© 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. 63016
    Prev
    Next
    INDEX
    Explanations

    best followed by a noun

    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
    çIJ¼
    -0.09
    AML
    -0.09
    éģĵ士
    -0.09
    esti
    -0.09
    åIJŀ
    -0.08
    å®Ľ
    -0.08
    å»ĵ
    -0.08
    wis
    -0.08
    é©°
    -0.08
     Wis
    -0.08
    POSITIVE LOGITS
    ä¸ĭæ²ī
    0.10
    ä¸ĭåįĬ
    0.09
    ä»»ä½ķå½¢å¼ı
    0.09
     whitelist
    0.08
    å±Ī
    0.08
    æµģåIJij
    0.08
     Sink
    0.08
    GES
    0.08
    cale
    0.08
     Worldwide
    0.08
    Activations Density 0.011%

    No Known Activations