© 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. 22-GEMMASCOPE-RES-131K
    7. 34122
    Prev
    Next
    INDEX
    Explanations

    founder, director, brothers

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

    The main thing this neuron does is detect explicit numerical values and quantity-related terms in the text.

    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
     new
    -1.32
     those
    -1.30
     at
    -1.28
     so
    -1.19
     as
    -1.19
    遺跡
    -1.18
     with
    -1.16
     in
    -1.16
     people
    -1.13
     increase
    -1.10
    POSITIVE LOGITS
     maniere
    1.41
     fonde
    1.17
     marte
    1.14
     attente
    1.13
     raiſ
    1.12
     offrir
    1.11
     occuper
    1.09
     唯美
    1.09
     diriger
    1.09
     vocale
    1.07
    Activations Density 0.150%

    No Known Activations