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. 40650
    Prev
    Next
    INDEX
    Explanations

    narrative

    np_max-act · gemini-2.0-flash

    The neuron detects personal-biography snippets—phrases detailing someone’s own life events (e.g. “his early years,” “his moniker,” “his friendships,” “death,” “overdose”).

    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
     Crossing
    -0.07
    Prepare
    -0.06
    ramid
    -0.06
     ----------------------------------------------------------------
    -0.06
     path
    -0.06
    eggies
    -0.06
     Hornets
    -0.06
    -CP
    -0.06
     Bölüm
    -0.06
     pyramid
    -0.06
    POSITIVE LOGITS
    |#
    0.07
    continent
    0.07
     tö
    0.06
    ibase
    0.06
    stad
    0.06
    няв
    0.06
    �
    0.06
    hesive
    0.06
    ısıyla
    0.06
    omin
    0.06
    Activations Density 0.081%

    No Known Activations