© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8183
    Prev
    Next
    INDEX
    Explanations

    legal or illegal activities

    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
    ãģ¯ãĤĤãģ¡ãĤįãĤĵ
    -0.09
    éĹ©
    -0.09
    olle
    -0.09
    åı¯ä¾Ľ
    -0.08
    ìĦ¹
    -0.08
    REW
    -0.08
    山寨
    -0.08
     Malk
    -0.08
    nore
    -0.08
    ORM
    -0.08
    POSITIVE LOGITS
    ä½ľä¸ºä¸Ģç§į
    0.12
    çĤ¹ç¼Ģ
    0.09
    æĺ¯ä¸Ģç§į
    0.09
    端æŃ£
    0.09
    æ½ĩæ´Ĵ
    0.09
    ban
    0.09
    åĺ²ç¬ij
    0.09
    好äºĭ
    0.08
    常æĢģåĮĸ
    0.08
    秽
    0.08
    Activations Density 0.180%

    No Known Activations