© 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. 114878
    Prev
    Next
    INDEX
    Explanations

    Questions

    np_max-act · gemini-2.0-flash

    yes/no questions that begin with an auxiliary or modal verb, often directed at the assistant (e.g., “are you,” “do you”).

    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
    dez
    -0.07
     DN
    -0.07
    rne
    -0.07
     Teams
    -0.07
    禾
    -0.06
    ഫ
    -0.06
     rer
    -0.06
    خت
    -0.06
     suspense
    -0.06
    牢固树立
    -0.06
    POSITIVE LOGITS
    バイ
    0.06
     jedną
    0.06
    ünü
    0.06
    เสา
    0.06
    Issues
    0.06
    _NULL
    0.06
    ocado
    0.06
     hanno
    0.06
    agas
    0.06
    Euro
    0.06
    Activations Density 0.067%

    No Known Activations