© 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. 30724
    Prev
    Next
    INDEX
    Explanations

    industrial, social, technical, presidential

    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
    _HS
    -0.10
    vig
    -0.09
    obi
    -0.08
    antal
    -0.08
     Tir
    -0.08
    并ä¸įèĥ½
    -0.07
    å¢ĥ
    -0.07
    ibles
    -0.07
     sal
    -0.07
    vard
    -0.07
    POSITIVE LOGITS
    åįģ足
    0.11
     ÙĦÙĦغاÙĬØ©
    0.11
     nature
    0.10
    ç½ijç«Ļåľ°åĽ¾
    0.10
    nature
    0.10
    èĢĮä¸į
    0.09
    ité
    0.09
    æľªå©ļ
    0.09
    ä¸Ģç«Ļ
    0.09
    åį³æŶåıijçĶŁ
    0.09
    Activations Density 0.178%

    No Known Activations