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. 16175
    Prev
    Next
    INDEX
    Explanations

    This neuron fires on PHP object‐oriented code constructs, especially the “->” property/method access patterns.

    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
     vinden
    -0.06
     Andy
    -0.06
    combination
    -0.06
     چهار
    -0.06
     Ast
    -0.06
    GetName
    -0.06
     currentUser
    -0.06
     onlar
    -0.06
    MASTER
    -0.06
    XmlAttribute
    -0.06
    POSITIVE LOGITS
     vascular
    0.07
    .getLast
    0.06
    ρας
    0.06
    (Dense
    0.06
    “These
    0.06
    ']");↵
    0.06
    >Loading
    0.06
    λεί
    0.06
     exits
    0.06
    -sub
    0.06
    Activations Density 0.082%

    No Known Activations