© 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. 7806
    Prev
    Next
    INDEX
    Explanations

    how to

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

    The neuron detects interrogative question words (e.g. “how,” “what”) marking that a sentence is asking 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
    jenigen
    -1.66
     compleanno
    -1.64
     ederim
    -1.53
     மற்ற
    -1.52
     wünsche
    -1.48
     miljö
    -1.46
     innych
    -1.44
     veramente
    -1.42
     cotone
    -1.41
     anderen
    -1.40
    POSITIVE LOGITS
    .
    1.49
    先是
    1.39
     what
    1.38
    راحل
    1.33
    子を
    1.31
    1.31
    1.31
     how
    1.30
    縞
    1.27
    一群
    1.27
    Activations Density 0.050%

    No Known Activations