© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 50839
    Prev
    Next
    INDEX
    Explanations

    street followed by related terms

    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
    Ùij
    -0.09
    urd
    -0.09
    à¹Ī
    -0.09
    åĽ¾çº¸
    -0.09
    ê¹IJ
    -0.09
    ous
    -0.09
     Brend
    -0.08
     Sunder
    -0.08
    queued
    -0.08
    oded
    -0.08
    POSITIVE LOGITS
    lights
    0.12
    ë§Ŀ
    0.10
    è¯Ŀ说
    0.09
    ä½ķåĨµ
    0.09
    å··
    0.09
    cars
    0.09
    walker
    0.09
    iful
    0.09
    blocks
    0.08
    lining
    0.08
    Activations Density 0.023%

    No Known Activations