© 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. 92615
    Prev
    Next
    INDEX
    Explanations

    names

    np_max-act · gemini-2.0-flash

    The neuron activates on capitalized personal names—particularly the heading name in “given name” or disambiguation entries.

    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
    usty
    -0.07
    brain
    -0.07
    様
    -0.07
    :Add
    -0.07
     tạm
    -0.07
     hilar
    -0.07
     착
    -0.07
    .reloadData
    -0.06
     congr
    -0.06
     کیفیت
    -0.06
    POSITIVE LOGITS
     odpověd
    0.06
    ovaného
    0.06
     leap
    0.06
     exercising
    0.06
    _ACCEPT
    0.06
    Fore
    0.06
     зак
    0.06
     Yun
    0.05
     columnName
    0.05
    Professional
    0.05
    Activations Density 0.010%

    No Known Activations