© 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. 72425
    Prev
    Next
    INDEX
    Explanations

    documentation and code comments

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

    The neuron fires most strongly on numeric tokens (e.g. version numbers, years, digit‐only literals) in source or documentation.

    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
     Krzysz
    -1.05
    ENAS
    -0.90
    četně
    -0.89
     Pièces
    -0.89
     hauts
    -0.88
    ENEF
    -0.88
    ॲ
    -0.88
     his
    -0.87
    termine
    -0.86
    ਕ
    -0.86
    POSITIVE LOGITS
     to
    1.02
     from
    1.02
    Aligned
    0.98
     with
    0.96
    deprecated
    0.94
    過ぎて
    0.89
     also
    0.88
    sian
    0.87
    llary
    0.82
     stesse
    0.82
    Activations Density 0.038%

    No Known Activations