© 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. 17328
    Prev
    Next
    INDEX
    Explanations

    violence and attacks

    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
    éĦ£
    -0.11
    åįİ缼
    -0.09
    åħļåĴĮ
    -0.09
    çŀŃ解
    -0.09
    åĸĺ
    -0.09
    俵
    -0.09
    ä¹Łåıªèĥ½
    -0.08
    FRING
    -0.08
    Descending
    -0.08
     Assass
    -0.08
    POSITIVE LOGITS
    举
    0.10
    çŃĸåĪĴ
    0.10
    æĪij羣çļĦ
    0.09
    触
    0.09
    æīĵæŃ»
    0.09
    æijģ
    0.09
    触åıĬ
    0.09
    触åıij
    0.08
    触åĬ¨
    0.08
    ä¸ĢåIJį
    0.08
    Activations Density 0.122%

    No Known Activations