© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50659
    Prev
    Next
    INDEX
    Explanations

    organ followed by specific suffixes

    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
    sic
    -0.11
    tle
    -0.10
    ayers
    -0.09
    tings
    -0.09
    /from
    -0.09
    ¤
    -0.08
    rogen
    -0.08
    ãĤīãģĦ
    -0.08
    涮
    -0.08
    plevel
    -0.08
    POSITIVE LOGITS
    ically
    0.12
    ized
    0.12
    izational
    0.12
    powered
    0.10
    ization
    0.10
    جاز
    0.09
    åĨħçļĦ
    0.09
    çĺ¤
    0.09
    icism
    0.09
    owie
    0.09
    Activations Density 0.026%

    No Known Activations