© 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. 63174
    Prev
    Next
    INDEX
    Explanations

    toxic compounds or conditions

    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
    /cop
    -0.10
    ersh
    -0.09
     vu
    -0.09
     обеÑģп
    -0.09
    unk
    -0.09
    è¾Ķ
    -0.09
    ourcing
    -0.08
    çľ¼ç§ij
    -0.08
    аÑħ
    -0.08
     поÑĤеÑĢÑı
    -0.08
    POSITIVE LOGITS
    ities
    0.10
    人éĻħåħ³ç³»
    0.09
    æľīæ¯Ĵ
    0.09
    .Exec
    0.09
    æ¯Ĵ
    0.09
    éĶı
    0.09
    æķ£åıijçĿĢ
    0.09
     poisonous
    0.08
    åıĭè°Ĭ
    0.08
    ç½IJ
    0.08
    Activations Density 0.046%

    No Known Activations