© 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. 67707
    Prev
    Next
    INDEX
    Explanations

    collections and descriptions

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

    The neuron fires on the main content words—longer, information‐bearing nouns, verbs, and adjectives—rather than on short function words.

    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
    BeforeAll
    -1.36
    🧍
    -1.34
    🧑
    -1.30
    ℰ
    -1.30
     phénom
    -1.26
    ᅲ
    -1.25
    ABAD
    -1.19
    Дан
    -1.17
     francisco
    -1.16
    formatie
    -1.14
    POSITIVE LOGITS
     where
    1.88
     confortável
    1.30
     people
    1.30
     you
    1.30
     this
    1.29
     assust
    1.29
     one
    1.27
     our
    1.23
     what
    1.23
     tools
    1.23
    Activations Density 0.034%

    No Known Activations