© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50863
    Prev
    Next
    INDEX
    Explanations

    amateur in various fields

    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
    unto
    -0.09
    ä¸įå®ī
    -0.09
    ungs
    -0.09
    iani
    -0.09
     unten
    -0.08
    itto
    -0.08
    /if
    -0.08
    åIJĮå¿Ĺ们
    -0.08
    æ£ļ
    -0.08
    ocab
    -0.08
    POSITIVE LOGITS
    ishly
    0.10
    theless
    0.10
    iful
    0.10
    bib
    0.09
    lic
    0.09
    ships
    0.09
    转载请
    0.09
    grounds
    0.09
    hood
    0.09
    iliary
    0.08
    Activations Density 0.005%

    No Known Activations