© 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. 34-GEMMASCOPE-RES-131K
    7. 79296
    Prev
    Next
    INDEX
    Explanations

    code, programming, objects

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

    The neuron fires on indentation whitespace (leading spaces/tabs) 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_34/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
     sorprender
    -0.93
     ücretsiz
    -0.91
     katun
    -0.82
    \"></
    -0.81
     Hivatalos
    -0.81
    Unmarshaller
    -0.81
    himo
    -0.81
    żyw
    -0.80
     každ
    -0.80
     stále
    -0.79
    POSITIVE LOGITS
     this
    1.09
    if
    0.94
     if
    0.86
    ilangkan
    0.85
     surely
    0.85
     esse
    0.84
    abetes
    0.83
     certainly
    0.82
     cav
    0.82
     conspiracy
    0.82
    Activations Density 0.002%

    No Known Activations