© 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. 38954
    Prev
    Next
    INDEX
    Explanations

    processing data or signals

    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
    çŃ®
    -0.09
    eras
    -0.09
     Dra
    -0.09
    aldo
    -0.08
    ãĤ¿ãĤ¤
    -0.08
    erc
    -0.08
    .ide
    -0.08
     Braz
    -0.08
    isse
    -0.08
    åħĭ
    -0.08
    POSITIVE LOGITS
    åĬłå·¥
    0.20
     processing
    0.17
    æ·±åĬłå·¥
    0.15
     Processing
    0.14
    processing
    0.14
    å¤ĦçIJĨ
    0.14
    åĬłå·¥åİĤ
    0.14
    Processing
    0.13
     processed
    0.13
    -processing
    0.13
    Activations Density 0.177%

    No Known Activations