© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 67720
    Prev
    Next
    INDEX
    Explanations

    Category Wikipedia entries

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

    The neuron is specially triggered by the “Category” lines at the ends of Wikipedia‐style entries—i.e. it spots the metadata category tags (and their year/name tokens).

    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
     actrice
    -1.28
    が無い
    -1.18
     LIRE
    -1.15
     MINERAL
    -1.13
     kayaks
    -1.05
     VIDEOT
    -1.05
     hunde
    -1.05
     handicaps
    -1.05
     oublié
    -1.04
     vanil
    -1.03
    POSITIVE LOGITS
     with
    1.23
     правительства
    1.05
     after
    1.03
     as
    1.01
    there
    0.99
     there
    0.99
    휘
    0.98
     İ
    0.97
    mal
    0.96
    sear
    0.96
    Activations Density 0.007%

    No Known Activations