© 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. 9067
    Prev
    Next
    INDEX
    Explanations

    programming constructs

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

    The neuron fires on code identifier tokens (variable and function names) rather than on normal language words.

    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
     why
    -1.44
    nologue
    -1.27
     explicado
    -1.22
    OREA
    -1.20
    льності
    -1.19
     to
    -1.18
     divertida
    -1.15
    
    -1.14
     década
    -1.13
    ینک
    -1.13
    POSITIVE LOGITS
    :
    1.34
    それは
    1.26
    épar
    1.23
    最後に
    1.20
     sebelumnya
    1.19
    ”
    1.19
    初めて
    1.17
     was
    1.14
     vollen
    1.13
    <
    1.13
    Activations Density 0.031%

    No Known Activations