© 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. 21314
    Prev
    Next
    INDEX
    Explanations

    asking or answering questions

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

    The neuron spots occurrences of “answer,” “answering,” and especially “questions,” i.e. segments related to asking or answering questions.

    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
    -2.11
    ’
    -1.62
     be
    -1.38
    '
    -1.37
     vielleicht
    -1.37
     rampant
    -1.37
     do
    -1.34
     tutaj
    -1.31
    astră
    -1.31
     нескольких
    -1.30
    POSITIVE LOGITS
     about
    1.80
    翾
    1.69
    代わりに
    1.67
    ⦊
    1.52
    这是什么
    1.48
     aceea
    1.45
    vají
    1.42
    鏵
    1.42
     liée
    1.41
    佧
    1.41
    Activations Density 0.018%

    No Known Activations