© 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. 61444
    Prev
    Next
    INDEX
    Explanations

    spyware, spy club, spy plane

    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
    -bodied
    -0.09
    ality
    -0.08
    ê¶Į
    -0.08
    ary
    -0.08
    end
    -0.08
    ibrator
    -0.08
     Gregg
    -0.08
    ion
    -0.08
    è¹Ī
    -0.08
     ÑģайÑĤа
    -0.08
    POSITIVE LOGITS
    glass
    0.11
    writing
    0.11
    gram
    0.10
    TEX
    0.10
    abwe
    0.09
    æĢ»å±Ģ
    0.09
    antha
    0.09
    æľĽçĿĢ
    0.09
    pping
    0.09
    icates
    0.09
    Activations Density 0.004%

    No Known Activations