© 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. 15-RESID-POST-AA
    11. 125324
    Prev
    Next
    INDEX
    Explanations

    iny

    np_max-act · gemini-2.0-flash

    technical, high-salience tokens such as acronyms, parameter-style identifiers, percentages, and intense abstract evaluative terms, often appearing in lists or enumerations.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    鹊
    -0.07
    Dep
    -0.07
    paRepository
    -0.07
     başladı
    -0.07
    냐
    -0.07
     hadde
    -0.07
    Ant
    -0.06
     hüküm
    -0.06
     Rounded
    -0.06
     shall
    -0.06
    POSITIVE LOGITS
    ओ
    0.07
    Tambah
    0.07
     citrus
    0.07
    只是
    0.07
    柳州
    0.06
     Rice
    0.06
    为基础
    0.06
     coconut
    0.06
     عالية
    0.06
    itemid
    0.06
    Activations Density 17.878%

    No Known Activations