© 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. 58819
    Prev
    Next
    INDEX
    Explanations

    then followed by a verb

    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
    åĪĽå®¢
    -0.11
    ฤ
    -0.10
    mys
    -0.09
    apsed
    -0.09
    æĹ©æĻļ
    -0.09
    MMM
    -0.09
    èij±
    -0.09
    å¸Į
    -0.09
     Ful
    -0.09
    ths
    -0.09
    POSITIVE LOGITS
    etu
    0.09
    oux
    0.08
    ylene
    0.08
    ãģĹãģĭãģªãģĦ
    0.08
    ick
    0.07
    s
    0.07
     consequ
    0.07
    ull
    0.07
    ¤
    0.07
    .tv
    0.07
    Activations Density 0.024%

    No Known Activations