© 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. 5661
    Prev
    Next
    INDEX
    Explanations

    describing the process

    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
    们
    -0.11
    spath
    -0.10
    åĢij
    -0.10
    éĿ¢åŃIJ
    -0.10
    çļĦåīįæıIJä¸ĭ
    -0.10
    伸æīĭ
    -0.10
    ÑģÑĮ
    -0.10
    åīįæıIJ
    -0.09
    stands
    -0.09
    çļĦåīįæıIJ
    -0.09
    POSITIVE LOGITS
    sing
    0.12
    ively
    0.10
    ioned
    0.09
    sed
    0.09
    æµģç¨ĭ
    0.09
    лаÑģ
    0.09
    çijľ
    0.08
    é¢Ĩè·ij
    0.08
    ors
    0.08
    edi
    0.08
    Activations Density 0.033%

    No Known Activations