© 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. 4524
    Prev
    Next
    INDEX
    Explanations

    singing and carols

    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
    ault
    -0.10
    loon
    -0.10
    oen
    -0.09
    eway
    -0.09
    æķ¢
    -0.08
    .msg
    -0.08
     mk
    -0.08
    EXTERN
    -0.08
    ê¶ģ
    -0.08
    _mk
    -0.08
    POSITIVE LOGITS
    åĶ±æŃĮ
    0.12
    åĶ±
    0.11
    æŃĮ声
    0.10
    ä¸Ģé¦ĸ
    0.10
     sheet
    0.10
    /GPL
    0.09
    æŃĮæĽ²
    0.09
    è¿Ļé¦ĸæŃĮ
    0.09
     firm
    0.09
    表æĥħ
    0.09
    Activations Density 0.038%

    No Known Activations