© 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. 33694
    Prev
    Next
    INDEX
    Explanations

    direct or directly

    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
    æ³®
    -0.09
    .mac
    -0.09
    ä¸įçŁ¥ä¸įè§ī
    -0.09
    æ´»è·ĥ
    -0.09
    /socket
    -0.09
     окол
    -0.08
    åĬ¿
    -0.08
    æµħ
    -0.08
     Conc
    -0.08
    æĪIJé¾Ļ
    -0.08
    POSITIVE LOGITS
    缴æİ¥
    0.26
     directly
    0.19
     direct
    0.18
    direct
    0.18
    åı¯ä»¥çĽ´æİ¥
    0.17
     Direct
    0.16
    Direct
    0.16
     direkt
    0.16
     ì§ģìłij
    0.15
     trá»±c
    0.15
    Activations Density 0.095%

    No Known Activations