© 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. 10-GEMMASCOPE-RES-131K
    7. 86882
    Prev
    Next
    INDEX
    Explanations

    scanning electron

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

    The neuron fires on tokens that look like technical identifiers—uppercase acronyms, class or variable names, version numbers, and other alphanumeric “code-style” or metadata 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_10/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
     to
    -2.16
     some
    -1.27
     those
    -1.20
     次元
    -1.15
     gesagt
    -1.11
     vermutlich
    -1.10
    لقاء
    -1.08
     particularly
    -1.07
     sufficient
    -1.05
     discrepan
    -1.05
    POSITIVE LOGITS
    plotlib
    1.20
    なのです
    1.10
     igény
    1.08
     kysy
    1.08
     Euró
    1.06
     quando
    1.05
    
    1.05
    URG
    1.05
     dù
    1.05
     exemp
    1.04
    Activations Density 0.034%

    No Known Activations