© 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. 58983
    Prev
    Next
    INDEX
    Explanations

    first line of defense

    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
    端æŃ£
    -0.11
    æ¯Ķè¾ĥå¤ļ
    -0.10
    ç«ŀäºī
    -0.09
    izable
    -0.09
    çļĦç«ŀäºī
    -0.09
    (er
    -0.08
     Neb
    -0.08
     GANG
    -0.08
    ä¸ĢåĪ»
    -0.08
    éĿŀ常å¤ļ
    -0.08
    POSITIVE LOGITS
    éĺ²çº¿
    0.19
    ä¸Ģéģĵ
    0.17
    å±ıéļľ
    0.16
    å±Ĥå±Ĥ
    0.12
    å¤ļéĩį
    0.12
    ä¸Ģå±Ĥ
    0.12
    éĺ²æĬ¤
    0.12
    åıĮéĩį
    0.12
     defense
    0.11
    éĺ²åį«
    0.11
    Activations Density 0.069%

    No Known Activations