© 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. 110573
    Prev
    Next
    INDEX
    Explanations

    methods, functions, or variables

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

    the closing angle bracket in generic type declarations (like `List<T>`).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron activates on the angle‐bracket tokens (“<” and “>”) that mark generic type parameters in code.

    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
    Toute
    -1.81
    Elő
    -1.80
    teilung
    -1.67
    Quoi
    -1.67
     meuble
    -1.65
    Selon
    -1.64
     plonge
    -1.63
    久しぶりの
    -1.63
    Pratique
    -1.62
     vendeur
    -1.60
    POSITIVE LOGITS
     of
    2.55
     to
    2.44
     it
    2.23
     this
    2.17
     You
    2.16
     from
    2.13
     out
    2.00
     as
    1.96
     I
    1.91
     using
    1.89
    Activations Density 0.005%

    No Known Activations