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. 16158
    Prev
    Next
    INDEX
    Explanations

    herits

    np_max-act · gemini-2.0-flash

    The neuron activates on calls to an “inherits” function (e.g. util.inherits or oop.inherits), i.e. spots prototypal‐inheritance setup.

    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
    Bucket
    -0.07
    ivative
    -0.06
     Roll
    -0.06
     cartridge
    -0.06
     Sweat
    -0.06
     Defender
    -0.06
     Drivers
    -0.06
    occ
    -0.06
    Bl
    -0.06
     senator
    -0.06
    POSITIVE LOGITS
    !).↵↵
    0.07
     {
    ↵
    ↵
    ↵
    0.07
    [X
    0.06
     RHS
    0.06
     κον
    0.06
     gdy
    0.06
     ‎
    0.06
    entlich
    0.06
    inherits
    0.06
    .ActionBar
    0.06
    Activations Density 0.001%

    No Known Activations