© 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. 84821
    Prev
    Next
    INDEX
    Explanations

    Business and research

    np_max-act · gemini-2.0-flash

    abstract, formal technical terms referring to configurations, protections, resources, policies, regulations, and other system or compliance concepts

    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
    Middle
    -0.07
     Device
    -0.07
    _UNSUPPORTED
    -0.07
    quez
    -0.07
     meat
    -0.07
     AFTER
    -0.07
     vending
    -0.06
     Loader
    -0.06
     Yemen
    -0.06
     cripp
    -0.06
    POSITIVE LOGITS
    ניות
    0.07
    Altern
    0.07
    ourage
    0.07
    完美的
    0.07
    рев
    0.06
     людей
    0.06
     passions
    0.06
    asyarakat
    0.06
    olated
    0.06
     самые
    0.06
    Activations Density 5.540%

    No Known Activations