© 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. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 19410
    Prev
    Next
    INDEX
    Explanations

    exploit, exploit, attack, vulnerability

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    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
    orme
    -0.07
    Ñħоз
    -0.06
     belast
    -0.05
    vet
    -0.05
    æĭĸ
    -0.05
    è¿ĻæĶ¯
    -0.05
    due
    -0.05
    нÑĥли
    -0.05
    æijĦ
    -0.05
    æ´
    -0.05
    POSITIVE LOGITS
    atas
    0.06
    atta
    0.06
    actus
    0.06
    ATA
    0.06
    atee
    0.05
     linker
    0.05
    åĿļåĨ³
    0.05
    å°º
    0.05
    寸
    0.05
    ciato
    0.05
    Activations Density 0.006%

    No Known Activations