© 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. 3238
    Prev
    Next
    INDEX
    Explanations

    header

    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
    ¼
    -0.10
    ä¹İ
    -0.10
    nap
    -0.10
    اب
    -0.10
    eur
    -0.09
     itemprop
    -0.09
    °
    -0.09
    rous
    -0.09
    adir
    -0.09
    cape
    -0.09
    POSITIVE LOGITS
    =headers
    0.10
    ä¸įæ¸ħ
    0.10
    ooks
    0.09
    edly
    0.09
    .stamp
    0.09
    çŀ©çĽ®
    0.09
    ooter
    0.09
     Aub
    0.09
    俯
    0.08
    楯
    0.08
    Activations Density 0.099%

    No Known Activations