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. 40496
    Prev
    Next
    INDEX
    Explanations

    Interactions between multiple actors

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of allocating coins (rewards) to “you” and to “others,” i.e. resource‐distribution statements specifying how many coins go to self versus another participant.

    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
     executions
    -0.06
    bruar
    -0.06
    τισ
    -0.06
    .backends
    -0.06
     intend
    -0.06
    него
    -0.06
    ****************************************************************
    -0.06
    Returning
    -0.06
     isso
    -0.06
     Naz
    -0.06
    POSITIVE LOGITS
     hlavy
    0.07
    俗
    0.07
    _OR
    0.06
    (Model
    0.06
    葡
    0.06
     Получ
    0.06
    舒
    0.06
     ={↵
    0.06
     أي
    0.06
    ısır
    0.06
    Activations Density 0.003%

    No Known Activations