© 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. 3512
    Prev
    Next
    INDEX
    Explanations

    optimization

    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
    polator
    -0.21
     '">
    -0.20
    -Methods
    -0.20
    .ObjectMeta
    -0.18
    æıĦ
    -0.18
    icense
    -0.18
    èħĺ
    -0.17
    ="{!!
    -0.17
    oader
    -0.17
    unda
    -0.17
    POSITIVE LOGITS
    æľĢä¼ĺ
    0.43
    æľĢä½³
    0.43
     optimal
    0.43
    çļĦæľĢä½³
    0.40
    æľĢ大åĮĸ
    0.39
    ä¼ĺåĮĸ
    0.34
     optimize
    0.34
     maximize
    0.33
     optimum
    0.33
    _best
    0.33
    Activations Density 0.267%

    No Known Activations