© 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 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. 3113
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    Detecting special dialogue/control tokens that mark speaker turns or system/assistant/user message boundaries.

    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
    ديدة
    -0.08
    IDL
    -0.07
     Build
    -0.07
    ROL
    -0.07
    kb
    -0.07
    ليل
    -0.07
    有很多
    -0.07
    _students
    -0.07
    ließlich
    -0.07
    .o
    -0.07
    POSITIVE LOGITS
     varargin
    0.07
    𝓾
    0.07
    ﯮ
    0.07
    متابعة
    0.07
     Intelli
    0.07
    _TRAIN
    0.06
    llvm
    0.06
    CppMethodPointer
    0.06
    ₦
    0.06
    כוונ
    0.06
    Activations Density 0.120%

    No Known Activations