© 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. 2057
    Prev
    Next
    INDEX
    Explanations

    will

    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
     future
    -0.22
    buat
    -0.22
     fut
    -0.21
    âĬĦ
    -0.19
    -v
    -0.19
     Wyn
    -0.19
    çļĦ帮åĬ©
    -0.18
     اÙĦعÙĦÙĬ
    -0.18
    -b
    -0.18
    oun
    -0.18
    POSITIVE LOGITS
    Xã
    0.26
    ffield
    0.25
    /Dk
    0.24
    herits
    0.23
    ometown
    0.23
    -validator
    0.22
     ?>"/>↵
    0.22
    (LogLevel
    0.21
    leine
    0.21
    CRY
    0.21
    Activations Density 0.235%

    No Known Activations