© 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. 30247
    Prev
    Next
    INDEX
    Explanations

    con followed by various 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
    emand
    -0.10
    .Butter
    -0.09
    FTER
    -0.09
    æĸŁ
    -0.09
     WTF
    -0.09
    èµ°åĩºæĿ¥
    -0.09
    itarian
    -0.09
    ÑįÑĤомÑĥ
    -0.09
    URITY
    -0.09
    licated
    -0.09
    POSITIVE LOGITS
    éĶĢ
    0.11
    (con
    0.10
    /con
    0.09
    jug
    0.09
    建åħļ
    0.09
    orb
    0.08
    olly
    0.08
    å¾ģæľį
    0.08
    éĢĤ
    0.08
    形象
    0.08
    Activations Density 0.028%

    No Known Activations