© 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. 56530
    Prev
    Next
    INDEX
    Explanations

    contractions like workin' or swingin'

    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
    s
    -0.09
    ing
    -0.09
    ×ķת
    -0.08
    gne
    -0.08
    ä½Ĩä¸įéĻIJ
    -0.08
    nt
    -0.08
    -ish
    -0.08
    ags
    -0.08
    ors
    -0.07
    es
    -0.07
    POSITIVE LOGITS
    asti
    0.10
    iated
    0.09
    媳
    0.09
    าย
    0.09
    stractions
    0.09
    ÑıÑĤ
    0.09
    trainer
    0.08
    eways
    0.08
    åѤåĦ¿
    0.08
    èĢģ人家
    0.08
    Activations Density 0.026%

    No Known Activations