© 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. 17712
    Prev
    Next
    INDEX
    Explanations

    proud of

    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
    yre
    -0.10
    èĩ´
    -0.09
    çĶ¨å¿ĥ
    -0.09
    èĸª
    -0.09
    å®ŀäºĭ
    -0.09
    ä»Ģä¹Īä¸ľè¥¿
    -0.09
    åıĸå¾Ĺ
    -0.09
    è°ij
    -0.09
    ä»Ģä¹Īäºĭ
    -0.09
    çļĦçĥŃæĥħ
    -0.09
    POSITIVE LOGITS
    uby
    0.10
    dit
    0.09
    è¡Įä¸ļåıijå±ķ
    0.09
    åĨįä¸Ģ次
    0.09
    ause
    0.09
    人éĢł
    0.08
    æĩĴ
    0.08
    人为
    0.08
    URT
    0.08
    /help
    0.08
    Activations Density 0.013%

    No Known Activations