© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 92626
    Prev
    Next
    INDEX
    Explanations

    Narratives using "you"

    np_max-act · gemini-2.0-flash

    The neuron activates on second‐person references (particularly “you” and related forms).

    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
    _mut
    -0.06
     Breath
    -0.06
    .skills
    -0.06
     econ
    -0.06
     من
    -0.06
     grunt
    -0.06
     Lockheed
    -0.06
     Fur
    -0.06
    вар
    -0.06
    不要
    -0.05
    POSITIVE LOGITS
     celle
    0.07
    eltas
    0.06
     Avg
    0.06
    '))
↵
    0.06
     trí
    0.06
    čit
    0.06
    ्ग
    0.06
     heb
    0.06
     dirección
    0.06
    enticator
    0.06
    Activations Density 0.035%

    No Known Activations