© 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. 27475
    Prev
    Next
    INDEX
    Explanations

    its / I / single / it / how / you / why

    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
    åĥĸ
    -0.09
    2
    -0.08
    1
    -0.08
    ä»İèĢĮ使
    -0.08
    ä¸İæѤåIJĮæŶ
    -0.08
    ä¸ĢèĪ¬æĿ¥è¯´
    -0.08
    éĻ¤æѤä¹ĭå¤ĸ
    -0.08
    4
    -0.08
    3
    -0.08
    åģĵ
    -0.08
    POSITIVE LOGITS
    éĹ®éĹ®
    0.10
    _then
    0.09
    è¦ģä¹Ī
    0.09
    å°±æĬĬ
    0.09
    INCT
    0.08
    ضر
    0.08
     either
    0.08
    cup
    0.08
    allon
    0.08
    åħ¨æĺ¯
    0.08
    Activations Density 0.372%

    No Known Activations