© 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. 1699
    Prev
    Next
    INDEX
    Explanations

    specific terms

    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
    nya
    -0.45
    ness
    -0.36
    r
    -0.36
    m
    -0.35
    n
    -0.34
    ra
    -0.33
    mente
    -0.33
    ni
    -0.32
    na
    -0.31
    d
    -0.30
    POSITIVE LOGITS
    Ùĭ
    0.39
    irlines
    0.32
    eh
    0.27
    antino
    0.25
    eid
    0.24
    efficient
    0.24
    upert
    0.24
    ï
    0.23
    ev
    0.23
    å°Ķ
    0.23
    Activations Density 6.146%

    No Known Activations