© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 110574
    Prev
    Next
    INDEX
    Explanations

    to be, in

    np_max-act · gemini-2.0-flash

    narrative elements related to beauty and danger.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron fires on first-person or direct experiential descriptions – especially “X was like…,” “I feel like…,” or similar narrative/imagery comparisons.

    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
     инвести
    -0.07
     stocking
    -0.06
    .COLOR
    -0.06
    _pages
    -0.06
     Resolve
    -0.06
    IDES
    -0.06
     admire
    -0.06
     lies
    -0.06
     близ
    -0.06
    .CASCADE
    -0.06
    POSITIVE LOGITS
     Dish
    0.07
     Became
    0.07
     bä
    0.07
     #{@
    0.06
    죽
    0.06
    (ast
    0.06
    fel
    0.06
     nex
    0.06
    (D
    0.06
     comparisons
    0.06
    Activations Density 0.011%

    No Known Activations