© 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. 22-GEMMASCOPE-RES-131K
    4. 93256
    Prev
    Next
    INDEX
    Explanations

    people in professional roles

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

    The neuron fires on capitalized occupational titles or role names (e.g. “Producer,” “Executive producer,” “Line producer,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     itself
    -1.23
     its
    -1.20
    Istorija
    -1.16
    έ
    -1.13
     it
    -1.13
    licante
    -1.12
    लिखित
    -1.10
     препо
    -1.09
     which
    -1.07
     contains
    -1.06
    POSITIVE LOGITS
     who
    3.16
     role
    1.38
     whom
    1.34
    是一位
    1.26
     someone
    1.23
     hän
    1.18
    에게
    1.16
     himself
    1.15
    Who
    1.15
    们
    1.14
    Activations Density 0.079%

    No Known Activations