© 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. 35946
    Prev
    Next
    INDEX
    Explanations

    impress, impressed, impressible

    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
    åĴª
    -0.09
    æ¾Į
    -0.09
    éĥ¨åĪĨåĨħ容
    -0.08
     Kirby
    -0.08
    ä¸Ģ个éĹ®é¢ĺ
    -0.08
    ization
    -0.08
     Liv
    -0.08
    taÅŁ
    -0.08
    åħħåĢ¼
    -0.08
    çIJ¶
    -0.08
    POSITIVE LOGITS
    ç¼Ń
    0.10
    ingly
    0.10
    dc
    0.10
    ively
    0.10
    и
    0.09
    ses
    0.09
    èĭ£
    0.09
    æģĭçĪ±
    0.09
    éĶħ
    0.09
    åĿĬ
    0.09
    Activations Density 0.003%

    No Known Activations