© 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. 46029
    Prev
    Next
    INDEX
    Explanations

    U

    np_max-act · gemini-2.0-flash

    the presence of the token "B" (a capital letter B) — often a message/turn label or prefix at the start of user lines.

    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
    Josh
    -0.08
    _udp
    -0.07
    תז
    -0.07
    翡
    -0.07
    咋
    -0.07
    _Code
    -0.07
    Cod
    -0.06
     discussions
    -0.06
     Komm
    -0.06
    𝒾
    -0.06
    POSITIVE LOGITS
     targets
    0.08
    Treatment
    0.07
    uilt
    0.07
     ↵    ↵
    0.07
     choice
    0.07
     Waiting
    0.07
     Thr
    0.07
     ).↵↵
    0.07
     :=↵
    0.07
    ensitive
    0.07
    Activations Density 0.018%

    No Known Activations