© 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. 85793
    Prev
    Next
    INDEX
    Explanations

    This neuron detects classic story‐opening phrases, especially “Once upon a time” and similar introductory sequences.

    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
    (command
    -0.07
     plutôt
    -0.07
     ridden
    -0.07
     spins
    -0.06
     farmer
    -0.06
    _dependencies
    -0.06
     dismissed
    -0.06
     重
    -0.06
     Não
    -0.06
    ionic
    -0.06
    POSITIVE LOGITS
     banker
    0.07
    _MM
    0.07
    ::|
    0.06
     Medicaid
    0.06
     hút
    0.06
     часу
    0.06
    =""↵
    0.06
    Emma
    0.06
    \Database
    0.06
     neckline
    0.06
    Activations Density 0.224%

    No Known Activations