© 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. 34521
    Prev
    Next
    INDEX
    Explanations

    tax

    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
    æĿ¥çĶµ
    -0.10
    его
    -0.10
    ìĥģìĿĦ
    -0.09
    ocket
    -0.09
    abei
    -0.09
    èĨĪ
    -0.08
    ç©Ĺ
    -0.08
    atter
    -0.08
    æĢ»ä¹¦è®°
    -0.08
    åıĹ访
    -0.08
    POSITIVE LOGITS
    çŃ¹åĪĴ
    0.09
    æľįä»İ
    0.09
    onomy
    0.09
    onomic
    0.08
    ä¼Ĺ
    0.08
    emm
    0.08
    fur
    0.08
     Lance
    0.08
     dock
    0.08
    rms
    0.08
    Activations Density 0.044%

    No Known Activations