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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 60851
    Prev
    Next
    INDEX
    Explanations

    This neuron detects requests or statements about improving oneself or one's skills — expressions of intent/goals to become better (self-improvement).

    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
     นาย
    -0.07
    retain
    -0.07
     payouts
    -0.07
    Af
    -0.07
    feit
    -0.06
     politics
    -0.06
     worksheets
    -0.06
     IsValid
    -0.06
     using
    -0.06
     seria
    -0.06
    POSITIVE LOGITS
    [][
    0.08
    (SQLException
    0.08
    -red
    0.07
    Backing
    0.07
    (dir
    0.07
    🌦
    0.07
     #[
    0.07
    أخبار
    0.07
    tracked
    0.07
    htmlspecialchars
    0.07
    Activations Density 0.143%

    No Known Activations