© 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. 49094
    Prev
    Next
    INDEX
    Explanations

    MPL or AOL emails

    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
    attle
    -0.09
    æļ´
    -0.08
     Clinic
    -0.08
    itch
    -0.08
    主人
    -0.08
    åľ¨å¤ĸéĿ¢
    -0.08
    zelf
    -0.07
    èµĦ产è¯Ħä¼°
    -0.07
     consequ
    -0.07
    仪
    -0.07
    POSITIVE LOGITS
    loyd
    0.11
    ä½İä¸ĭ
    0.10
    ches
    0.09
    ê·Ģ
    0.09
    诰
    0.09
    éĶ¢
    0.09
    igne
    0.09
    åħ¶éĹ´
    0.08
    çªŁ
    0.08
    inha
    0.08
    Activations Density 0.001%

    No Known Activations