© 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. 3616
    Prev
    Next
    INDEX
    Explanations

    in style, usual, done, eye, tracks, guns, job, occasion, side

    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
    icot
    -0.11
    .uml
    -0.10
    /ajax
    -0.10
    azzi
    -0.10
    usted
    -0.09
    agus
    -0.09
    ocked
    -0.09
    undy
    -0.09
    abei
    -0.09
    ullo
    -0.09
    POSITIVE LOGITS
    5
    0.09
    å¸ļ
    0.09
    4
    0.09
    èĪ¬çļĦ
    0.09
    6
    0.09
    2
    0.09
    çĿ¹
    0.09
    8
    0.09
    1
    0.08
    7
    0.08
    Activations Density 0.519%

    No Known Activations