© 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. 436
    Prev
    Next
    INDEX
    Explanations

    entertainment companies and management

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

    The neuron fires on words used in those “X is a …” biography/introduction lines—tokens that mark an entity’s role, affiliation, location, or status (e.g. “is,” “under,” “based,” “from,” “former,” 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_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
    сіб
    -0.74
    omination
    -0.71
     när
    -0.71
    ‸
    -0.70
     any
    -0.69
    apatkan
    -0.69
    Bộ
    -0.68
     when
    -0.68
     alc
    -0.67
     porcel
    -0.67
    POSITIVE LOGITS
     agency
    1.69
     agencies
    1.34
     company
    1.34
    公司
    1.23
     entertainment
    1.21
     brokerage
    1.16
    company
    1.10
     агент
    1.09
     management
    1.07
    entertainment
    1.06
    Activations Density 0.009%

    No Known Activations