© 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. 5222
    Prev
    Next
    INDEX
    Explanations

    mail to or mailbox

    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
    人æ°Ķ
    -0.09
    èĢķåľ°
    -0.08
    éĩijçīĮ
    -0.08
    edia
    -0.08
    ixer
    -0.08
    æű
    -0.08
    ä¿Ŀå§Ĩ
    -0.08
    çľĭäºĨä¸Ģ
    -0.08
    _AHB
    -0.08
    qli
    -0.08
    POSITIVE LOGITS
    chimp
    0.14
    room
    0.12
    boxes
    0.12
    bag
    0.11
    ings
    0.11
    ableObject
    0.11
    rooms
    0.10
    otte
    0.10
    worm
    0.10
    æŃĩ
    0.10
    Activations Density 0.022%

    No Known Activations