© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 3339
    Prev
    Next
    INDEX
    Explanations

    position

    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
    )./
    -0.38
    hoot
    -0.32
    åĹIJ
    -0.29
    âļĹ
    -0.29
    printStats
    -0.29
    utow
    -0.28
     $č↵
    -0.28
     sodom
    -0.27
     cuckold
    -0.27
    /Dk
    -0.27
    POSITIVE LOGITS
     position
    0.30
     Position
    0.28
    (position
    0.27
     positions
    0.26
    è·Łå¥¹
    0.25
    (Position
    0.24
    è¦ģä¸įçĦ¶
    0.24
    åĶģ
    0.24
    position
    0.24
    .position
    0.24
    Activations Density 0.142%

    No Known Activations