© 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. 57797
    Prev
    Next
    INDEX
    Explanations

    simple way

    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
    amp
    -0.11
    edo
    -0.09
    wards
    -0.09
    éĵ®
    -0.09
    apo
    -0.09
    uo
    -0.09
    liqu
    -0.09
    çī¹åĪ«å£°æĺİ
    -0.09
    umph
    -0.09
    æŃ»è§Ĵ
    -0.09
    POSITIVE LOGITS
    -minded
    0.20
     minded
    0.17
    xes
    0.15
    tons
    0.14
    ctic
    0.14
    /simple
    0.12
     enough
    0.12
    æī¼
    0.12
    lest
    0.11
    mente
    0.10
    Activations Density 0.067%

    No Known Activations