© 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. 1930
    Prev
    Next
    INDEX
    Explanations

    bitter flavors or sensations

    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
    ounter
    -0.10
    ç¿©
    -0.10
    ever
    -0.09
    ULAR
    -0.09
    rick
    -0.09
    bens
    -0.09
    inka
    -0.09
    uos
    -0.09
    ucha
    -0.09
    ç¿ķ
    -0.08
    POSITIVE LOGITS
    ness
    0.11
    fully
    0.10
    è¡·
    0.10
    濯
    0.10
    reat
    0.10
    LY
    0.09
    è°ģçŁ¥éģĵ
    0.09
    éħĬ
    0.09
    çĶŁäº§åĴĮ
    0.09
    è°ĥåīĤ
    0.09
    Activations Density 0.005%

    No Known Activations