© 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. 10-GEMMASCOPE-RES-131K
    7. 75573
    Prev
    Next
    INDEX
    Explanations

    carcinogenic and mutagenic potential

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

    words related to toxicity, specifically terms like "carcinogenic," "mutagenic," and "teratogenic" that describe harmful effects of substances.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron flags technical terms naming chemical hazards or toxic effects (e.g., carcinogenic, mutagenic, teratogenic, genotoxic).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    了许多
    -1.37
    co
    -1.36
    了不少
    -1.35
     tinte
    -1.33
     welchen
    -1.32
    us
    -1.32
    了很多
    -1.31
    菪
    -1.30
    king
    -1.27
    lin
    -1.27
    POSITIVE LOGITS
     ersten
    1.37
     you
    1.34
     efekt
    1.23
     reklama
    1.22
     kérdés
    1.22
     zeggen
    1.21
    Inggris
    1.21
    braz
    1.20
    mercado
    1.15
    emplares
    1.15
    Activations Density 0.098%

    No Known Activations