© 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. 10802
    Prev
    Next
    INDEX
    Explanations

    masses and multitudes

    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
     Braun
    -0.11
    åĪĿè¡·
    -0.09
    ä¸Ģä½į
    -0.09
    ä¼ļ使
    -0.08
    loc
    -0.08
    icas
    -0.08
    å¤ļä½į
    -0.08
    abled
    -0.08
    ode
    -0.08
    lobal
    -0.08
    POSITIVE LOGITS
    енн
    0.12
    å¼ĢæĿ¥
    0.11
    é¢Ĩè¢ĸ
    0.09
    mma
    0.09
    åįī
    0.09
    èĭ±çī¹
    0.09
    çĨŁçŁ¥
    0.09
    cripts
    0.09
    ingers
    0.09
    èĤ¤
    0.09
    Activations Density 0.023%

    No Known Activations