© 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. 30905
    Prev
    Next
    INDEX
    Explanations

    frequency identification

    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
    _VO
    -0.09
    ç»Ħç»ĩçĶŁæ´»
    -0.08
    urer
    -0.08
    ç²¾å¯Ĩ
    -0.08
    ðŁı»
    -0.08
    åĴĮåıijå±ķ
    -0.08
    osit
    -0.08
    ä¼Ļ
    -0.08
     Kemp
    -0.08
    æĦ¿
    -0.08
    POSITIVE LOGITS
    è¦ĭãģ¦
    0.10
    åħĢ
    0.09
    оде
    0.09
    ØŃا
    0.09
    ãģİ
    0.09
    åIJİæŀľ
    0.09
     Äijá»Ļc
    0.09
    仿
    0.09
    æĪijèĩªå·±
    0.08
    лев
    0.08
    Activations Density 0.064%

    No Known Activations