© 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. 34104
    Prev
    Next
    INDEX
    Explanations

    copyright (

    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
    ä¸İæѤ
    -0.11
    åĬŁè¯¾
    -0.10
    éĨī
    -0.09
    ÑĢма
    -0.09
    engo
    -0.09
    lander
    -0.09
    绿èī²åıijå±ķ
    -0.08
    为åħ¶
    -0.08
    éħĿ
    -0.08
    æĺ¾å¾Ĺ
    -0.08
    POSITIVE LOGITS
    izen
    0.09
    çĶŁæ¶¯
    0.08
    icode
    0.08
     Maurice
    0.08
    zy
    0.08
    ÅĤe
    0.08
    iar
    0.08
    è¿Ļ个ä¸ĸçķĮ
    0.08
    æī¿åĮħ
    0.08
    èĪĽ
    0.08
    Activations Density 0.101%

    No Known Activations