© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 103066
    Prev
    Next
    INDEX
    Explanations

    storage containers or code repositories

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

    nouns referring to technical objects, systems, or structured containers (like repositories, devices, databases, hoppers, arrays, and applications).

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires on multi‐word technical or domain‐specific noun phrases—i.e. the names of studied objects or entities (like “glomerular epithelial cells,” “houseboat,” or “liquid crystal display device”).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     yesterday
    -1.22
     sobot
    -1.16
     pią
    -1.16
     Ngoài
    -1.14
     mismos
    -1.13
     Mereka
    -1.11
     dovrebbero
    -1.09
     construite
    -1.09
    はもちろん
    -1.07
    íř
    -1.07
    POSITIVE LOGITS
     in
    1.80
     into
    1.48
    ərbaycan
    1.26
     inversa
    1.07
    内の
    1.07
     have
    1.07
     suitable
    1.04
    ύ
    1.03
     așa
    1.03
    føre
    1.03
    Activations Density 0.651%

    No Known Activations