© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8399
    Prev
    Next
    INDEX
    Explanations

    attraction or action

    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
    åŀĽ
    -0.10
    rys
    -0.10
    æİ¥éĢģ
    -0.09
    íıIJ
    -0.09
    zilla
    -0.09
    大åıij
    -0.09
    ãĢħ
    -0.09
    -padding
    -0.09
    ACITY
    -0.08
    pressor
    -0.08
    POSITIVE LOGITS
    ively
    0.15
    iveness
    0.10
    ¶
    0.09
     Incredible
    0.09
    ees
    0.09
    IBUT
    0.08
    tion
    0.08
    ék
    0.08
    IBUTES
    0.08
    chine
    0.08
    Activations Density 0.025%

    No Known Activations