© 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. 128023
    Prev
    Next
    INDEX
    Explanations

    people's names

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

    The neuron fires on capitalized named entities—proper nouns such as people’s names, show titles, or character names.

    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
     этого
    -1.28
    tieron
    -1.22
     accompagné
    -1.20
     betont
    -1.18
     that
    -1.17
    tiéndose
    -1.14
     этой
    -1.13
     sababu
    -1.13
     pubblici
    -1.12
     этом
    -1.11
    POSITIVE LOGITS
     to
    1.43
     følgende
    1.38
     according
    1.34
     besø
    1.26
     ferdig
    1.24
     lecteur
    1.16
     løpet
    1.15
     valgt
    1.12
     mengunjungi
    1.12
     forbindelse
    1.12
    Activations Density 0.006%

    No Known Activations