© 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. 11701
    Prev
    Next
    INDEX
    Explanations

    again

    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
    oom
    -0.09
    _rt
    -0.09
    éŀ
    -0.08
    æĮŁ
    -0.08
    å¿ĥæĦ¿
    -0.08
    mino
    -0.08
    ä¸ĢæĹı
    -0.08
     Conte
    -0.08
    _Interface
    -0.08
    oki
    -0.08
    POSITIVE LOGITS
    indre
    0.09
    ä¸Ńå°ı
    0.09
    æ¸ħåĩī
    0.08
    æļij
    0.08
    åįĵè¶Ĭ
    0.08
    è¿Ļæī¹
    0.08
    大æĸ¹
    0.08
    佶
    0.08
    åĨįéĢł
    0.08
    ryn
    0.08
    Activations Density 0.003%

    No Known Activations