© 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. 11-RESID-POST-AA
    11. 85314
    Prev
    Next
    INDEX
    Explanations

    Emoticons and hashtags

    np_max-act · gemini-2.0-flash

    This neuron detects kawaii/cutesy emotive markers — emojis, emoticons, and playful interjections (e.g., "UwU", "Ny aa", "(^ω^)"-style faces).

    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
    שחק
    -0.08
    CppTypeDefinition
    -0.07
    Future
    -0.07
    据悉
    -0.07
    mnop
    -0.07
     ספר
    -0.07
    며
    -0.07
    (expected
    -0.07
     losers
    -0.07
    _desc
    -0.07
    POSITIVE LOGITS
    ém
    0.07
    ách
    0.07
    IDE
    0.07
    具
    0.07
     dryer
    0.07
    柱
    0.07
     register
    0.07
    בן
    0.06
    跪
    0.06
     Maze
    0.06
    Activations Density 0.151%

    No Known Activations