© 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. 4816
    Prev
    Next
    INDEX
    Explanations

    printer-friendly

    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
    NESS
    -0.11
    家ä¼Ļ
    -0.10
    åľĥ
    -0.09
    æµģ失
    -0.09
    ãģ¾ãģ¾
    -0.09
    Inset
    -0.09
    /plain
    -0.09
    åĨľä¸ļ大åѦ
    -0.09
    wing
    -0.09
    ·
    -0.09
    POSITIVE LOGITS
    -related
    0.09
    hana
    0.08
    lav
    0.08
    åŀ«
    0.08
    dar
    0.08
     Abyss
    0.08
    åĿıäºĭ
    0.08
    villa
    0.08
    tro
    0.08
    orp
    0.08
    Activations Density 0.026%

    No Known Activations