© 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. 50020
    Prev
    Next
    INDEX
    Explanations

    start, then action

    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
    觥
    -0.11
    owie
    -0.10
    éĢıæĺİ
    -0.09
    (PropertyName
    -0.09
    è¿ĻäºĽä¸ľè¥¿
    -0.09
    è¦ģåİ»
    -0.09
    éĩįå¿ĥ
    -0.08
    è¦ģçĶ¨
    -0.08
    èĪĮ头
    -0.08
    .install
    -0.08
    POSITIVE LOGITS
    积累
    0.09
    æ·±åħ¥äººå¿ĥ
    0.08
    马ä¸Ĭ
    0.08
    è¾ĺ
    0.08
    ç¨Ĺ
    0.08
    åħ¼èģĮ
    0.08
    edly
    0.08
     zug
    0.08
    visor
    0.08
    å¡«
    0.08
    Activations Density 0.025%

    No Known Activations