© 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. 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. 28274
    Prev
    Next
    INDEX
    Explanations

    assistant

    np_max-act · gemini-2.0-flash

    Tokens that mark or mention the assistant role/identity (speaker label or assistant name), across languages and capitalizations.

    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
     uc
    -0.07
     Lagos
    -0.07
    .display
    -0.07
     Des
    -0.07
     jeunes
    -0.07
    _debug
    -0.06
    引起
    -0.06
     Export
    -0.06
    _predict
    -0.06
    .REACT
    -0.06
    POSITIVE LOGITS
    串
    0.08
    ↵
    0.07
    OTO
    0.07
    consistent
    0.07
    ali
    0.07
    String
    0.07
    总的来说
    0.07
    肤色
    0.06
    ose
    0.06
    line
    0.06
    Activations Density 0.045%

    No Known Activations