© 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. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Qwen2.5-7B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 30565
    Prev
    Next
    INDEX
    Explanations

    Q&A forum posts

    np_max-act · gemini-2.0-flash

    It detects tokens that indicate the user is asking for key information (requests for main points, summaries, lists, specs, or use-case-style answers).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    達
    -0.07
    ``↵
    -0.07
    =""></
    -0.07
    只要你
    -0.07
    舒适的
    -0.07
    逄
    -0.07
    二十
    -0.06
    *)↵↵
    -0.06
     diferente
    -0.06
     sending
    -0.06
    POSITIVE LOGITS
    utschen
    0.08
     bếp
    0.07
     פרס
    0.07
     Bris
    0.07
    学习成绩
    0.07
    积淀
    0.07
    uforia
    0.07
     Grund
    0.07
     tslib
    0.07
    grading
    0.07
    Activations Density 0.300%

    No Known Activations