© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 109225
    Prev
    Next
    INDEX
    Explanations

    prayer

    np_max-act · gemini-2.0-flash

    This neuron detects vocative invocations—especially the standalone “O” at the start of prayer addresses (e.g. “O Mary,” “O God,” “O St. …”)

    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
    YD
    -0.07
    ellen
    -0.07
    .learn
    -0.06
     Πρό
    -0.06
    ��
    -0.06
     summaries
    -0.06
    _lift
    -0.06
    James
    -0.06
     niche
    -0.06
    (reg
    -0.06
    POSITIVE LOGITS
    (hw
    0.06
    pps
    0.06
     нали
    0.06
    μ
    0.06
    .Script
    0.06
     sklad
    0.06
     cinemas
    0.06
    发出
    0.06
     çünkü
    0.06
    ,proto
    0.05
    Activations Density 0.012%

    No Known Activations