© 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. 5642
    Prev
    Next
    INDEX
    Explanations

    salt-related terms

    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
    iston
    -0.10
    举
    -0.09
    qus
    -0.09
    .tk
    -0.09
    naire
    -0.09
    acity
    -0.09
    åĨħåľ°
    -0.09
     bucks
    -0.08
    éĦ¯
    -0.08
    ndo
    -0.08
    POSITIVE LOGITS
    ÑĤеÑĢ
    0.10
    /o
    0.09
    æĺ¨å¤©
    0.09
    ters
    0.09
    ting
    0.09
    ores
    0.09
    less
    0.08
    ening
    0.08
    inez
    0.08
    èĵ¬
    0.08
    Activations Density 0.005%

    No Known Activations