© 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. 1396
    Prev
    Next
    INDEX
    Explanations

    fixed rate, difference between

    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
    æ·ĺ
    -0.12
    å¸ĥ
    -0.11
    eded
    -0.09
    责令
    -0.09
    _topology
    -0.09
    åºĶæĢ¥ç®¡çIJĨ
    -0.09
    mani
    -0.08
    奥æĸ¯åį¡
    -0.08
    stops
    -0.08
     PARAMETERS
    -0.08
    POSITIVE LOGITS
    ç»Ĩ
    0.10
    ç»ħ
    0.09
    ä»Ħ
    0.09
    åıĺå¼Ĥ
    0.09
    åŁºåĩĨ
    0.08
    èĦijåŃIJ
    0.08
    ç´°
    0.08
     Arithmetic
    0.08
    ç¿Ĭ
    0.08
    èµŀæĪIJ
    0.08
    Activations Density 0.090%

    No Known Activations