© 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. 7230
    Prev
    Next
    INDEX
    Explanations

    stuffed animal

    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
    ãģĵãģĿ
    -0.08
    esty
    -0.08
    glass
    -0.08
    ä¸Ģç¬Ķ
    -0.08
    éªĪ
    -0.08
    ä¹ĭæĺŁ
    -0.08
    å°ıä¼Ļ
    -0.08
     Aston
    -0.08
    eps
    -0.08
     Terr
    -0.08
    POSITIVE LOGITS
    iffs
    0.10
    èĥĮéĥ¨
    0.09
    缸å½ĵ
    0.09
    lijk
    0.08
    åĵģ
    0.08
    箪
    0.08
    (circle
    0.08
    ï¸ı
    0.08
    èĥĮ
    0.08
    次ä¼ļè®®
    0.08
    Activations Density 0.007%

    No Known Activations