© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 127668
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on loop‐iteration constructs (e.g. “for,” “foreach,” “each” loops).

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    news headlines and scientific titles

    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
    Negative Logits
    sont
    -1.03
     pantal
    -1.02
     konfig
    -1.01
    𓆏
    -0.97
    毎回
    -0.94
     ytterligare
    -0.94
     tunik
    -0.94
     kokos
    -0.93
     rapporter
    -0.93
     impairments
    -0.93
    POSITIVE LOGITS
     strengthen
    1.19
     changes
    1.05
     unless
    1.04
     new
    1.02
     If
    1.00
    ’
    0.97
     Unless
    0.95
     strengthening
    0.95
     directly
    0.94
    Literatura
    0.93
    Activations Density 0.005%

    No Known Activations