© 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. 2771
    Prev
    Next
    INDEX
    Explanations

    so-called

    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
     pus
    -0.09
    èij©
    -0.09
    urable
    -0.09
    å°ıå¿ĥ
    -0.09
    >tag
    -0.08
    èĩªæ²»åĮº
    -0.08
    交éĶĻ
    -0.08
    åĬ¨
    -0.08
    çĶŁæľº
    -0.08
    _IOCTL
    -0.08
    POSITIVE LOGITS
    ä¸Ģ个éĹ®é¢ĺ
    0.10
    åľ¨æł¡
    0.10
    å¸ĤéĿ¢ä¸Ĭ
    0.09
    ä¸Ģè¶Ł
    0.09
    广åijĬæľįåĬ¡
    0.09
    umed
    0.09
    ignment
    0.08
    ilent
    0.08
    idth
    0.08
    енÑĮк
    0.08
    Activations Density 0.031%

    No Known Activations