© 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. 32731
    Prev
    Next
    INDEX
    Explanations

    separating or cutting across

    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
    urts
    -0.08
    ãģıãĤĬ
    -0.08
    窪
    -0.07
     berb
    -0.07
     orto
    -0.07
    转载èĩª
    -0.07
    åĬŁäºİ
    -0.07
    apped
    -0.07
    fonte
    -0.07
    æıīäºĨ
    -0.07
    POSITIVE LOGITS
    æĭ¦æĪª
    0.10
    æĪª
    0.08
    éĺ»æĸŃ
    0.08
    åħ³åı£
    0.08
     separating
    0.08
    åĪĩå¼Ģ
    0.08
    åĪĩæĸŃ
    0.08
     cutting
    0.08
     inward
    0.07
     chắn
    0.07
    Activations Density 0.008%

    No Known Activations