© 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. 34040
    Prev
    Next
    INDEX
    Explanations

    bulk followed by a noun

    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
    çħ²
    -0.10
    èĦī
    -0.09
    ä¸ĸä¸Ĭ
    -0.08
     Hurt
    -0.08
    robots
    -0.08
     bells
    -0.08
    åł¡åŀĴ
    -0.08
    寤
    -0.08
    å±¥
    -0.08
    ä¼ļç»Ļ
    -0.08
    POSITIVE LOGITS
    iliary
    0.11
    itarian
    0.11
    éºĴ
    0.10
    isateur
    0.10
    ier
    0.10
    åĪģ
    0.09
    ify
    0.09
    erness
    0.09
    éĨį
    0.09
    çļĦéĤ£ç§į
    0.09
    Activations Density 0.013%

    No Known Activations