© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Qwen2.5-7B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 127911
    Prev
    Next
    INDEX
    Explanations

    programming code

    np_max-act · gemini-2.0-flash

    Tokens used for direct, informal address or conversational openings—greetings, vocatives, and turn-taking phrases (e.g., "who", "let’s", "good morning", "wassup").

    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
     hư
    -0.08
    ようになりました
    -0.07
    .Sm
    -0.06
    降至
    -0.06
    感激
    -0.06
     Bright
    -0.06
     extravag
    -0.06
    长征
    -0.06
    其实就是
    -0.06
    想到了
    -0.06
    POSITIVE LOGITS
    	events
    0.07
    厂房
    0.07
     Zus
    0.07
     Gefühl
    0.06
    Endpoint
    0.06
     RCC
    0.06
    Pref
    0.06
    短短
    0.06
     ENTITY
    0.06
     self
    0.06
    Activations Density 0.424%

    No Known Activations