© 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. 28232
    Prev
    Next
    INDEX
    Explanations

    rooms within a house

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

    The neuron identifies mentions of architectural elements and specific rooms or features of a house (e.g., garage, porch, kitchen, foyer).

    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
     thereby
    -1.38
    さんへ
    -1.37
     mengatakan
    -1.30
     their
    -1.27
     podían
    -1.20
     triliun
    -1.19
    にします
    -1.13
     since
    -1.09
     květ
    -1.09
     weten
    -1.08
    POSITIVE LOGITS
     is
    1.73
     with
    1.56
    は
    1.23
    これも
    1.23
    はとても
    1.20
     fera
    1.20
    があり
    1.19
     possu
    1.16
    騭
    1.16
     bete
    1.15
    Activations Density 0.012%

    No Known Activations