© 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. 42671
    Prev
    Next
    INDEX
    Explanations

    online sales/shopping

    np_max-act · gemini-2.0-flash

    The neuron fires strongly on seller/buyer service instructions and calls to action—phrases like “message us,” “contact seller,” “please provide…,” or other user‐directed policy and support prompts.

    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
    pagination
    -0.07
     paginate
    -0.06
     suits
    -0.06
    _EDIT
    -0.06
    Components
    -0.06
    wan
    -0.06
     lbl
    -0.06
     end
    -0.06
    _RETURNTRANSFER
    -0.06
     pas
    -0.06
    POSITIVE LOGITS
    .Pl
    0.07
     �
    0.07
    γο
    0.06
    engeance
    0.06
    _Instance
    0.06
    intage
    0.06
    IFICATE
    0.06
     ◎
    0.06
     pouze
    0.06
     Ore
    0.06
    Activations Density 0.028%

    No Known Activations