© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 73229
    Prev
    Next
    INDEX
    Explanations

    collective entities

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

    The neuron fires strongly on multiword organization, institution, or consortium names (including acronyms) and related proper-noun titles.

    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
     people
    -1.26
     features
    -1.14
     and
    -1.14
     only
    -1.02
     things
    -1.02
     الشرطة
    -0.98
     staff
    -0.98
    Thousands
    -0.98
     time
    -0.98
    .
    -0.97
    POSITIVE LOGITS
     collectively
    1.50
     united
    1.30
     pooled
    1.27
    共同
    1.26
     pooling
    1.25
    colle
    1.23
    共通
    1.16
     unison
    1.16
     объеди
    1.14
    กลุ่ม
    1.11
    Activations Density 0.114%

    No Known Activations