© 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. 16944
    Prev
    Next
    INDEX
    Explanations

    in-house technical development

    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
    елÑĮ
    -0.09
    ä»İä¸Ń
    -0.09
    .persistent
    -0.09
     Baum
    -0.08
    _AI
    -0.08
    æīĢåľ¨
    -0.08
    *=*=
    -0.08
    HDR
    -0.08
    _cached
    -0.08
    åħ±åIJĮä½ĵ
    -0.08
    POSITIVE LOGITS
    (internal
    0.10
    itage
    0.10
    /Internal
    0.10
    人士
    0.09
    åĬŁè¯¾
    0.09
    /by
    0.09
    -Out
    0.08
    /internal
    0.08
    åħīæ»ij
    0.08
    icorn
    0.08
    Activations Density 0.013%

    No Known Activations