© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 1612
    Prev
    Next
    INDEX
    Explanations

    describing items and their qualities

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

    The neuron is detecting the main product or item names (e.g. “Mug,” “Watch,” “SUV,” “Dress,” “Tool Set,” “Mattress,” etc.) in commercial or catalog‐style text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     بودند
    -1.77
     vannak
    -1.56
    模样
    -1.54
    ходили
    -1.52
     lavorato
    -1.47
    teilte
    -1.47
     ſever
    -1.44
     voltak
    -1.41
     raconter
    -1.41
     magát
    -1.41
    POSITIVE LOGITS
     is
    4.09
     will
    3.03
     has
    2.86
     couldn
    2.55
     was
    2.55
     certainly
    2.36
     might
    2.31
     definitely
    2.23
     could
    2.20
     shouldn
    2.20
    Activations Density 0.159%

    No Known Activations