© 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. 22-GEMMASCOPE-RES-131K
    4. 68102
    Prev
    Next
    INDEX
    Explanations

    property name or ID

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    actually
    -1.23
     dando
    -1.21
     všechny
    -1.17
    igth
    -1.14
    ebly
    -1.12
    纐
    -1.10
    excluding
    -1.09
     trám
    -1.08
    rfan
    -1.06
     Befes
    -1.06
    POSITIVE LOGITS
     unique
    1.55
     uniquely
    1.19
     name
    1.16
     while
    1.12
     this
    1.12
     came
    1.07
     even
    1.05
     some
    1.03
    There
    1.02
     identifying
    1.02
    Activations Density 0.018%

    No Known Activations