© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 31258
    Prev
    Next
    INDEX
    Explanations

    online opinions and arguments

    np_max-act · gemini-2.0-flash

    sarcastic, contrast-heavy critiques that highlight hypocrisy or contradictions using negation, rhetorical questions, and aside-like commentary.

    oai_token-act-pair · gpt-5
    Triggered 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
    Negative Logits
     Investment
    -0.07
    ꂫ
    -0.07
     Imagine
    -0.07
     сообщил
    -0.07
     económico
    -0.07
    Subscribe
    -0.07
    .usage
    -0.07
    (mm
    -0.06
    👠
    -0.06
    COMPARE
    -0.06
    POSITIVE LOGITS
     errno
    0.08
     Natürlich
    0.08
     binaries
    0.08
    -tr
    0.07
    <len
    0.07
    ;if
    0.07
     deve
    0.07
     Config
    0.07
     pun
    0.07
    ര
    0.07
    Activations Density 0.107%

    No Known Activations