© 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. 48242
    Prev
    Next
    INDEX
    Explanations

    ending with ity/-ical/-ural/-ary

    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
    est
    -0.12
    ew
    -0.10
    ings
    -0.10
    æł·çļĦ
    -0.09
    ability
    -0.09
    oret
    -0.09
    tures
    -0.09
    estimated
    -0.09
    ocratic
    -0.09
    abilities
    -0.09
    POSITIVE LOGITS
    ITY
    0.19
    ities
    0.17
    mente
    0.17
    ity
    0.15
    ypse
    0.15
    ized
    0.14
    idad
    0.13
    ité
    0.13
    ness
    0.13
    itas
    0.13
    Activations Density 2.045%

    No Known Activations