© 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. 55314
    Prev
    Next
    INDEX
    Explanations

    end of clause punctuation

    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
     Spirits
    -0.10
    åĿ¦
    -0.10
    侨
    -0.10
    [block
    -0.10
    å¼Ģ端
    -0.10
     spirits
    -0.10
    SCALL
    -0.09
    ç¡ĸ
    -0.09
    SCII
    -0.09
    toa
    -0.09
    POSITIVE LOGITS
    orm
    0.08
    ict
    0.08
    åĬĽåѦ
    0.07
     pres
    0.07
    ä¸įçĦ¶
    0.07
    asel
    0.07
    å¾Ī
    0.07
    åIJĦé¡¹å·¥ä½ľ
    0.07
    ä¸į容æĺĵ
    0.07
    ä¸įæĺĵ
    0.07
    Activations Density 0.120%

    No Known Activations