© 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. 76198
    Prev
    Next
    INDEX
    Explanations

    the role label marking the assistant’s turn in chat-style markup (i.e., the start of assistant responses)

    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
     cà
    -0.08
     Petit
    -0.07
    .r
    -0.07
    北约
    -0.07
    Ŧ
    -0.07
    😤
    -0.07
    🐺
    -0.07
     Gina
    -0.07
     dt
    -0.07
    봅
    -0.06
    POSITIVE LOGITS
    技艺
    0.07
    sequence
    0.07
    Equ
    0.07
     błęd
    0.06
     JMenuItem
    0.06
     EVERY
    0.06
    -account
    0.06
    <Audio
    0.06
    _movement
    0.06
     источ
    0.06
    Activations Density 0.013%

    No Known Activations