© 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. 30801
    Prev
    Next
    INDEX
    Explanations

    percentage of

    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
    SIDE
    -0.09
    cape
    -0.09
    aday
    -0.09
     ведÑĮ
    -0.09
    enticator
    -0.08
    ickle
    -0.08
     Æ¡n
    -0.08
    ampa
    -0.08
    æĢ§è´¨
    -0.08
    ×ķת
    -0.08
    POSITIVE LOGITS
    人寿
    0.11
    alin
    0.09
    WISE
    0.09
    ëģĶ
    0.09
    sar
    0.09
     Mahm
    0.09
    unch
    0.09
    ewis
    0.09
    aus
    0.08
    -wise
    0.08
    Activations Density 0.023%

    No Known Activations