© 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. 27097
    Prev
    Next
    INDEX
    Explanations

    specialty fields and products

    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
    ness
    -0.11
    airo
    -0.10
    ly
    -0.10
    zon
    -0.09
    atchet
    -0.09
    asco
    -0.09
    -quarters
    -0.09
    oland
    -0.09
    aining
    -0.08
    patrick
    -0.08
    POSITIVE LOGITS
    chine
    0.10
    оÑĢ
    0.09
    aylor
    0.08
    oret
    0.08
    annah
    0.08
    suming
    0.08
    artment
    0.08
    ç¨Ĺ
    0.08
    dur
    0.08
    ana
    0.08
    Activations Density 0.004%

    No Known Activations