© 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. 111753
    Prev
    Next
    INDEX
    Explanations

    locations, dates, institutions

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

    The neuron fires on the parenthetical birthplace and birth/death date blocks (cities, years, and their commas) that follow a person’s name in biographical intros.

    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
    ۹
    -1.34
     gore
    -1.31
    чну
    -1.23
    illères
    -1.23
     jones
    -1.20
     Js
    -1.17
     dino
    -1.16
     apparaissent
    -1.15
     viktig
    -1.12
     Krzysztof
    -1.12
    POSITIVE LOGITS
     out
    1.46
     option
    1.40
     different
    1.36
     far
    1.33
     all
    1.33
     tramo
    1.30
     off
    1.27
     options
    1.27
    prising
    1.26
     four
    1.23
    Activations Density 0.050%

    No Known Activations