© 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. 61843
    Prev
    Next
    INDEX
    Explanations

    capital cities

    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
    ä¿Ŀå®ļ
    -0.10
    åıĪ被
    -0.09
    çī©æĸĻ
    -0.09
    社ä¼ļ稳å®ļ
    -0.09
    æĺ¯æľī
    -0.09
    é£İéĩĩ
    -0.08
    èĭ±æīį
    -0.08
    åħī
    -0.08
    bane
    -0.08
    _LOWER
    -0.08
    POSITIVE LOGITS
    walker
    0.10
    den
    0.09
    æĿ¥çĶµ
    0.09
    isto
    0.09
     hôm
    0.09
    serialized
    0.09
    æ´Ĵ
    0.08
    دس
    0.08
    ä¸Ĭæĸ¹
    0.08
    IPS
    0.08
    Activations Density 0.064%

    No Known Activations