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. 129184
    Prev
    Next
    INDEX
    Explanations

    management positions

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of sports team executives or coaches by their formal titles (e.g., general manager, GM, coach, president).

    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
    Encoded
    -0.07
     Backup
    -0.06
     wid
    -0.06
     vegetables
    -0.06
    umped
    -0.06
    말
    -0.06
    /basic
    -0.06
    idy
    -0.06
     Reform
    -0.06
    -addons
    -0.06
    POSITIVE LOGITS
     scout
    0.06
     nigeria
    0.06
     sebagai
    0.06
     Nhà
    0.06
    SHOT
    0.06
     hazardous
    0.06
     HF
    0.06
     treffen
    0.06
    Fore
    0.06
    pons
    0.06
    Activations Density 0.015%

    No Known Activations