© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 70680
    Prev
    Next
    INDEX
    Explanations

    Substrate, Raze Gaming, JOE, Retribution, SR, RUE

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

    the presence of proper names or other title-case tokens (i.e. named entities and branded terms)

    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
    engager
    -1.54
     Einige
    -1.31
    Goto
    -1.28
     refroid
    -1.28
     Wählen
    -1.20
    Faktor
    -1.20
    Renato
    -1.19
     églises
    -1.19
     enkele
    -1.18
    تحميل
    -1.16
    POSITIVE LOGITS
     și
    1.36
    糴
    1.34
    eados
    1.30
    URS
    1.26
    鹇
    1.20
    '
    1.20
    U
    1.20
     recha
    1.19
    зовый
    1.19
    上の
    1.18
    Activations Density 0.010%

    No Known Activations