© 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. 12569
    Prev
    Next
    INDEX
    Explanations

    opera

    np_max-act · gemini-2.0-flash

    This neuron activates on terms related to opera—names of operas, opera companies, and operatic voice types and performances.

    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
    algorithm
    -0.07
    -zone
    -0.07
     �
    -0.07
     Josh
    -0.06
     Light
    -0.06
     Lease
    -0.06
     κύ
    -0.06
    .GetKeyDown
    -0.06
     novo
    -0.06
    Dev
    -0.06
    POSITIVE LOGITS
     opera
    0.10
     Opera
    0.09
     объек
    0.06
     Opinion
    0.06
     içeri
    0.06
     забезпечення
    0.06
    isted
    0.06
     detached
    0.06
     Armenian
    0.06
    ,var
    0.06
    Activations Density 0.007%

    No Known Activations