© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 68195
    Prev
    Next
    INDEX
    Explanations

    how/what questions about attributes

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

    This neuron activates on interrogative question stems—especially “How” and “What” phrases (e.g. “How long,” “How much,” “What type,” “What else”).

    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
     menyen
    -2.30
     laiss
    -2.28
     berge
    -2.16
     regionale
    -2.08
     mêmes
    -2.05
    ครื่อง
    -2.03
     montrer
    -2.00
     flesta
    -1.98
     aient
    -1.96
     花纹
    -1.94
    POSITIVE LOGITS
     would
    2.64
     did
    2.22
    What
    2.20
     could
    2.19
     You
    2.17
     do
    2.14
     then
    2.11
    When
    2.08
     find
    2.05
     are
    2.03
    Activations Density 0.040%

    No Known Activations