© 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. 23121
    Prev
    Next
    INDEX
    Explanations

    last thing, last time, last lap

    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
    รà¸Ńà¸ĩ
    -0.10
    иÑĪ
    -0.10
    æĨ§æĨ¬
    -0.09
    fleet
    -0.09
     поÑģл
    -0.09
    ãģĨãģ¡ãģ«
    -0.09
    allery
    -0.09
    uds
    -0.09
    mediately
    -0.09
    chor
    -0.09
    POSITIVE LOGITS
    most
    0.13
    -ever
    0.11
    ä¸ĢæŃ¥
    0.09
    /latest
    0.09
     thing
    0.09
    -minute
    0.09
    /root
    0.09
     few
    0.09
     resort
    0.08
    born
    0.08
    Activations Density 0.031%

    No Known Activations