© 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. 12762
    Prev
    Next
    INDEX
    Explanations

    my

    np_max-act · gemini-2.0-flash

    The neuron detects words and phrases expressing the author’s personal intent or desire (e.g. “I,” “want,” “need,” “trying”).

    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
     Counts
    -0.06
    vironments
    -0.06
    ρων
    -0.06
     entra
    -0.06
    Err
    -0.06
    ELY
    -0.06
    кого
    -0.06
    .wall
    -0.06
     以
    -0.06
    (entities
    -0.05
    POSITIVE LOGITS
    trad
    0.07
     absurd
    0.07
    ेश
    0.06
     تاریخی
    0.06
     Red
    0.06
     대표
    0.06
     يو
    0.06
     Independ
    0.06
    ریف
    0.06
    _pic
    0.06
    Activations Density 0.101%

    No Known Activations