© 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. 54743
    Prev
    Next
    INDEX
    Explanations

    one at a time

    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
    ган
    -0.11
    âŁ
    -0.09
    éŃĶ
    -0.09
    DDR
    -0.09
    nets
    -0.09
    èĥ½çľĭåĩº
    -0.09
    åħ«åį¦
    -0.09
     gee
    -0.09
    cling
    -0.09
     munch
    -0.08
    POSITIVE LOGITS
    ä¸ĢçĤ¹çĤ¹
    0.10
    ä¸ĢåĿĹ
    0.09
    éĢł
    0.09
    çĽ²çĽ®
    0.09
    èĭij
    0.09
    æĻ·
    0.09
    好å¥ĩ
    0.09
    ä¸ŀ
    0.09
    踱
    0.09
    ç«¿
    0.08
    Activations Density 0.034%

    No Known Activations