© 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. 21580
    Prev
    Next
    INDEX
    Explanations

    staff members

    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
    ç»Īç»ĵ
    -0.09
    æĿ¥åĽŀ
    -0.08
    èĬ¥
    -0.08
    nant
    -0.08
    enet
    -0.08
     hob
    -0.08
     Gi
    -0.08
    ment
    -0.08
    æ±°
    -0.08
    ics
    -0.08
    POSITIVE LOGITS
    acionales
    0.11
    pha
    0.10
    regon
    0.09
    çŁ¥æĥħ
    0.09
    cntl
    0.09
    MouseButton
    0.09
    ä¸Ńæŀ¢
    0.09
    iments
    0.09
    è¹
    0.09
     pragma
    0.09
    Activations Density 0.009%

    No Known Activations