© 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. 30002
    Prev
    Next
    INDEX
    Explanations

    Software specifications/documentation

    np_max-act · gemini-2.0-flash

    self-referential boilerplate where the assistant identifies itself as an AI language model and describes its capabilities or limitations.

    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
     broccoli
    -0.08
     resembl
    -0.07
    James
    -0.07
    Im
    -0.07
    禀
    -0.07
    落实
    -0.07
     nowrap
    -0.07
    ilee
    -0.06
    的形式
    -0.06
    _USERS
    -0.06
    POSITIVE LOGITS
     dna
    0.07
     Paladin
    0.07
    dna
    0.07
     Rohing
    0.07
    低迷
    0.07
     Damn
    0.07
    фор
    0.07
    	direction
    0.07
     маст
    0.07
    滑雪
    0.07
    Activations Density 0.017%

    No Known Activations