© 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. 84551
    Prev
    Next
    INDEX
    Explanations

    free from, tested, and attributes

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

    The neuron detects marketing “badge” phrases that assert product certifications or “free‐from” and “tested” claims.

    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
    ighed
    -1.70
    nadequate
    -1.66
    ridescent
    -1.52
     immagine
    -1.42
    nappropriate
    -1.34
     for
    -1.31
    FontAwesomeIcon
    -1.30
     at
    -1.30
    carus
    -1.29
    vermectin
    -1.26
    POSITIVE LOGITS
     ܐ
    1.33
    ֺ
    1.31
     ویتامین
    1.31
    箭头
    1.27
     hvil
    1.26
     keten
    1.23
     benzin
    1.23
     perpé
    1.23
    Doesn
    1.21
    lampu
    1.19
    Activations Density 0.011%

    No Known Activations