© 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. 129314
    Prev
    Next
    INDEX
    Explanations

    The neuron flags high-information, content‐bearing words (proper nouns, key nouns, and other salient terms) rather than common function words.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    role and involvement

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    Médaille
    -0.90
    莅
    -0.90
    Какой
    -0.87
    FAKE
    -0.86
     шесть
    -0.85
     convertTo
    -0.85
    ︙
    -0.84
     assail
    -0.84
    gemaakt
    -0.84
     círculo
    -0.82
    POSITIVE LOGITS
     involvement
    2.23
     participation
    1.99
     role
    1.83
    particip
    1.77
     participate
    1.77
     roles
    1.60
    participation
    1.52
    参与
    1.51
     participates
    1.45
    volvement
    1.42
    Activations Density 0.098%

    No Known Activations