© 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. 61559
    Prev
    Next
    INDEX
    Explanations

    abstract:

    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
    tractive
    -0.09
    çĶŁäº§æĢ»åĢ¼
    -0.08
    åĨįä¹Łæ²¡æľī
    -0.08
    å¼ĢæĿ¥
    -0.08
     Philipp
    -0.08
     Cry
    -0.08
    åĵ¥ä¼¦
    -0.08
    çĿĢä»ĸ
    -0.08
    icz
    -0.08
    路边
    -0.08
    POSITIVE LOGITS
    ä½ĵç³»
    0.10
    å¹´éĻIJ
    0.09
    surname
    0.08
     Stevenson
    0.08
    cks
    0.08
    .onStart
    0.08
    endir
    0.08
    _species
    0.08
     rever
    0.08
    -document
    0.08
    Activations Density 0.011%

    No Known Activations