© 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. 8406
    Prev
    Next
    INDEX
    Explanations

    Proper nouns and associated words

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

    The neuron fires on capitalized tokens—especially names of organizations, people, or other proper nouns.

    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
    They
    -1.50
    There
    -1.49
    Many
    -1.38
    Very
    -1.37
    Sometimes
    -1.34
    With
    -1.32
     like
    -1.31
    Numerous
    -1.24
     catalana
    -1.23
    Other
    -1.23
    POSITIVE LOGITS
     this
    1.76
    ifikationer
    1.42
    tahui
    1.34
    側から
    1.31
    junctive
    1.31
    ghalaya
    1.27
     necesar
    1.26
     why
    1.25
     phun
    1.25
     appropri
    1.24
    Activations Density 0.236%

    No Known Activations