Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 130242
    Prev
    Next
    INDEX
    Explanations

    loyal

    np_max-act · gemini-2.0-flash

    The neuron activates on words expressing loyalty, protection, and caring attachments.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    _UNIQUE
    -0.06
    esinden
    -0.06
     بي
    -0.06
    ADDRESS
    -0.06
    extract
    -0.06
    .construct
    -0.06
    енного
    -0.06
    .list
    -0.06
     سیم
    -0.06
     machen
    -0.06
    POSITIVE LOGITS
     궁금
    0.08
    鹿
    0.07
     foundational
    0.07
    AW
    0.07
     Sweep
    0.06
    ếp
    0.06
     Mercer
    0.06
     Rudy
    0.06
     polov
    0.06
    (phase
    0.06
    Activations Density 0.013%

    No Known Activations