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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 1079
    Prev
    Next
    INDEX
    Explanations

    users

    np_max-act · gemini-2.0-flash

    This neuron activates on words naming the people using or involved in a service (e.g. users, runners, trainers, consumers, students).

    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
     whats
    -0.06
     Beau
    -0.06
     Himself
    -0.06
     Zur
    -0.06
     Editorial
    -0.06
    paused
    -0.06
     Much
    -0.06
     translated
    -0.06
     엘
    -0.06
     unfortunate
    -0.06
    POSITIVE LOGITS
    .IN
    0.07
    .setPreferredSize
    0.07
     numberOfRows
    0.06
    ًا
    0.06
     Popup
    0.06
    ẹp
    0.06
    jsx
    0.06
    Validation
    0.06
     letra
    0.06
    aven
    0.06
    Activations Density 0.081%

    No Known Activations