© 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. 39689
    Prev
    Next
    INDEX
    Explanations

    Sanj Chalab Boshqbak Chabab Chovancova Chaudhuri Chonburi Chaba

    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
    åĨįéĢł
    -0.09
    æĥ³åĥı
    -0.09
    æĥ³è±¡
    -0.09
    ãģ¤ãģĦ
    -0.09
    chal
    -0.09
    æľŁæľĽ
    -0.09
    addir
    -0.09
    oid
    -0.08
    ardo
    -0.08
    é¤IJ
    -0.08
    POSITIVE LOGITS
    åĤį
    0.10
    .SwingConstants
    0.09
    ACTER
    0.09
    à¤ĸ
    0.09
    sticks
    0.09
    edin
    0.09
    .fi
    0.09
    UGHT
    0.09
    mania
    0.08
    wick
    0.08
    Activations Density 0.088%

    No Known Activations