© 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. 72740
    Prev
    Next
    INDEX
    Explanations

    phrases often following specific starter words

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

    The neuron detects and flags the capitalized words that form the titles or names of events, articles, or other headings.

    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
    峮
    -1.35
    衄
    -1.21
     boissons
    -1.21
     žal
    -1.17
     Besoin
    -1.17
     autorisation
    -1.16
     męskie
    -1.16
    Jeffrey
    -1.16
    enzo
    -1.16
    ubahan
    -1.15
    POSITIVE LOGITS
     a
    1.32
    ata
    1.29
     an
    1.28
     different
    1.23
    clerosis
    1.21
     There
    1.20
     got
    1.18
    y
    1.18
     on
    1.16
    \_
    1.16
    Activations Density 0.074%

    No Known Activations