© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 65602
    Prev
    Next
    INDEX
    Explanations

    ages of people

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

    The neuron fires on numeric expressions—particularly age and quantity mentions (e.g. “two,” “17-year-old,” “18”).

    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
    篆
    -0.91
     that
    -0.88
     like
    -0.87
     in
    -0.84
     grigio
    -0.82
     +
    -0.82
    setAdapter
    -0.82
    texpr
    -0.80
    おいしい
    -0.80
    而在
    -0.79
    POSITIVE LOGITS
     address
    1.41
    අ
    1.27
     addresses
    1.25
     resident
    1.25
     alleged
    1.22
     unemployed
    1.22
    mployed
    1.20
     defendant
    1.16
     residents
    1.16
     tatuajes
    1.15
    Activations Density 0.067%

    No Known Activations