© 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. 33573
    Prev
    Next
    INDEX
    Explanations

    analytics platform features

    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
    ities
    -0.10
    ality
    -0.10
    è¾ĥ大çļĦ
    -0.10
    LY
    -0.10
    oral
    -0.10
    大éĩıçļĦ
    -0.09
    åĿĢ
    -0.09
     lá»Ńa
    -0.09
    æºIJæºIJ
    -0.09
    Ark
    -0.09
    POSITIVE LOGITS
    enic
    0.09
    ÑĢÑıд
    0.09
    çĿ¬
    0.09
    quot
    0.09
    -ready
    0.09
    angi
    0.09
    黥
    0.08
     quot
    0.08
    doll
    0.08
    xic
    0.08
    Activations Density 0.018%

    No Known Activations