© 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. 15-RESID-POST-AA
    11. 61741
    Prev
    Next
    INDEX
    Explanations

    source code headers and comments containing software licensing and copyright boilerplate (e.g., MIT/Apache notices, copyright lines, and LICENSE links).

    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
    ，
    -0.14
    ：
    -0.09
    、
    -0.09
    !
    -0.07
    �
    -0.07
    2
    -0.07
     site
    -0.06
    лы
    -0.06
    -0.06
    两个
    -0.06
    POSITIVE LOGITS
     hardcoded
    0.07
     DACA
    0.06
     swal
    0.06
     fucked
    0.06
     çarp
    0.06
    Unicode
    0.06
    舀
    0.06
    Ῐ
    0.06
    ᶅ
    0.06
    ɽ
    0.06
    Activations Density 36.901%

    No Known Activations