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

    Neuronpedia

    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. 90286
    Prev
    Next
    INDEX
    Explanations

    diff

    np_max-act · gemini-2.0-flash

    Text about consumer household appliances or plug-in scent/air-dispensing products (e.g., diffusers, air fresheners and similar small appliances).

    oai_token-act-pair · gpt-5-miniTriggered 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
     leadership
    -0.07
    捯
    -0.07
     ThreadPool
    -0.06
     Tìm
    -0.06
     centralized
    -0.06
    古人
    -0.06
    рен
    -0.06
    😖
    -0.06
     kol
    -0.06
     Runs
    -0.06
    POSITIVE LOGITS
    ﲠ
    0.07
    /****************************************************************************
    0.07
    污泥
    0.07
    (comment
    0.07
    ForObject
    0.07
    xDD
    0.07
    止め
    0.07
    .msg
    0.07
    NgModule
    0.07
    -spec
    0.07
    Activations Density 1.126%

    No Known Activations