© 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. 72844
    Prev
    Next
    INDEX
    Explanations

    finding patterns or sequences

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

    This neuron fires on prominent topic words in document and section headings (e.g. key terms in titles, lists, and headers).

    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
    numOf
    -1.24
     médec
    -1.23
    Utilisateur
    -1.21
    ab
    -1.20
     飲み
    -1.18
     распоря
    -1.17
     idéia
    -1.17
     romántico
    -1.17
    稍微
    -1.17
     marginally
    -1.17
    POSITIVE LOGITS
    וּ
    1.61
     článek
    1.36
     wła
    1.34
     takže
    1.31
    ían
    1.30
    logías
    1.30
     веке
    1.29
     всех
    1.29
    ungkap
    1.27
    tiktok
    1.27
    Activations Density 0.030%

    No Known Activations