© 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. 35415
    Prev
    Next
    INDEX
    Explanations

    leaching, trickle, runoff, diffusion

    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
    lico
    -0.09
    csr
    -0.09
    lical
    -0.08
    StackSize
    -0.08
    matic
    -0.08
    å§Ķä¼ļ
    -0.08
    次ä¼ļè®®
    -0.08
    åIJĪãĤıãģĽ
    -0.08
    æ¢Ń
    -0.08
    黾
    -0.08
    POSITIVE LOGITS
    æ¼ı
    0.11
    into
    0.11
    _into
    0.10
    è¿ĩæĿ¥
    0.10
     into
    0.10
    åħ¨æ°ij
    0.09
    夫
    0.09
    ivate
    0.09
    ä½ķåĨµ
    0.09
    Into
    0.09
    Activations Density 0.034%

    No Known Activations