© 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. 67405
    Prev
    Next
    INDEX
    Explanations

    code keys and properties

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

    The neuron fires on programming “key” identifiers (attribute or parameter names) such as id, name, url, type, get, ForKey, etc.

    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
    któ
    -1.12
     kontinu
    -1.06
    Roboto
    -0.99
     porsi
    -0.97
     this
    -0.97
    Whatever
    -0.96
     Ně
    -0.95
    🔃
    -0.94
     Če
    -0.94
     in
    -0.94
    POSITIVE LOGITS
    などが
    1.30
     estabelec
    1.18
     '
    1.18
     focuses
    1.14
    ということです
    1.13
     mentre
    1.11
    などを
    1.09
    などは
    1.09
     REALLY
    1.09
    特有
    1.09
    Activations Density 0.006%

    No Known Activations