© 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. 18382
    Prev
    Next
    INDEX
    Explanations

    Additionally, similarly, or special characters

    np_acts-logits-general · gemini-2.5-flash-lite

    legal and technical disclaimers, particularly sentences containing conditional statements and restrictions about liability or usage rights.

    oai_token-act-pair · claude-4-5-haikuTriggered by @yooniel31

    sentences or fragments from legal, copyright, licensing, or policy/disclaimer text (formal legal wording).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    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
     pil
    -0.08
    ointment
    -0.07
    æĻ®éĢļçļĦ
    -0.07
    æıŃ
    -0.07
    Pref
    -0.07
    pile
    -0.07
     spo
    -0.07
    æĬĵ
    -0.07
    ÂŃ
    -0.07
    好åĩłä¸ª
    -0.07
    POSITIVE LOGITS
    éĸ£
    0.10
    éªļ
    0.09
    çĿ
    0.09
     PaÅĦstwo
    0.09
     ë°ĺëĵľìĭľ
    0.08
    侵害
    0.08
    æĽĻåħī
    0.08
    æĨ
    0.08
    èĿ
    0.08
    å¦Ĥæľī
    0.08
    Activations Density 0.179%

    No Known Activations