© 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. 66111
    Prev
    Next
    INDEX
    Explanations

    programming language keywords

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

    The neuron fires on tokens used in declaring or defining functions and methods (e.g. “fun,” “override,” return-type keywords, method-signature markers).

    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
     və
    -1.04
     each
    -0.99
    ELTS
    -0.98
    laria
    -0.97
     materiais
    -0.96
     all
    -0.96
    
    -0.95
    bsen
    -0.95
    ContentAsync
    -0.94
     Heiligen
    -0.93
    POSITIVE LOGITS
    首先
    1.11
     Israël
    0.98
    地點
    0.98
    שלום
    0.96
    ないけど
    0.92
    授業
    0.90
    腦
    0.88
     канал
    0.86
    まず
    0.86
     sambal
    0.85
    Activations Density 0.006%

    No Known Activations