© 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. 10-GEMMASCOPE-RES-131K
    7. 128640
    Prev
    Next
    INDEX
    Explanations

    code separators or path components

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

    This neuron fires on technical identifiers and proper‐noun tokens in code or config—e.g. library/class names, uppercase constants, function names, and other code‐style keywords.

    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
    教你
    -1.51
     our
    -1.46
     celana
    -1.45
     nasze
    -1.43
     (!__
    -1.38
    するための
    -1.36
     skolan
    -1.35
     naše
    -1.35
    oncesto
    -1.34
    (!__
    -1.33
    POSITIVE LOGITS
    だが
    1.60
     
    1.45
     Pueden
    1.38
     Encuentra
    1.31
    笑
    1.30
            
    1.28
     Finalmente
    1.27
     Cualquier
    1.27
        
    1.25
     Récup
    1.23
    Activations Density 0.059%

    No Known Activations