© 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. 98662
    Prev
    Next
    INDEX
    Explanations

    made of fine materials

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

    The neuron fires on language describing how a product is made or what materials it’s made of (e.g. “made from,” “crafted,” “manufactured,” etc.).

    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
    vermectin
    -1.45
     to
    -1.32
    ebly
    -1.30
     Bemerkungen
    -1.28
     until
    -1.23
    ridescent
    -1.23
    ighed
    -1.21
     at
    -1.20
    igth
    -1.20
    unno
    -1.18
    POSITIVE LOGITS
     soigneusement
    1.30
     provenant
    1.21
     indeb
    1.13
    wój
    1.12
     endast
    1.12
     remis
    1.11
    чки
    1.07
     entusias
    1.06
    海报
    1.05
     الفلك
    1.05
    Activations Density 0.004%

    No Known Activations