© 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. 49666
    Prev
    Next
    INDEX
    Explanations

    he

    np_max-act · gemini-2.0-flash

    This neuron primarily detects the token "he" (third-person singular male pronoun).

    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
     '('
    -0.08
     MSI
    -0.08
    	menu
    -0.07
     learned
    -0.07
    Ϫ
    -0.07
    README
    -0.07
    _trampoline
    -0.07
     lz
    -0.07
    =sys
    -0.07
    warnings
    -0.06
    POSITIVE LOGITS
    Specific
    0.08
    抓捕
    0.07
     melhores
    0.07
     proprietary
    0.07
    ]initWithFrame
    0.07
    Fine
    0.07
     сахар
    0.06
    serve
    0.06
     Frame
    0.06
     Efficient
    0.06
    Activations Density 0.030%

    No Known Activations