© 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. 28145
    Prev
    Next
    INDEX
    Explanations

    symmetric symmetry

    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
    efs
    -0.09
    ç¬ij声
    -0.09
    ohl
    -0.08
    å¼Ĥè®®
    -0.08
    vre
    -0.08
    ests
    -0.08
    ocre
    -0.08
    å£ij
    -0.08
    å¥ĸåѦéĩij
    -0.08
     grocery
    -0.08
    POSITIVE LOGITS
    ally
    0.12
    -minded
    0.10
    ÑĤÑĢо
    0.09
    persistent
    0.09
    ulla
    0.09
    /as
    0.09
    èĽĽ
    0.09
    metrical
    0.08
    æłª
    0.08
    èĻ¹
    0.08
    Activations Density 0.046%

    No Known Activations