© 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. 55338
    Prev
    Next
    INDEX
    Explanations

    risk factors

    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
    éģŃ
    -0.09
    æİĪ
    -0.09
    éĢ¢
    -0.09
     pac
    -0.08
    åħļæł¡
    -0.08
    æĺł
    -0.08
    Gratis
    -0.08
    æī«
    -0.08
    èįī
    -0.08
    nz
    -0.08
    POSITIVE LOGITS
     factors
    0.11
    åĽłç´ł
    0.11
    æ°Ľ
    0.10
    åĽłåŃIJ
    0.10
    risk
    0.09
    Ñĥни
    0.09
    erre
    0.09
    Ļ¨
    0.09
    æĭİ
    0.09
     Ris
    0.09
    Activations Density 0.300%

    No Known Activations