© 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. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 15-RESID-POST-AA
    11. 42212
    Prev
    Next
    INDEX
    Explanations

    elaborating

    np_max-act · gemini-2.0-flash

    second-person, reader-directed statements (especially with contractions and advisory phrasing).

    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_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
     he
    -0.10
     I
    -0.10
     He
    -0.09
     we
    -0.08
     it
    -0.08
    ชาย
    -0.08
     this
    -0.07
     my
    -0.07
     they
    -0.07
     our
    -0.07
    POSITIVE LOGITS
    jm
    0.07
     UserData
    0.06
    _was
    0.06
    _region
    0.06
     jenom
    0.06
     출장
    0.06
     ES
    0.06
    _stub
    0.06
    Campo
    0.06
     plumber
    0.06
    Activations Density 0.646%

    No Known Activations