© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10072
    Prev
    Next
    INDEX
    Explanations

    reader utility

    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
    eds
    -0.10
    ments
    -0.10
    urve
    -0.09
    åĬŀåħ¬åİħ
    -0.09
    æĺŁæĺŁ
    -0.09
    cheme
    -0.09
    æĥ¨
    -0.09
    è¾ĥéĩı
    -0.08
    bie
    -0.08
    è°´
    -0.08
    POSITIVE LOGITS
    odyn
    0.11
    ulet
    0.11
    ymb
    0.10
    ERT
    0.09
     Mage
    0.09
    uhn
    0.09
    errat
    0.09
    天ä¸ĭ
    0.09
    æľĭåıĭ们
    0.09
    atical
    0.08
    Activations Density 0.030%

    No Known Activations