© 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. 4196
    Prev
    Next
    INDEX
    Explanations

    assistance

    np_max-act · gemini-2.0-flash

    statements in system prompts that define the model’s role/identity as a helpful assistant

    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
    Figure
    -0.08
    举例
    -0.08
    endDate
    -0.07
    klass
    -0.07
    _SIGNAL
    -0.07
     э
    -0.07
    .goods
    -0.07
    手术
    -0.07
    	html
    -0.07
    =%.
    -0.07
    POSITIVE LOGITS
    挫
    0.08
    -Fi
    0.08
    スキ
    0.07
     preserving
    0.07
    ;left
    0.07
     mini
    0.07
     Griffith
    0.07
     �
    0.07
    urf
    0.07
     club
    0.07
    Activations Density 0.047%

    No Known Activations