© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7764
    Prev
    Next
    INDEX
    Explanations

    never give

    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
    å·²ç»ıæľī
    -0.11
    æŃ£åľ¨è¿Ľè¡Į
    -0.11
    éĺ¡
    -0.11
    å·²ç»ıæĺ¯
    -0.10
    Ñĥж
    -0.10
    å·²æľī
    -0.10
    å·²æĺ¯
    -0.09
     already
    -0.09
    å·²ç»ıåľ¨
    -0.09
    ä»ĸå·²ç»ı
    -0.09
    POSITIVE LOGITS
    theless
    0.14
     EVER
    0.13
    never
    0.12
    -ending
    0.12
     NEVER
    0.11
    Never
    0.10
    æľīä»»ä½ķ
    0.10
    ä»İæľª
    0.10
     Never
    0.10
     never
    0.10
    Activations Density 0.106%

    No Known Activations