© 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. 19609
    Prev
    Next
    INDEX
    Explanations

    programming language data types

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

    The neuron activates on occurrences of “type” (in any casing or context)—for example in TYPE macros, the typeof keyword, typedefs, and other tokens containing “type.”

    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
     into
    -1.17
    鋏
    -1.10
     venture
    -1.10
    したのは
    -1.09
    儔
    -1.08
     was
    -1.07
    introdu
    -1.05
    ⽉
    -1.05
     crianças
    -1.05
     between
    -1.05
    POSITIVE LOGITS
     sagde
    1.41
     vincit
    1.35
     suele
    1.31
     also
    1.31
    when
    1.31
    Usually
    1.29
     Usually
    1.27
    畀
    1.26
    Because
    1.24
     usually
    1.23
    Activations Density 0.026%

    No Known Activations