© 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. 22-GEMMASCOPE-RES-131K
    4. 75021
    Prev
    Next
    INDEX
    Explanations

    register or provide details

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

    The neuron activates on formal service or organization names and their associated program titles (e.g. “Children’s Services,” “kinship care,” 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_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
     garante
    -1.22
     actuellement
    -1.14
    classify
    -1.07
     melk
    -1.07
     ἄ
    -1.04
    髋
    -1.04
     ventes
    -1.03
     garan
    -1.02
     applau
    -1.01
    (");
    -1.01
    POSITIVE LOGITS
     your
    1.88
     then
    1.70
     budete
    1.44
    וּ
    1.40
     approval
    1.38
     before
    1.34
     necessary
    1.33
     documentation
    1.30
     wtedy
    1.28
     vermeld
    1.28
    Activations Density 0.029%

    No Known Activations