© 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. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 7-RESID-POST-AA
    11. 36424
    Prev
    Next
    INDEX
    Explanations

    political arguments

    np_max-act · gemini-2.0-flash

    language about organized groups and in‑group alignment or control—especially unions or political parties pushing an agenda, demanding conformity, or framing “us vs. them.”

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_7/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.7.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     şirket
    -0.06
     VM
    -0.06
    .blogspot
    -0.06
    …the
    -0.06
     squads
    -0.06
    ANDROID
    -0.06
    Charles
    -0.06
    _Release
    -0.06
     Sioux
    -0.06
     sanat
    -0.06
    POSITIVE LOGITS
    _est
    0.07
    .hide
    0.07
     uten
    0.06
     traj
    0.06
     Judgment
    0.06
    /helper
    0.06
    ?',
    0.06
    _AN
    0.06
     ar
    0.06
    frica
    0.06
    Activations Density 0.311%

    No Known Activations