© 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. 57398
    Prev
    Next
    INDEX
    Explanations

    wonder, marvel, miracle

    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
    uting
    -0.09
    æĦıè§ģ
    -0.09
    æĦŁæĥħ
    -0.09
    ting
    -0.09
    éĺ´
    -0.08
    åĢ¾
    -0.08
    æľįåĬ¡åijĺ
    -0.08
    æ±
    -0.08
    presso
    -0.08
    URITY
    -0.08
    POSITIVE LOGITS
     marvel
    0.13
    ç¥ŀå¥ĩ
    0.13
    å¥ĩ迹
    0.12
    æĥĬå¥ĩ
    0.11
    ç¥ŀéĢļ
    0.11
     wonder
    0.11
     wonders
    0.11
    Amazing
    0.10
    æĥĬåı¹
    0.10
    adm
    0.10
    Activations Density 0.039%

    No Known Activations