© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 85169
    Prev
    Next
    INDEX
    Explanations

    web code

    np_max-act · gemini-2.0-flash

    numeric digit sequences and number-heavy patterns such as phone numbers, prices, years, and list indices, especially hyphenated formats like 1-800-style strings.

    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
    FetchRequest
    -0.08
     chat
    -0.07
     danmark
    -0.07
    Zero
    -0.07
    .register
    -0.07
    榍
    -0.07
    剐
    -0.07
    鹙
    -0.07
     Joined
    -0.07
    낱
    -0.07
    POSITIVE LOGITS
    гляд
    0.07
    奶油
    0.07
     PSD
    0.07
     pem
    0.07
    لغ
    0.07
    שי
    0.07
    .sym
    0.06
    ệm
    0.06
    委
    0.06
     смысл
    0.06
    Activations Density 0.177%

    No Known Activations