© 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. 26211
    Prev
    Next
    INDEX
    Explanations

    names and identifiers

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

    The neuron specifically fires on the token “name” when it appears as an XML (or Android manifest) attribute name.

    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
     create
    -1.00
     get
    -0.99
     offset
    -0.95
     where
    -0.92
    了的
    -0.90
     electronically
    -0.87
    形の
    -0.86
     ambiguous
    -0.85
     install
    -0.83
     help
    -0.81
    POSITIVE LOGITS
     figliu
    0.98
     gepub
    0.98
     امریک
    0.97
     morire
    0.95
     azzurro
    0.95
     cucchiai
    0.94
     язы
    0.94
    acabana
    0.94
    ⅽ
    0.93
    的一切
    0.93
    Activations Density 0.000%

    No Known Activations