© 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. 10-GEMMASCOPE-RES-131K
    4. 9131
    Prev
    Next
    INDEX
    Explanations

    describe or list concepts

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

    The neuron fires on the main imperative question words (e.g. “Describe,” “List,” “Map,” etc.) at the start of a question.

    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
     to
    -1.48
    椅子
    -1.39
     témoignage
    -1.34
     into
    -1.32
     same
    -1.25
     mesmos
    -1.20
    UMS
    -1.19
    enschappe
    -1.18
     mêmes
    -1.16
    specified
    -1.16
    POSITIVE LOGITS
     other
    1.45
     that
    1.44
     Papst
    1.40
     買取
    1.34
     Lengkap
    1.31
     how
    1.29
     staff
    1.28
     because
    1.27
     But
    1.24
    者は
    1.23
    Activations Density 0.074%

    No Known Activations