© 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. 16725
    Prev
    Next
    INDEX
    Explanations

    comparison of values

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

    The neuron strongly activates on uppercase identifiers, constants, and macro‐style tokens in code (e.g. preprocessor symbols or named constants).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     katk
    -1.10
     dáms
    -1.09
    cepts
    -1.05
     naka
    -1.02
     notamment
    -1.02
     tecnologias
    -1.00
     konfigura
    -1.00
    springen
    -1.00
    Feels
    -0.99
    deta
    -0.98
    POSITIVE LOGITS
     problems
    1.23
     any
    1.14
     this
    1.13
     all
    1.10
     still
    1.05
     supremely
    1.03
     in
    1.02
     Regardless
    1.02
     more
    1.01
     Often
    1.01
    Activations Density 0.001%

    No Known Activations