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. 86048
    Prev
    Next
    INDEX
    Explanations

    greetings

    np_max-act · gemini-2.0-flash

    the neuron detects conversational check-in phrases expressing well-being (e.g. “I’m doing well, thanks”).

    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
     Caucas
    -0.06
     Light
    -0.06
    vac
    -0.06
    .surface
    -0.06
     Ra
    -0.06
     кожи
    -0.06
     getUsername
    -0.06
    Alarm
    -0.06
     wav
    -0.06
    Alexander
    -0.06
    POSITIVE LOGITS
     *__
    0.07
     wish
    0.07
    ReadWrite
    0.07
    飞
    0.06
    _fixture
    0.06
    iciente
    0.06
     Duterte
    0.06
    _unpack
    0.06
     enjoyed
    0.06
    itel
    0.06
    Activations Density 0.012%

    No Known Activations