© 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. 2920
    Prev
    Next
    INDEX
    Explanations

    code snippets

    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
     beÄŁen
    -0.29
    ombs
    -0.24
    qing
    -0.23
     citiz
    -0.23
    -UA
    -0.23
    rvine
    -0.23
    isseur
    -0.22
    ersistence
    -0.22
    uckland
    -0.22
     Jerseys
    -0.21
    POSITIVE LOGITS
    缩水
    0.22
    çŁ¿æ³īæ°´
    0.19
    .mipmap
    0.19
    licence
    0.19
     atlas
    0.19
    èĻı
    0.18
     “
    0.18
    å¸ļ
    0.18
    çĬ¶æĢģ
    0.18
    æ±²
    0.18
    Activations Density 0.005%

    No Known Activations