© 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. 5360
    Prev
    Next
    INDEX
    Explanations

    it all

    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
     nipples
    -0.10
    pp
    -0.09
    lider
    -0.08
    ä¿Ł
    -0.08
    .inline
    -0.08
    TAIL
    -0.08
    enburg
    -0.08
    unders
    -0.08
    ders
    -0.08
    thur
    -0.08
    POSITIVE LOGITS
    ä¸ĢåĪĩ
    0.33
    ä¸ĢåĪĩéĥ½
    0.28
     everything
    0.27
    çļĦä¸ĢåĪĩ
    0.25
    everything
    0.25
    ä¸ĢåĪĩéĥ½æĺ¯
    0.23
     Everything
    0.23
    è¿Ļä¸ĢåĪĩ
    0.23
    Everything
    0.22
     tudo
    0.22
    Activations Density 0.031%

    No Known Activations