© 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. 57302
    Prev
    Next
    INDEX
    Explanations

    tangy, rich, tart flavors

    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
    æī¹
    -0.09
    æĿ¯
    -0.09
    arger
    -0.08
    家åĸ»
    -0.08
    纵横
    -0.08
    éĻ¤å¤ĸ
    -0.08
    沫
    -0.08
     Piece
    -0.08
    åķ¤éħĴ
    -0.08
     Jog
    -0.08
    POSITIVE LOGITS
    éļIJéļIJ
    0.09
    æľīçĤ¹
    0.09
    ith
    0.09
    åĩłåĪĨ
    0.09
    è¾Ł
    0.09
    edor
    0.09
     somewhat
    0.08
    ä¹Ĥ
    0.08
    干线
    0.08
    éļ½
    0.08
    Activations Density 0.017%

    No Known Activations