© 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. 36352
    Prev
    Next
    INDEX
    Explanations

    kneeling, crawling, or lying down

    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
    仡
    -0.09
    utan
    -0.09
     Feeling
    -0.09
    éĶ£
    -0.09
    wide
    -0.08
    æĥ³èµ·
    -0.08
    å¼łè´´
    -0.08
    åĬ¨æīĭ
    -0.08
    wid
    -0.08
    沫
    -0.08
    POSITIVE LOGITS
    èĻ±
    0.10
    lings
    0.09
    ấn
    0.09
    ä¸Ģå±Ĥ
    0.09
    æĮĩå°ĸ
    0.09
    æĭľ
    0.09
    主ç¼ĸ
    0.08
    .synthetic
    0.08
    é»Ħéĩij
    0.08
    олÑı
    0.08
    Activations Density 0.009%

    No Known Activations