© 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. 30578
    Prev
    Next
    INDEX
    Explanations

    engine, piston, prop

    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
    é¢ł
    -0.11
    骨
    -0.10
    èµ°å»Ĭ
    -0.09
    åĪĴ
    -0.09
     gang
    -0.08
    å¹²æ¶ī
    -0.08
    bones
    -0.08
     fif
    -0.08
    æĥħ趣
    -0.08
     Red
    -0.08
    POSITIVE LOGITS
    çĸ£
    0.10
    æŁł
    0.10
    aca
    0.09
    -anchor
    0.09
    mant
    0.09
    éĩįä¸Ń
    0.09
    å«ī
    0.09
    离ä¸įå¼Ģ
    0.08
    alink
    0.08
    anut
    0.08
    Activations Density 0.122%

    No Known Activations