© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 352
    Prev
    Next
    INDEX
    Explanations

    technical code usage

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    chanind/qwen2.5-7B-it-layer-20-saes/pile/matryoshka/k-100
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    oriously
    -0.23
    oldem
    -0.23
    olem
    -0.22
    eworthy
    -0.21
    \application
    -0.21
     ×Ķ×ķ×ĵ
    -0.20
    odial
    -0.20
    lete
    -0.20
    uben
    -0.19
    alez
    -0.19
    POSITIVE LOGITS
    ’B
    0.19
    åİŁæĸĩåľ°åĿĢ
    0.19
    éĥħ
    0.19
    ons
    0.18
    unitÃł
    0.18
    vertiser
    0.18
    Ø£Ùģر
    0.18
    prene
    0.17
     corner
    0.17
    itation
    0.17
    Activations Density 0.367%

    No Known Activations