© 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. 32132
    Prev
    Next
    INDEX
    Explanations

    wreak havoc, inflict damage, consequences

    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
    çļĦä¸Ģ大
    -0.10
    ÑģÑĤой
    -0.10
    缴è§Ĥ
    -0.08
    ÑĪки
    -0.08
    consum
    -0.08
    worth
    -0.08
     lain
    -0.08
    çĤ«èĢĢ
    -0.08
    bable
    -0.08
    ches
    -0.08
    POSITIVE LOGITS
     havoc
    0.13
    zeug
    0.10
    iting
    0.10
    ä»Ģä¹Īæł·çļĦ
    0.10
    uate
    0.10
    osg
    0.10
    å·´æĭī
    0.10
    tam
    0.09
    åĽŀå¿Ĩ
    0.09
    rips
    0.09
    Activations Density 0.016%

    No Known Activations