© 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. Qwen3.5-9B Base
    5. 15-QWENSCOPE-RES-64K
    7. 33020
    Prev
    Next
    INDEX
    Explanations

    both/all/each followed by noun or had/of/can/are

    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
    人æĿ¥è¯´
    -0.07
    uitar
    -0.07
    èĢĮå®ļ
    -0.06
    æķħåIJį
    -0.06
    adá
    -0.06
    Spy
    -0.06
    ibri
    -0.06
    å¯ĩ
    -0.06
    adu
    -0.06
     gotten
    -0.06
    POSITIVE LOGITS
     delas
    0.10
    家ä¼Ļ
    0.09
    ussin
    0.08
     undergo
    0.08
     deles
    0.08
    ossi
    0.08
    umba
    0.08
    å®ĥä»¬åľ¨
    0.08
    ayani
    0.08
     Grosso
    0.08
    Activations Density 0.100%

    No Known Activations