© 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. 10-GEMMASCOPE-RES-131K
    7. 8103
    Prev
    Next
    INDEX
    Explanations

    specific nouns or technical terms

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

    The neuron activates on capitalized proper names and titles (e.g. band names, product or software names, show titles).

    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
    ஂ
    -1.67
     magari
    -1.54
     hende
    -1.50
    趿
    -1.49
     etek
    -1.49
     fossero
    -1.46
     estremamente
    -1.46
     henne
    -1.45
     gambe
    -1.44
     innamor
    -1.44
    POSITIVE LOGITS
     in
    1.52
     is
    1.52
     Since
    1.43
     describe
    1.41
     my
    1.38
     must
    1.37
    .
    1.34
    酝
    1.34
    tiary
    1.33
     decipher
    1.33
    Activations Density 0.152%

    No Known Activations