© 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. Qwen3.5-9B Base
    5. 15-QWENSCOPE-RES-64K
    7. 60610
    Prev
    Next
    INDEX
    Explanations

    damage to components

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ugno
    -0.08
    ä¸įå½±åĵį
    -0.08
    -BEGIN
    -0.08
    .kr
    -0.08
    .Disabled
    -0.07
    è¿·ç³Ĭç³Ĭ
    -0.07
    å¹²æī°
    -0.07
    кÑĥÑĢ
    -0.07
    讨
    -0.07
    çĻ¼éħµ
    -0.07
    POSITIVE LOGITS
     damage
    0.12
    æī¿åıĹçļĦ
    0.10
    为äºĨä¿ĿæĬ¤
    0.10
    æįŁä¼¤
    0.10
     overload
    0.09
    æī¿åıĹ
    0.09
     Damage
    0.09
     изна
    0.09
     overst
    0.09
     damaging
    0.09
    Activations Density 0.118%

    No Known Activations