© 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. 5499
    Prev
    Next
    INDEX
    Explanations

    pre- or post- prefixes

    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
     Ø¥ÙĨÙĩ
    -0.09
    rogen
    -0.09
    è¯Ŀ说
    -0.08
    ngen
    -0.08
    pton
    -0.08
    riel
    -0.08
    åģ¶
    -0.08
    æķıæį·
    -0.08
    çĨ¨
    -0.08
    .qml
    -0.08
    POSITIVE LOGITS
    éľĵ
    0.10
    å¤ļåįĬ
    0.10
    åĮĸè¿Ľç¨ĭ
    0.09
    ç¼ļ
    0.09
    ulates
    0.09
    ·»
    0.09
    çī¹æ®ĬæĥħåĨµ
    0.09
    asca
    0.09
    å®ŀä½ĵ
    0.09
    å®ŀè´¨æĢ§
    0.09
    Activations Density 0.042%

    No Known Activations