© 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. 31837
    Prev
    Next
    INDEX
    Explanations

    permission to write external

    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
    isl
    -0.09
     GER
    -0.09
    ãng
    -0.08
     Maurice
    -0.08
    fort
    -0.08
    assage
    -0.08
    Revision
    -0.08
    si
    -0.08
    èŀ³
    -0.08
     charging
    -0.08
    POSITIVE LOGITS
    æĺİçıł
    0.09
    uddy
    0.09
    trag
    0.09
     obscene
    0.08
    忽
    0.08
    è¾ļ
    0.08
     Pratt
    0.08
    åıĤä¸İ
    0.08
    举é£İ
    0.08
    Incre
    0.08
    Activations Density 0.022%

    No Known Activations