© 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. 3369
    Prev
    Next
    INDEX
    Explanations

    treatment failed

    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
    çĮ¿
    -0.13
    _PAD
    -0.10
     erotica
    -0.10
    ç¾½æ¯Ľ
    -0.09
    oyer
    -0.09
     ká»ĭp
    -0.09
    ODB
    -0.09
    æ·±åĪĩ
    -0.09
    Õª
    -0.09
    ä¹Łæľīä¸ĢäºĽ
    -0.09
    POSITIVE LOGITS
    æĹłæķĪ
    0.16
    屡
    0.14
    å°Ŀè¯ķ
    0.14
     attempts
    0.13
     failed
    0.12
    å¤ļ次
    0.11
    没æľīåĬŀæ³ķ
    0.11
    失败
    0.11
     fails
    0.11
     fail
    0.11
    Activations Density 0.237%

    No Known Activations