Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-2B
    3. 20-GEMMASCOPE-TRANSCODER-16K
    4. 14094
    Prev
    Next
    INDEX
    Explanations

    words associated with facilities and processes, encompassing legal, sports, penal, maintenance and travel contexts, and verbs ending with 'ing'.

    oai_token-act-pair · gemini-2.0-flash

    room, house

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope/layer_20/width_16k
    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
    RectangleBorder
    -0.52
    ształ
    -0.49
    êmica
    -0.47
    utilisons
    -0.47
    GIVEREF
    -0.45
    ύπ
    -0.45
    뀐
    -0.43
    omock
    -0.43
    haped
    -0.42
    ("}\
    -0.42
    POSITIVE LOGITS
     room
    1.26
     rooms
    1.20
    house
    1.17
    room
    1.17
     house
    1.16
    rooms
    1.09
     Room
    1.03
     ROOM
    1.02
    houses
    1.02
     Rooms
    1.01
    Activations Density 1.987%

    No Known Activations