© 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. 22-GEMMASCOPE-RES-131K
    7. 14680
    Prev
    Next
    INDEX
    Explanations

    named entities and their descriptors

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

    The neuron strongly activates on formal, encyclopedic language—particularly institutional, organizational, and location names and related technical nouns.

    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
     créer
    -1.67
     currently
    -1.63
     ২০১
    -1.56
     today
    -1.54
     смарт
    -1.36
    今年
    -1.35
     isn
    -1.35
     aujourd
    -1.33
     yesterday
    -1.32
     ۲۰۱
    -1.30
    POSITIVE LOGITS
     тогда
    1.66
     damals
    1.55
     when
    1.52
    estos
    1.38
     even
    1.38
     still
    1.35
    鏖
    1.34
     had
    1.32
     There
    1.31
     would
    1.30
    Activations Density 0.056%

    No Known Activations