© 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. 58159
    Prev
    Next
    INDEX
    Explanations

    how works

    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
    çĪ±ç¾İ
    -0.10
    æķ´æķ´
    -0.10
    ä»Ĭ天å°ıç¼ĸ
    -0.09
    è´¨éĩıåĴĮ
    -0.09
    è¿Ļ家ä¼Ļ
    -0.09
    åįģäºĮäºĶ
    -0.09
    çĮ¬
    -0.09
    ä¸Ģä¸Ŀ
    -0.09
    æķ´å½¢
    -0.09
    ÃĤ
    -0.09
    POSITIVE LOGITS
    OUS
    0.10
    _HOOK
    0.10
    ï¼Ł↵↵
    0.10
    ixon
    0.09
    (&$
    0.08
    ðŁĩµ
    0.08
    Mir
    0.08
     Pip
    0.08
    Virgin
    0.08
    undi
    0.08
    Activations Density 0.966%

    No Known Activations