© 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. 3693
    Prev
    Next
    INDEX
    Explanations

    eyelids or related suffixes

    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
    ick
    -0.10
    s
    -0.09
    sad
    -0.08
    enal
    -0.08
    enic
    -0.08
    iles
    -0.08
    en
    -0.08
    on
    -0.08
    ä¿Ŀåį«
    -0.08
    ¤
    -0.08
    POSITIVE LOGITS
    rary
    0.11
    rides
    0.10
    borg
    0.09
    åİ¢
    0.09
    åİļåİļ
    0.09
    å¹Ķ
    0.09
    ãĤĩ
    0.09
    otes
    0.09
    åħ³ä¹İ
    0.09
    ustral
    0.09
    Activations Density 0.014%

    No Known Activations