© 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. 40337
    Prev
    Next
    INDEX
    Explanations

    programming constructs and notation

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

    The neuron activates on long runs of leading whitespace (indentation) tokens.

    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
     for
    -2.72
     bolster
    -2.17
     with
    -2.16
     seven
    -2.11
     phenomenal
    -2.09
     burgeoning
    -2.09
     Providing
    -2.08
     if
    -2.06
     by
    -2.05
     at
    -2.02
    POSITIVE LOGITS
     isolamento
    2.38
     continually
    2.36
     molte
    2.34
     semelh
    2.33
     amazingly
    2.28
    蚡
    2.28
     molti
    2.23
     remarkably
    2.16
     insanely
    2.16
     berço
    2.11
    Activations Density 0.005%

    No Known Activations