© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 33356
    Prev
    Next
    INDEX
    Explanations

    promoting and preserving

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

    The neuron identifies infinitive goal‐ or mission‐statement phrases (e.g. “to foster,” “to encourage,” “to promote”) that express an organization’s purpose.

    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
    ícia
    -1.55
    breviations
    -1.38
     consente
    -1.27
    ႐
    -1.23
    kkor
    -1.23
    แค่
    -1.23
     antemano
    -1.23
    columnHeader
    -1.20
     continuación
    -1.19
    ệu
    -1.16
    POSITIVE LOGITS
    Noten
    1.32
     its
    1.10
    itism
    1.08
     byen
    1.08
     related
    1.07
    Thro
    1.06
     TÉCNICA
    1.05
     线条
    1.04
     promouvoir
    1.04
    Ingrédients
    1.04
    Activations Density 0.071%

    No Known Activations