© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 86121
    Prev
    Next
    INDEX
    Explanations

    knowing what

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

    The neuron is keyed to occurrences of “what” (especially in “know what …” constructions) that introduce clauses about understanding or explaining something.

    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
    1
    -2.02
    5
    -1.87
    6
    -1.82
    The
    -1.80
    7
    -1.78
     have
    -1.75
    焄
    -1.70
    when
    -1.67
     What
    -1.66
    3
    -1.66
    POSITIVE LOGITS
     to
    2.08
    asiswa
    1.75
    🤡
    1.63
    PreferredItem
    1.62
    และ
    1.56
     edifici
    1.55
     normaal
    1.48
    ,’”
    1.48
    🔥🔥🔥
    1.48
    
    1.47
    Activations Density 0.025%

    No Known Activations