© 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. 18909
    Prev
    Next
    INDEX
    Explanations

    lie or down followed by location

    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
    éĢ®
    -0.11
    oise
    -0.09
    å°ģ
    -0.09
    åĩłåı¥
    -0.09
    移
    -0.09
    ovich
    -0.08
    è²Į
    -0.08
    çĵ´
    -0.08
    umpt
    -0.08
     ÑģÑĤÑĢем
    -0.08
    POSITIVE LOGITS
    æŀķ
    0.09
    èĬĹ
    0.08
    大éģĵ
    0.08
    åIJį声
    0.08
    ÑģÑı
    0.08
    berman
    0.08
     phóng
    0.08
    ecome
    0.08
    Enumerator
    0.08
    Popular
    0.08
    Activations Density 0.019%

    No Known Activations