© 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. 53572
    Prev
    Next
    INDEX
    Explanations

    really a

    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
    PageRoute
    -0.08
    CONTEXT
    -0.08
    åĴĮåħ¶ä»ĸ
    -0.08
     bill
    -0.07
     ’
    -0.07
    é¢Ķ
    -0.07
    åľ¨åħ¶
    -0.07
    доб
    -0.07
    çĻ»
    -0.07
     Disconnect
    -0.07
    POSITIVE LOGITS
    éļĽ
    0.10
    æī¶
    0.09
    å®ŀä½ĵç»ıæµİ
    0.09
    æ¿¡
    0.08
    BOSE
    0.08
    çīĪæĿĥå½Ĵ
    0.08
    æµ·åĨħå¤ĸ
    0.08
    mı
    0.08
    olley
    0.08
    inde
    0.08
    Activations Density 0.338%

    No Known Activations