© 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. 18643
    Prev
    Next
    INDEX
    Explanations

    sack

    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
    缨
    -0.10
    常å§Ķ
    -0.09
    çļĦ身ä½ĵ
    -0.09
    ìłĢ
    -0.09
    ise
    -0.09
    ike
    -0.08
    iment
    -0.08
    说æľį
    -0.08
    ken
    -0.08
    ìĤ¬íķŃ
    -0.08
    POSITIVE LOGITS
    cloth
    0.12
    å¥ł
    0.11
    æİī
    0.10
     hạng
    0.09
    ties
    0.09
    regation
    0.09
    utton
    0.09
    ses
    0.09
    nowled
    0.09
    æºŁ
    0.09
    Activations Density 0.006%

    No Known Activations