© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9700
    Prev
    Next
    INDEX
    Explanations

    concerto, accounts, theft, function, formulation, computers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    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
    äºĶè¡Į
    -0.11
    _mt
    -0.09
    kur
    -0.08
    erala
    -0.08
    æ¶Įçİ°åĩº
    -0.08
    çĥ½
    -0.08
    æĹłå£°
    -0.08
    è¿Ĥ
    -0.08
    èĻ®
    -0.08
     Olsen
    -0.08
    POSITIVE LOGITS
    ä¸ĢåĿĹ
    0.10
    orum
    0.10
    æīĭä¸Ĭ
    0.09
    åIJ¾
    0.09
    æķ£
    0.09
    æ³µ
    0.09
    çĴŀ
    0.08
    æ·¡
    0.08
    è´¡
    0.08
    jour
    0.08
    Activations Density 0.105%

    No Known Activations