© 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. 2818
    Prev
    Next
    INDEX
    Explanations

    spherical shapes

    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
    å¥ĩ
    -0.09
    è¤
    -0.09
    ees
    -0.09
    pac
    -0.09
    uitive
    -0.08
    aurus
    -0.08
    esson
    -0.08
    -heavy
    -0.08
    rch
    -0.08
    ye
    -0.08
    POSITIVE LOGITS
    丸
    0.11
    辫
    0.10
    æĹħ游度åģĩ
    0.10
    磶
    0.10
    èĥį
    0.09
    ä»Ģä¹ĪåİŁåĽł
    0.09
    ä¸ĢåĽ¢
    0.09
    heits
    0.09
    èĹ»
    0.09
    庵
    0.09
    Activations Density 0.048%

    No Known Activations