© 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. 48368
    Prev
    Next
    INDEX
    Explanations

    reflects and shows

    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
    liness
    -0.12
    fight
    -0.09
    frey
    -0.09
    oe
    -0.09
    illas
    -0.09
    è¯Ŀ说
    -0.08
    å®ķ
    -0.08
    ç¹ģ
    -0.08
    ÑıÑĤи
    -0.08
    иÑĤеÑĤ
    -0.08
    POSITIVE LOGITS
    ively
    0.13
    äºĨæĪijçļĦ
    0.12
    lops
    0.10
    ivity
    0.09
    åıijå±ķæĪĺçķ¥
    0.09
    OrCreate
    0.09
    .MouseEvent
    0.09
    äºĨ个
    0.09
    ives
    0.09
     Gore
    0.09
    Activations Density 0.012%

    No Known Activations