© 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. 62921
    Prev
    Next
    INDEX
    Explanations

    the end-of-message marker for assistant replies, especially after short greeting or help-offer responses (often multilingual)

    oai_token-act-pair · gpt-5Triggered by @vetterc0

    polite assistant greeting/opening messages that invite the user to ask for help

    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
     Norfolk
    -0.07
     kuk
    -0.06
     [@
    -0.06
     decor
    -0.06
    	Render
    -0.06
     FN
    -0.06
     Routing
    -0.06
    解脱
    -0.06
     disconnected
    -0.06
     STA
    -0.06
    POSITIVE LOGITS
     mạnh
    0.08
    dlg
    0.07
    嵌
    0.07
    🐉
    0.07
     Ages
    0.07
    igr
    0.07
    个月
    0.07
    YPRE
    0.07
    blood
    0.07
     xấu
    0.07
    Activations Density 0.083%

    No Known Activations