© 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. 118879
    Prev
    Next
    INDEX
    Explanations

    Davies, Davison, Dennison

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

    The neuron strongly activates on occurrences of the substring "Dav"—in particular proper names like Davies, Davidson, Dennison, etc.

    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
     или
    -2.97
     eerie
    -2.56
     второй
    -2.47
     ache
    -2.45
     glaubt
    -2.44
    </b>
    -2.39
     nuova
    -2.38
    ৢ
    -2.34
    麁
    -2.33
     oder
    -2.31
    POSITIVE LOGITS
    鹮
    3.11
    螄
    2.52
     放題
    2.48
    朥
    2.48
     russ
    2.39
     花纹
    2.38
    жуть
    2.33
    !';
    2.28
     cristo
    2.27
    s
    2.27
    Activations Density 0.005%

    No Known Activations