© 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. 50588
    Prev
    Next
    INDEX
    Explanations

    scaffolding

    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
    iking
    -0.09
    å¼Ĥåij³
    -0.08
    大æĪ·
    -0.08
    æ®ĸ
    -0.08
    elt
    -0.08
    olec
    -0.08
    éī´
    -0.08
     thù
    -0.08
    ased
    -0.08
    lect
    -0.08
    POSITIVE LOGITS
    mates
    0.11
    ural
    0.10
    yard
    0.10
    ä¸ĭéĿ¢æĺ¯å°ı
    0.09
    ermo
    0.09
    alık
    0.09
     Brend
    0.09
    ings
    0.09
    æľĹ
    0.09
    ucus
    0.09
    Activations Density 0.006%

    No Known Activations