© 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. 33579
    Prev
    Next
    INDEX
    Explanations

    beauty and related concepts

    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
    ible
    -0.10
    ion
    -0.10
    举缣
    -0.09
    Synopsis
    -0.09
    sWith
    -0.09
    çĶ·å¥³
    -0.08
    éĥ¨åĪĨåĨħ容
    -0.08
    ster
    -0.08
    bons
    -0.08
    POSITIVE LOGITS
    fully
    0.12
    ful
    0.11
    çĸĻ
    0.11
    lection
    0.09
    ะ
    0.09
    lessness
    0.09
    folk
    0.09
    anded
    0.09
    shots
    0.09
     Gör
    0.08
    Activations Density 0.010%

    No Known Activations