© 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. 27485
    Prev
    Next
    INDEX
    Explanations

    previous/next navigation

    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
    ç´¯
    -0.09
    -times
    -0.09
    åī
    -0.09
    彤
    -0.09
    人群ä¸Ń
    -0.08
    hound
    -0.08
    åīįç½®
    -0.08
    å¿Ļç¢Į
    -0.08
    ç½ijæ°ij
    -0.08
    æĵħ
    -0.08
    POSITIVE LOGITS
    (previous
    0.09
    previous
    0.09
    atical
    0.09
    å®ŀè´¨æĢ§
    0.09
     Previous
    0.08
    _Des
    0.08
    roller
    0.08
    ãĥıãĥ³
    0.08
     previous
    0.08
    éĺµåľ°
    0.08
    Activations Density 0.033%

    No Known Activations