© 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. 69112
    Prev
    Next
    INDEX
    Explanations

    titles, headings, or lists items

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

    This neuron detects headings, i.e. tokens naming or marking HTML heading levels (like “Heading 1,” “Heading 2,” or <h1>–<h6>).

    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
     conquistar
    -1.14
    "}";
    -1.08
     informó
    -1.08
    ὸ
    -1.05
     két
    -1.05
     explí
    -1.04
    色々な
    -1.02
     );
    -1.02
     offensichtlich
    -1.02
     และ
    -1.01
    POSITIVE LOGITS
     refugio
    1.07
     our
    1.05
     particules
    1.02
     sexto
    1.01
     escritor
    0.99
     junto
    0.98
     text
    0.98
    спа
    0.95
     Folge
    0.95
     structure
    0.95
    Activations Density 0.010%

    No Known Activations