© 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. 18549
    Prev
    Next
    INDEX
    Explanations

    romantic followed by genre or relationship terms

    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
    idor
    -0.09
    ÑĩаÑĤ
    -0.09
    oui
    -0.09
    æĶ¹éĿ©åıijå±ķ
    -0.09
    ima
    -0.08
    亮çĤ¹
    -0.08
    agher
    -0.08
    ses
    -0.08
    imal
    -0.08
    igner
    -0.08
    POSITIVE LOGITS
    主ä¹īæĢĿæĥ³
    0.14
    ting
    0.12
    ized
    0.11
    ismo
    0.11
    izing
    0.10
    ization
    0.10
    ism
    0.10
    主ä¹ī
    0.10
    æ¹¾åĮº
    0.10
    ÃŃsimo
    0.10
    Activations Density 0.003%

    No Known Activations