© 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. 28216
    Prev
    Next
    INDEX
    Explanations

    verbs introducing claims or beliefs

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

    The neuron fires on words that attribute ideas or opinions—i.e. verbs like “assume,” “believe,” “argue,” “thought,” “named,” “view,” etc., which introduce or report theories, beliefs, or judgments.

    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
    Tabela
    -1.09
     {},
    -1.07
     [],
    -1.05
    ഔ
    -1.04
    ַּ
    -1.03
    laştır
    -1.02
     různ
    -1.00
    ִּ
    -0.98
    ᾧ
    -0.98
     diferite
    -0.95
    POSITIVE LOGITS
     that
    1.27
     even
    1.19
     only
    1.16
    стоит
    1.09
     chré
    1.06
    埜
    1.01
     Unterricht
    1.00
     to
    0.99
     too
    0.99
    omorphisms
    0.99
    Activations Density 0.024%

    No Known Activations