© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52905
    Prev
    Next
    INDEX
    Explanations

    make the most 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
    amba
    -0.09
    æĢ»ä»·
    -0.09
    æŃ£æĸĩ
    -0.08
    æģķ
    -0.08
    ENTER
    -0.08
    gent
    -0.08
    FromBody
    -0.08
     treÅĽci
    -0.08
     Warfare
    -0.08
    afil
    -0.08
    POSITIVE LOGITS
    éĻIJ度
    0.10
    è¿Ļ个æĹ¶ä»£
    0.10
    éĻIJ
    0.09
    APER
    0.09
     Existing
    0.09
    åıijæĮ¥
    0.09
    inish
    0.09
    ablish
    0.09
    ä»ĬçĶŁ
    0.08
    çİ°æľīçļĦ
    0.08
    Activations Density 0.007%

    No Known Activations