© 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. 22-GEMMASCOPE-RES-131K
    7. 68687
    Prev
    Next
    INDEX
    Explanations

    do stuff

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

    This neuron fires on informal action-oriented phrases in code comments (e.g. “do what you want with …”).

    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
     máme
    -1.26
    Mirage
    -1.14
     颜色
    -1.13
     konfer
    -1.12
    ようで
    -1.11
     köy
    -1.11
    tapete
    -1.10
    مقاله
    -1.09
    Lij
    -1.09
     ľ
    -1.08
    POSITIVE LOGITS
     here
    1.45
     would
    1.33
     now
    1.30
     other
    1.27
     some
    1.26
     whatever
    1.20
     additional
    1.19
    D
    1.17
     or
    1.14
     still
    1.14
    Activations Density 0.045%

    No Known Activations