© 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. 35139
    Prev
    Next
    INDEX
    Explanations

    announcement

    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
    å³Ļ
    -0.13
    ä¾Ī
    -0.10
    -0.10
    宽容
    -0.10
    ousse
    -0.10
    åĮĸè¿Ľç¨ĭ
    -0.09
    ä¸įä½ı
    -0.09
    ivation
    -0.09
    Ĭ
    -0.09
    ãĢħ
    -0.09
    POSITIVE LOGITS
    ä¸ĸ人
    0.10
    دÙĨ
    0.09
    ãĥ¡ãĥ³ãĥĪ
    0.09
    паÑĢ
    0.09
    ions
    0.09
    roy
    0.08
    ments
    0.08
    odable
    0.08
    骥
    0.08
    è·¸
    0.08
    Activations Density 0.022%

    No Known Activations