© 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. 103766
    Prev
    Next
    INDEX
    Explanations

    Confederate, Bay, film, bal

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

    This neuron responds strongly to biomedical substance and condition terms—particularly peptide or drug names (e.g. “calcitonin”) and toxicity/adverse‐event words.

    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
    tween
    -1.26
    こいつ
    -1.23
    </h2>
    -1.23
     wszystkich
    -1.19
     =
    -1.17
     its
    -1.15
    他の
    -1.15
     więks
    -1.13
     miljoen
    -1.13
    すでに
    -1.12
    POSITIVE LOGITS
     there
    1.63
     ruine
    1.42
     having
    1.32
     encontrarás
    1.31
     any
    1.31
     trovano
    1.28
     doInBackground
    1.28
     Conch
    1.26
    [
    1.25
     trov
    1.24
    Activations Density 0.120%

    No Known Activations