© 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. 32716
    Prev
    Next
    INDEX
    Explanations

    LA Weekly Institute Club Project

    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
    otec
    -0.10
    æĿĵ
    -0.09
    fone
    -0.09
    /pm
    -0.09
    æݨå¼Ģ
    -0.08
    ëįĺ
    -0.08
    obs
    -0.08
    æĬ¤èº«
    -0.08
    éªĹåŃIJ
    -0.08
    ERM
    -0.08
    POSITIVE LOGITS
    æĢĿ
    0.10
    çĤħ
    0.10
    çķĮ
    0.09
    è¿Ļ个è¡Įä¸ļ
    0.09
    ible
    0.09
    è¡ĮæĶ¿éĥ¨éŨ
    0.09
    说æĺ¯
    0.08
     Brothers
    0.08
    uze
    0.08
    æįī
    0.08
    Activations Density 0.002%

    No Known Activations