© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 49594
    Prev
    Next
    INDEX
    Explanations

    employee and employer related matters

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

    The neuron strongly activates on sequences of numeric or alphanumeric codes laid out like table/data fields (e.g. multi‐digit numbers, dates, government fare/eligibility codes). It is essentially detecting structured numeric data entries rather than ordinary text.

    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
    נד
    -0.82
    yntax
    -0.78
     作業
    -0.76
     disassembly
    -0.74
    접
    -0.73
    äger
    -0.73
    Emir
    -0.73
    תו
    -0.73
    тивно
    -0.72
    bria
    -0.71
    POSITIVE LOGITS
     pension
    0.92
    Retirement
    0.84
     promotion
    0.82
    Promotion
    0.82
     Pension
    0.82
     retirement
    0.82
    pension
    0.81
    retirement
    0.76
     employee
    0.75
     retire
    0.75
    Activations Density 0.015%

    No Known Activations