© 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. 58869
    Prev
    Next
    INDEX
    Explanations

    spatial capacity

    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
    丰æĶ¶
    -0.10
    æĢİä¹Ī说
    -0.09
    ä¸įå®ļ
    -0.09
     nouve
    -0.09
    ạn
    -0.09
    actory
    -0.09
    å°Ŀ
    -0.09
    ValueChanged
    -0.09
    bilit
    -0.09
    建çŃijçī©
    -0.09
    POSITIVE LOGITS
    壮大
    0.09
    éħį
    0.09
    ish
    0.08
    èĢĮæĿ¥
    0.08
     chiá»ģu
    0.08
    pal
    0.08
     Mitchell
    0.08
    7
    0.08
     rapport
    0.08
     Riders
    0.08
    Activations Density 0.403%

    No Known Activations