© 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. 11492
    Prev
    Next
    INDEX
    Explanations

    but not

    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
    åŃ©
    -0.09
    erek
    -0.09
     ав
    -0.09
    åıĸè¯ģ
    -0.09
    人æµģ
    -0.08
    æĬĺèħ¾
    -0.08
    éĢĭ
    -0.08
    é¢ĺæĿIJ
    -0.08
    轺
    -0.08
    è°Ģ
    -0.08
    POSITIVE LOGITS
    åĨ²åĪº
    0.11
    .sul
    0.09
    å¿ħè¦ģ
    0.09
    å®ŀè´¨
    0.09
     necessarily
    0.09
    éľ°
    0.09
    åħ¶ä»ĸ人
    0.09
    hower
    0.09
    KO
    0.08
    äºĭå®ŀä¸Ĭ
    0.08
    Activations Density 0.091%

    No Known Activations