© 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. 3801
    Prev
    Next
    INDEX
    Explanations

    folding and unfolding

    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
     ÑħоÑĤÑĮ
    -0.10
    _HT
    -0.10
    unc
    -0.10
    пÑĥ
    -0.09
    ä¸Ģéĥ¨
    -0.09
    è¡Ķ
    -0.09
    éĨĴ
    -0.09
    èĸ°
    -0.08
    æ²¥
    -0.08
     Alv
    -0.08
    POSITIVE LOGITS
    æĬĺåıł
    0.10
    褶
    0.10
    .fold
    0.10
     Folding
    0.09
    -fold
    0.09
     Fold
    0.09
     folded
    0.08
     folding
    0.08
    enet
    0.08
     orig
    0.08
    Activations Density 0.025%

    No Known Activations