© 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. 12948
    Prev
    Next
    INDEX
    Explanations

    Darren and sleeping

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

    This neuron is sensitive to tokens that are capitalized (proper names or sentence-initial words).

    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
    糴
    -2.69
    蓠
    -2.64
    人
    -2.45
     berdua
    -2.25
     سيارة
    -2.22
     coûts
    -2.22
     моей
    -2.17
    鵙
    -2.16
    
    -2.16
    c
    -2.16
    POSITIVE LOGITS
    ↵↵
    3.25
     For
    3.25
    "
    2.94
     Eigentü
    2.69
     Also
    2.36
     encompass
    2.31
     It
    2.30
     möge
    2.28
     gewähr
    2.27
     Präsidenten
    2.27
    Activations Density 0.003%

    No Known Activations