© 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. 49904
    Prev
    Next
    INDEX
    Explanations

    "-less" or "free" followed by a noun

    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
    å°ĺ
    -0.09
    交åıĭ
    -0.08
    idon
    -0.08
    aban
    -0.08
    Hunter
    -0.08
    lev
    -0.08
     not
    -0.08
    éģķãģĦ
    -0.08
    æİ´
    -0.08
    pac
    -0.08
    POSITIVE LOGITS
    æ¡İ
    0.10
    éĩįç£ħ
    0.09
    _TC
    0.09
    ç«ŀäºīåĬĽ
    0.09
    chant
    0.09
    chal
    0.09
    osen
    0.09
    æ±Ŀ
    0.09
    å®ŀäºĭæ±Ĥæĺ¯
    0.08
     Blackburn
    0.08
    Activations Density 0.087%

    No Known Activations