© 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. 422
    Prev
    Next
    INDEX
    Explanations

    always

    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
     currently
    -0.28
     possibly
    -0.23
     many
    -0.23
    åĪĿæŃ¥
    -0.22
    缮åīį
    -0.22
    æ¯ı天
    -0.22
    /legal
    -0.22
    æĻ®éģį
    -0.21
     often
    -0.21
    exels
    -0.20
    POSITIVE LOGITS
     Asked
    0.23
     egret
    0.23
     EÄŁitim
    0.22
     Alloy
    0.22
     Dysfunction
    0.22
    -ending
    0.21
    被èªī
    0.21
    ìĶ©
    0.21
    theless
    0.20
    -times
    0.20
    Activations Density 1.668%

    No Known Activations