© 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. 57753
    Prev
    Next
    INDEX
    Explanations

    (India) or (Japan)

    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
    åħ³æ³¨
    -0.10
    ä¸Ģæł¹
    -0.09
    (coder
    -0.09
     ÐļÑĥ
    -0.09
    éĤł
    -0.09
    /sbin
    -0.09
    纡
    -0.09
    argar
    -0.09
    egl
    -0.09
    azar
    -0.09
    POSITIVE LOGITS
    éħIJ
    0.09
    è±ģ
    0.09
     Yak
    0.09
    æ§ģ
    0.08
     structured
    0.08
    èĤĩ
    0.08
    鹧
    0.08
    osh
    0.08
    erts
    0.07
     Dra
    0.07
    Activations Density 0.160%

    No Known Activations