© 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. 34-GEMMASCOPE-RES-131K
    7. 12003
    Prev
    Next
    INDEX
    Explanations

    people counts

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

    The neuron activates on numeric tokens representing counts or measurements—especially population sizes, seating capacities, and other large numerical figures.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     crews
    -1.20
     groups
    -1.17
     teams
    -1.13
     and
    -1.02
    uttore
    -0.96
     group
    -0.96
    -)
    -0.94
    object
    -0.93
    group
    -0.91
    tett
    -0.91
    POSITIVE LOGITS
     people
    2.20
     persons
    1.84
     souls
    1.48
     pessoas
    1.45
     person
    1.38
    people
    1.35
     человек
    1.32
     individuals
    1.30
     personnes
    1.30
     personen
    1.30
    Activations Density 0.089%

    No Known Activations