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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen2.5-7B-IT
    5. 20-MATRYOSHKA-65K
    7. 1355
    Prev
    Next
    INDEX
    Explanations

    common suffixes or vowels

    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
    établissement
    -0.31
    hôtel
    -0.30
    qli
    -0.30
    intérieur
    -0.29
    ropdown
    -0.29
    åħ¬å®īå±Ģ
    -0.28
    échange
    -0.27
    améli
    -0.27
    éĢħ
    -0.27
    atatype
    -0.27
    POSITIVE LOGITS
    a
    0.83
    o
    0.79
    i
    0.77
    e
    0.72
    er
    0.63
    y
    0.58
    es
    0.53
    ed
    0.53
    s
    0.52
    ا
    0.51
    Activations Density 9.942%

    No Known Activations