© 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. 32218
    Prev
    Next
    INDEX
    Explanations

    fantastic followed by positive descriptor

    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
    felt
    -0.10
    iem
    -0.09
    ãģĦãģ¦
    -0.09
    ings
    -0.09
    boat
    -0.09
    ä¹ĭæĪĺ
    -0.09
    ein
    -0.09
    æ¯Ķè¾ĥç®Ģåįķ
    -0.08
    åĴĮå®ĮåĸĦ
    -0.08
    inton
    -0.08
    POSITIVE LOGITS
    ally
    0.15
    ALLY
    0.11
    elli
    0.10
    -looking
    0.10
     Airways
    0.10
    ä¸İåIJ¦
    0.09
    管å±Ģ
    0.09
    ness
    0.09
     manera
    0.09
    ÑģÑĤеÑĢ
    0.09
    Activations Density 0.005%

    No Known Activations