© 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. 128817
    Prev
    Next
    INDEX
    Explanations

    The neuron fires strongly on the main topic noun that concludes or titles each document section.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    nouns followed by prepositions

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     キッズ
    -1.14
    tentu
    -1.09
     genoux
    -1.06
     vêtements
    -1.05
     tratamientos
    -1.04
     retraso
    -1.04
     boissons
    -1.02
    mische
    -1.02
     rayures
    -1.02
     cadeaux
    -1.00
    POSITIVE LOGITS
     by
    2.81
     to
    2.23
     with
    1.77
     into
    1.64
     via
    1.52
     through
    1.45
     mediante
    1.33
     as
    1.31
     according
    1.30
     bằng
    1.30
    Activations Density 0.107%

    No Known Activations