© 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. 10821
    Prev
    Next
    INDEX
    Explanations

    Sloan

    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
    ties
    -0.13
    icit
    -0.09
    TL
    -0.09
     Äijá»±ng
    -0.08
    ites
    -0.08
    æĪijä¸Ģ缴
    -0.08
    bies
    -0.08
    大ä¼ļ
    -0.08
    bam
    -0.08
    colo
    -0.08
    POSITIVE LOGITS
    utow
    0.11
    eur
    0.10
    åĽŀäºĭ
    0.10
    erb
    0.09
    ожно
    0.09
    空åīį
    0.09
    omas
    0.09
    uby
    0.09
    اÙħØ©
    0.09
    spy
    0.09
    Activations Density 0.001%

    No Known Activations