY
    1.09
     of
    1.09
    揍
    1.08
    H
    1.08
    \_
    1.05
    Activations Density 0.007%

    No Known Activations