© 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. 53434
    Prev
    Next
    INDEX
    Explanations

    to win

    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
    寥
    -0.09
    cerr
    -0.09
    ivas
    -0.09
    æ®·
    -0.09
    ãģĦãģŁãģł
    -0.09
    ilog
    -0.09
    çİĭåŃIJ
    -0.08
    æīįæĺ¯
    -0.08
    éĤ£å°±
    -0.08
    è¾ĵäºĨ
    -0.08
    POSITIVE LOGITS
    两èĢħ
    0.10
    åĩ¡
    0.09
    icit
    0.09
    äºĨä¸Ģ
    0.09
    缴è¾ĸå¸Ĥ
    0.09
    ä¿İ
    0.09
    线ä¸Ĭ线ä¸ĭ
    0.08
    èĢĮæĪIJ
    0.08
    arme
    0.08
    æĮ¤
    0.08
    Activations Density 0.087%

    No Known Activations