© 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. 26489
    Prev
    Next
    INDEX
    Explanations

    organization or group affiliation

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

    The neuron fires on prominent content nouns—especially institution-, group-, or country-related nouns (e.g. “department,” “paper,” “country,” “team”)—often marking key entities in the 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_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
     its
    -1.34
     who
    -1.20
     their
    -1.20
     it
    -1.09
     his
    -1.07
     EACH
    -1.04
    を読む
    -1.03
    ようになります
    -1.03
    っていました
    -1.00
     obstáculos
    -1.00
    POSITIVE LOGITS
    「
    1.06
    著者
    1.05
     colleagues
    1.05
    几个人
    1.05
    wide
    1.02
     teammate
    1.02
    mate
    1.01
    先日
    0.99
     peers
    0.97
    MATE
    0.97
    Activations Density 0.032%

    No Known Activations