© 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. 56781
    Prev
    Next
    INDEX
    Explanations

    special

    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
    خاÙħ
    -0.09
    _UNSUPPORTED
    -0.09
     outright
    -0.09
     ÑģобÑģÑĤвенно
    -0.08
    积水
    -0.08
    æĦ¦
    -0.08
    amed
    -0.08
    hir
    -0.08
    letes
    -0.08
    ä¸Ģä½ĵåĮĸ
    -0.08
    POSITIVE LOGITS
    çī¹æ®Ĭ
    0.14
     special
    0.14
    -special
    0.14
     SPECIAL
    0.13
    .Special
    0.13
    Special
    0.12
     Special
    0.12
    çī¹æ®ĬçļĦ
    0.12
    special
    0.12
    .special
    0.11
    Activations Density 0.123%

    No Known Activations