Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Qwen2.5-7B-IT
    4. Resid Post - 131k
    5. 19-RESID-POST-AA
    6. 72412
    Prev
    Next
    INDEX
    Explanations

    code using threading

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-qwen2.5-7b-instruct/resid_post_layer_19/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.19.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    printw
    -0.07
    PLUS
    -0.06
     PPC
    -0.06
     QColor
    -0.06
     shaky
    -0.06
    当然
    -0.06
    benh
    -0.06
    看了看
    -0.06
    פופולרי
    -0.06
     crippled
    -0.06
    POSITIVE LOGITS
    等因素
    0.07
    ],$
    0.07
     $
    0.07
     Identity
    0.07
    .mozilla
    0.07
    )"
    0.07
    也将
    0.06
    ears
    0.06
    发送
    0.06
     ↵    ↵
    0.06
    Activations Density 0.001%

    No Known Activations