© 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. 1534
    Prev
    Next
    INDEX
    Explanations

    path

    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
    afür
    -0.30
    uforia
    -0.27
     tranny
    -0.27
     prostit
    -0.25
     spoilers
    -0.24
    ;br
    -0.23
     libertin
    -0.23
    ucch
    -0.23
     helicopt
    -0.22
     UIAlert
    -0.22
    POSITIVE LOGITS
     path
    0.25
    åij¸
    0.24
    UTILITY
    0.23
    ways
    0.23
    WAY
    0.22
    way
    0.22
     paths
    0.22
     modalitÃł
    0.21
    ournemouth
    0.21
    èĩ³æѤ
    0.21
    Activations Density 6.066%

    No Known Activations