© 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. 63182
    Prev
    Next
    INDEX
    Explanations

    institutions and affiliations

    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
    .bp
    -0.10
    æĹ¶ä»£ä¸ŃåĽ½
    -0.09
    åĮ»ç§ij
    -0.08
    æ³¾
    -0.08
    rah
    -0.08
    å§Ĭ
    -0.08
    ä»»ä½ķä¸Ģ个
    -0.08
    ippy
    -0.08
    æĬĬèĩªå·±çļĦ
    -0.08
    æĴ®
    -0.08
    POSITIVE LOGITS
    _dirty
    0.10
    .dirty
    0.09
    BOSE
    0.09
    è¿Ī
    0.09
    ваÑĤ
    0.09
    LOB
    0.08
    asco
    0.08
    _BOOT
    0.08
    -LAST
    0.08
    idy
    0.08
    Activations Density 0.151%

    No Known Activations