© 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. 21325
    Prev
    Next
    INDEX
    Explanations

    debate or discussion

    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
    xis
    -0.09
    æ©±æŁľ
    -0.09
    _compiler
    -0.09
    ä¿¡æģ¯æľįåĬ¡
    -0.08
    çĽ¸ä¼¼
    -0.08
    æī¿åĬŀ
    -0.08
    åĢı
    -0.08
    -nav
    -0.08
    åºı
    -0.08
    çī¹æľĹ
    -0.08
    POSITIVE LOGITS
     debate
    0.26
    äºī论
    0.25
    讨论
    0.23
    辩论
    0.22
     debates
    0.22
     Debate
    0.22
     debated
    0.20
     discussion
    0.18
    deb
    0.17
     discussions
    0.17
    Activations Density 0.096%

    No Known Activations