© 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. 82154
    Prev
    Next
    INDEX
    Explanations

    Conversational text

    np_max-act · gemini-2.0-flash

    assistant self-referential replies about identity, consciousness, or intelligence in chat transcripts, especially first‑person statements denying or asserting sentience.

    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
    向东
    -0.08
     debunk
    -0.07
     Türkçe
    -0.07
    מינ
    -0.07
     abort
    -0.07
    重新
    -0.07
     MIPS
    -0.06
     literature
    -0.06
    럿
    -0.06
    А
    -0.06
    POSITIVE LOGITS
    _comm
    0.08
    区委
    0.07
    使用权
    0.07
     Rankings
    0.07
    ////↵
    0.07
    quia
    0.07
     Naked
    0.07
     Equip
    0.07
    ={'
    0.07
    ящ
    0.07
    Activations Density 0.096%

    No Known Activations