© 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. 52049
    Prev
    Next
    INDEX
    Explanations

    bowling, bowler, bowls

    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
    å¼Ģä¼ļ
    -0.10
    æ¿Ĵ
    -0.09
    åĩºå¤Ħ
    -0.09
     pil
    -0.09
     Pil
    -0.09
    代åĬŀ
    -0.09
    ç¼Ļ
    -0.09
    ä¼ļè°Ī
    -0.09
    棨
    -0.09
    erah
    -0.09
    POSITIVE LOGITS
     bowling
    0.10
    èŁĭ
    0.10
    arel
    0.10
    èĬ¦
    0.10
     Bowling
    0.09
    sek
    0.09
    大éģĵ
    0.09
    åİħ
    0.09
    åºĦ
    0.09
    æű
    0.08
    Activations Density 0.058%

    No Known Activations