© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 12213
    Prev
    Next
    INDEX
    Explanations

    buchbinder, buchholz, buchannon, bryant

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

    Explanation of neuron 4 behavior: the main thing this neuron does is find tokens containing the substring "Buch" (e.g. Buchbinder, Buchowetzki, Buch).

    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
    ↵
    -2.50
    骜
    -2.44
    的说道
    -2.33
    がこの
    -2.25
    ’
    -2.23
    
    -2.22
    न
    -2.19
    Ꮠ
    -2.11
    并未
    -2.09
     ''
    -2.02
    POSITIVE LOGITS
    to
    2.89
    and
    2.55
     値下げ
    2.31
    }$
    2.27
    颳
    2.25
    鸰
    2.25
    々は
    2.22
    𓄹
    2.17
    ()):
    2.16
    韞
    2.14
    Activations Density 0.005%

    No Known Activations