© 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
    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. 28658
    Prev
    Next
    INDEX
    Explanations

    code and file paths

    np_max-act · gemini-2.0-flash

    subword character sequences common in Romance-language words (and similar letter clusters in identifiers), especially those centered on “ro/ra/ru”-type patterns.

    oai_token-act-pair · gpt-5Triggered 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
    ꪻ
    -0.07
    _yellow
    -0.07
    .aspx
    -0.07
     ImageView
    -0.07
    Graphic
    -0.07
    [row
    -0.06
     Obesity
    -0.06
    	pm
    -0.06
     outstanding
    -0.06
     Kathleen
    -0.06
    POSITIVE LOGITS
    拼
    0.08
    皈
    0.08
    ⇣
    0.08
     Sco
    0.07
     Disco
    0.07
    (help
    0.07
     disarm
    0.07
    朕
    0.07
     cate
    0.07
    Crypt
    0.07
    Activations Density 0.139%

    No Known Activations