© 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. 34-GEMMASCOPE-RES-131K
    7. 11775
    Prev
    Next
    INDEX
    Explanations

    entity followed by location

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

    The neuron fires strongly on salient topic words and named entities—key content-bearing nouns (like “Gaza,” “astrology,” “bitcoin,” “Niagara Falls,” product or person names, technical terms, 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_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
     stedet
    -1.25
     getragen
    -1.06
     zieken
    -1.05
    歺
    -1.04
     invités
    -1.02
     tellement
    -1.01
     fabriqué
    -1.01
     spectateurs
    -1.00
    ఝ
    -0.99
     espesor
    -0.98
    POSITIVE LOGITS
     in
    3.27
     at
    2.48
     on
    2.36
     within
    2.02
     throughout
    1.95
     during
    1.80
     across
    1.60
     elsewhere
    1.55
    ใน
    1.55
     via
    1.32
    Activations Density 0.683%

    No Known Activations