© 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. 52521
    Prev
    Next
    INDEX
    Explanations

    XY, XXX, X followed by words

    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
    并éĿŀ
    -0.09
    IOD
    -0.09
    羣å®ŀçļĦ
    -0.09
    iglia
    -0.08
    cw
    -0.08
    èªī
    -0.08
    eva
    -0.08
    Presence
    -0.08
    uto
    -0.08
    MMC
    -0.08
    POSITIVE LOGITS
    æŁIJæŁIJ
    0.17
    æŁIJ
    0.17
    æŁIJ个
    0.15
    æŁIJä¸Ģ
    0.13
     blah
    0.12
    blah
    0.11
     XYZ
    0.11
     etc
    0.11
    æŁIJç§į
    0.10
    çŃīçŃī
    0.10
    Activations Density 0.365%

    No Known Activations