© 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. 68610
    Prev
    Next
    INDEX
    Explanations

    vote

    np_max-act · gemini-2.0-flash

    This neuron activates on first‐person self‐references and opinion statements (e.g. “I…,” “my,” “voted,” “vote for,” “I’ll vote,” “I’m going to…”).

    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
     бак
    -0.06
     mobs
    -0.06
     deductible
    -0.06
    -answer
    -0.06
     facebook
    -0.06
     CORS
    -0.06
     від
    -0.06
    adh
    -0.06
    _tar
    -0.06
    _accessible
    -0.06
    POSITIVE LOGITS
    ΑΜ
    0.07
     федера
    0.07
    λλι
    0.07
     λο
    0.07
     hg
    0.06
    时代
    0.06
     CR
    0.06
    еления
    0.06
    830
    0.06
    �
    0.06
    Activations Density 0.013%

    No Known Activations