© 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. 10620
    Prev
    Next
    INDEX
    Explanations

    temporary

    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
    _TRANSACTION
    -0.09
    zug
    -0.08
    åĶ±
    -0.08
    بدا
    -0.08
     increasingly
    -0.08
    brew
    -0.08
    æħ¢æħ¢çļĦ
    -0.08
    avez
    -0.08
     Americ
    -0.08
    çķħ
    -0.08
    POSITIVE LOGITS
    临æŶ
    0.13
    .tmp
    0.12
    æļĤ
    0.11
    /temp
    0.11
    èIJİ
    0.11
     temporarily
    0.10
    æĬ¬
    0.10
     temporary
    0.10
     tạm
    0.09
    [temp
    0.09
    Activations Density 0.090%

    No Known Activations