© 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. 11-RESID-POST-AA
    11. 68775
    Prev
    Next
    INDEX
    Explanations

    quotation marks

    np_max-act · gemini-2.0-flash

    Tokens that mark strong opinions or moral/value judgments (especially evaluative words often shown in quotes, like "best", "better", "right"/"wrong", "hate", or emphatic negations).

    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_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Size
    -0.08
     תל
    -0.07
     נש
    -0.07
    อาจาร
    -0.07
    Sz
    -0.07
     "");
↵
    -0.07
     inflater
    -0.07
    peak
    -0.07
    马刺
    -0.06
    Dataset
    -0.06
    POSITIVE LOGITS
     getting
    0.06
    完整
    0.06
    ün
    0.06
    忙碌
    0.06
     пров
    0.06
     gratuits
    0.06
    坎坷
    0.06
    갉
    0.06
    艋
    0.06
    🤩
    0.06
    Activations Density 0.257%

    No Known Activations