© 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. 57547
    Prev
    Next
    INDEX
    Explanations

    dug up or with

    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
    æѦè£ħ
    -0.11
     condom
    -0.10
    ignant
    -0.09
    eren
    -0.09
     кваÑĢ
    -0.09
    æĢ§å¼º
    -0.09
    .nodeType
    -0.09
    è°Ķ
    -0.09
    æħ°
    -0.08
    .Sequential
    -0.08
    POSITIVE LOGITS
    佬
    0.09
    åĬ¡
    0.08
    湫
    0.08
    ocks
    0.08
    ä¸Ģè§Ĵ
    0.08
    0.08
    æĥħæĬ¥
    0.08
    lops
    0.08
    oppel
    0.08
    нак
    0.08
    Activations Density 0.002%

    No Known Activations