© 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. 39128
    Prev
    Next
    INDEX
    Explanations

    Francis Informa Ltd, Polity Books

    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
    èī¯
    -0.09
    usat
    -0.09
    ä¸įèī¯
    -0.09
    itle
    -0.09
    emet
    -0.08
    å®®
    -0.08
     kê
    -0.08
    á»®
    -0.08
    ÑĥÑĪ
    -0.08
    itters
    -0.08
    POSITIVE LOGITS
    æīĪ
    0.09
    èĥĮä¸Ĭ
    0.09
    è¿«
    0.08
    -submit
    0.08
    èĩªè¡Į
    0.08
    _CRITICAL
    0.08
    _critical
    0.08
    çĹŀ
    0.08
    èĽĩ
    0.08
    çľĭå¾Ĺ
    0.08
    Activations Density 0.018%

    No Known Activations