© 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. 55590
    Prev
    Next
    INDEX
    Explanations

    Forum posting rules

    np_max-act · gemini-2.0-flash

    This neuron detects direct second-person instructional or command phrases (imperatives addressing “you”) in forum‐style rules, such as “Do not,” “Keep,” “Stay,” etc.

    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
    .cut
    -0.07
    sty
    -0.07
    .getSystemService
    -0.07
    报
    -0.06
    รด
    -0.06
    ünst
    -0.06
    eteor
    -0.06
    uilder
    -0.06
    اجه
    -0.06
    yu
    -0.06
    POSITIVE LOGITS
     lifelong
    0.06
     Alps
    0.06
     spirituality
    0.06
    Sort
    0.06
     Wayne
    0.06
     folly
    0.06
    .ToList
    0.06
     Theme
    0.06
     정
    0.06
    college
    0.06
    Activations Density 0.014%

    No Known Activations