© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 60738
    Prev
    Next
    INDEX
    Explanations

    Todd and toddlers

    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
    Negative Logits
    odos
    -0.09
    zug
    -0.09
    uft
    -0.09
    ми
    -0.09
    nte
    -0.09
    ÑĩаÑĤ
    -0.09
    dos
    -0.09
    fully
    -0.09
    anity
    -0.08
     года
    -0.08
    POSITIVE LOGITS
    ä¼ļå±ķ
    0.09
    ormsg
    0.09
    amate
    0.09
    æŀģ端
    0.09
    ement
    0.09
    ·
    0.09
    aje
    0.09
    èľľ
    0.09
    ably
    0.08
    InSection
    0.08
    Activations Density 0.018%

    No Known Activations