© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 17009
    Prev
    Next
    INDEX
    Explanations

    colors

    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
    Disclosure
    -0.10
    æĢ§å¼º
    -0.09
     Laur
    -0.09
    åºĬä¸Ĭ
    -0.09
    resa
    -0.09
    itat
    -0.09
    妻
    -0.09
    Ñģл
    -0.09
    åĹij
    -0.08
     pena
    -0.08
    POSITIVE LOGITS
    ìĥī
    0.14
    â¾Ĭ
    0.12
    åŁºè°ĥ
    0.12
    /or
    0.12
    -colored
    0.10
    é²ľèī³
    0.10
    -haired
    0.10
    èī²
    0.10
    ê¹Ķ
    0.10
    /black
    0.10
    Activations Density 0.103%

    No Known Activations