© 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. 33018
    Prev
    Next
    INDEX
    Explanations

    pipeline type, sampling clock, LT3081

    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
    åĽ¢
    -0.09
    æ³±
    -0.08
    onom
    -0.08
     machine
    -0.08
     disc
    -0.08
    俾
    -0.08
    æIJŀ好
    -0.08
    æ³¥
    -0.08
    sphere
    -0.08
    EST
    -0.07
    POSITIVE LOGITS
    iens
    0.11
     Ven
    0.09
    WithMany
    0.09
    è´§æºIJ
    0.09
    ç®
    0.09
    çĥŃæIJľ
    0.09
    .ba
    0.09
    _aspect
    0.09
    /li
    0.09
     Spirits
    0.09
    Activations Density 0.198%

    No Known Activations