© 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-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 11084
    Prev
    Next
    INDEX
    Explanations

    var declarations

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

    The neuron fires strongly on Go declaration keywords—especially “var” (and to a lesser extent “import”) at the top level.

    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
    蟊
    -1.51
    úsica
    -1.47
    mainly
    -1.43
    
    -1.37
     และ
    -1.37
    เพื่อ
    -1.36
    睃
    -1.32
    ఽ
    -1.31
     helicóptero
    -1.31
     عناصر
    -1.31
    POSITIVE LOGITS
    </h1>
    1.55
     very
    1.48
    a
    1.42
     (
    1.41
     appropriate
    1.39
     truly
    1.27
     profoundly
    1.27
    1.24
     или
    1.24
     two
    1.24
    Activations Density 0.007%

    No Known Activations