© 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. 3848
    Prev
    Next
    INDEX
    Explanations

    female

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of people—personal names and other tokens referring to individual persons (person entities).

    oai_token-act-pair · gpt-5-miniTriggered 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
    icipant
    -0.07
    宁夏
    -0.07
    メーカ
    -0.07
    臑
    -0.07
    旅客
    -0.07
     Everton
    -0.07
    菜肴
    -0.06
    ellido
    -0.06
    爷爷
    -0.06
    ᴶ
    -0.06
    POSITIVE LOGITS
    vection
    0.08
     confronted
    0.07
     diz
    0.07
     declares
    0.07
     Ms
    0.07
     mime
    0.07
     patched
    0.07
     oppressed
    0.07
    сим
    0.07
    Recursive
    0.07
    Activations Density 2.150%

    No Known Activations