© 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. 50184
    Prev
    Next
    INDEX
    Explanations

    can't or can say

    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
    æĨ©
    -0.11
    él
    -0.10
    秦çļĩ
    -0.10
    GROUND
    -0.09
    æ±Ĭ
    -0.09
    ickey
    -0.09
    åIJįåı«
    -0.08
    ubb
    -0.08
    çļĦä¸Ģ款
    -0.08
    èĹIJ
    -0.08
    POSITIVE LOGITS
     Kamp
    0.09
     бÑĥм
    0.08
    æĦŁè§īåĪ°
    0.08
    æ±ĤãĤģ
    0.08
    cat
    0.08
    åĨįå¤ļ
    0.08
    æĦ¿
    0.08
    èŁł
    0.08
    è¿ĺä¸įå¤Ł
    0.08
    åIJĮæĦı
    0.08
    Activations Density 0.043%

    No Known Activations