© 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. 22493
    Prev
    Next
    INDEX
    Explanations

    open late winter

    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
    èĢį
    -0.10
     Burl
    -0.09
    ượ
    -0.09
    اØŃ
    -0.08
    é©®
    -0.08
    éļľ
    -0.08
     cuer
    -0.08
    siblings
    -0.08
    尾巴
    -0.08
    xc
    -0.08
    POSITIVE LOGITS
    Primitive
    0.09
    Appear
    0.09
    intl
    0.09
    å°ıç»Ħ
    0.09
    çĶ¨æīĭ
    0.08
    å¥ĸåĬ±
    0.08
    ðŁĮ¸
    0.08
     Primitive
    0.08
     THR
    0.08
    饿
    0.08
    Activations Density 0.034%

    No Known Activations