© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 4634
    Prev
    Next
    INDEX
    Explanations

    variable names and code syntax

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

    The neuron strongly activates on code‐style identifiers and numeric literals (e.g. variable names like “fields” and numbers such as “1”) in programming snippets.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    	
    1.51
     =
    1.45
    ↵
    1.40
     /*
    1.38
     //
    1.36
     ==
    1.30
    		
    1.27
     json
    1.27
     \
    1.17
    //
    1.16
    POSITIVE LOGITS
     अल्को
    1.35
    industrie
    1.22
    ेंद्रीय
    1.20
     निजामाच्या
    1.20
    आधी
    1.18
     सूत्रकृ
    1.17
     रक्षाबंधनाच्या
    1.17
    विषयी
    1.16
    श्मीर
    1.13
    enhancing
    1.12
    Activations Density 0.070%

    No Known Activations