© 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. 4563
    Prev
    Next
    INDEX
    Explanations

    detention, killings, disappearance

    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
    ê½ĥ
    -0.10
    ascade
    -0.10
     Fancy
    -0.09
    æľ¬æĬ¥
    -0.09
    mium
    -0.09
    ä½ĵåĪ¶æľºåĪ¶
    -0.09
    ëĦĺ
    -0.09
    èıľ
    -0.09
    xis
    -0.09
    æ·»åĬł
    -0.09
    POSITIVE LOGITS
    èĤ©èĨĢ
    0.09
    åĬłåĢį
    0.09
    MUX
    0.09
    åħ¥åħļ
    0.08
    å«Įçĸij
    0.08
    åıĺåİĭ
    0.08
    æİĴæŁ¥
    0.08
     interrog
    0.08
    åŁİå¸Ĥ管çIJĨ
    0.08
    黾
    0.08
    Activations Density 0.073%

    No Known Activations