© 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 EvalsExportsAPI 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. 3565
    Prev
    Next
    INDEX
    Explanations

    Random web content

    np_max-act · gemini-2.0-flash

    programming and developer-related artifacts such as code snippets, library/package references, API/repository links, and technical Markdown/URL structures.

    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
     Spice
    -0.07
    ihan
    -0.07
    ++]
    -0.07
    默认
    -0.07
     BOT
    -0.07
    _Ch
    -0.06
    ritic
    -0.06
     CharSet
    -0.06
    udad
    -0.06
    司机
    -0.06
    POSITIVE LOGITS
    ﯠ
    0.07
    묄
    0.07
     Flying
    0.07
    _blocking
    0.07
    菘
    0.07
     plays
    0.06
     movies
    0.06
    行业中
    0.06
     tệ
    0.06
     directory
    0.06
    Activations Density 14.261%

    No Known Activations