© 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. 3657
    Prev
    Next
    INDEX
    Explanations

    words followed by numbers or punctuation

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

    The neuron detects components of formal named entities—especially multi‐token person names, official titles, and associated dates.

    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
     после
    -1.58
    Ār
    -1.47
     сначала
    -1.27
     dopo
    -1.26
     cualquier
    -1.25
     like
    -1.25
    umque
    -1.25
     Setelah
    -1.23
     normalmente
    -1.21
     such
    -1.20
    POSITIVE LOGITS
     these
    2.20
     lately
    1.32
    These
    1.28
     which
    1.26
    在这里
    1.24
     escribiendo
    1.23
    these
    1.20
    ėd
    1.19
     since
    1.17
    COVID
    1.16
    Activations Density 0.034%

    No Known Activations