© 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. 60661
    Prev
    Next
    INDEX
    Explanations

    domain-specific technical terms and acronyms, especially compound or hyphenated nouns across scientific, medical, and product contexts.

    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
    $update
    -0.07
    ਪ
    -0.07
     customary
    -0.07
    position
    -0.07
    ψ
    -0.07
    厦门市
    -0.07
    евой
    -0.07
     Drill
    -0.07
    重庆
    -0.06
     firmly
    -0.06
    POSITIVE LOGITS
    صحف
    0.07
    咝
    0.07
    AttributedString
    0.07
    alue
    0.07
     inkl
    0.07
    垛
    0.06
    海外
    0.06
    과정
    0.06
    çı
    0.06
    根源
    0.06
    Activations Density 1.023%

    No Known Activations