Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 54821
    Prev
    Next
    INDEX
    Explanations

    The neuron is sensitive to descriptive references to a woman’s sexualized body features—most strongly words about breasts (and to a lesser extent butt and body measurements).

    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
    .Task
    -0.07
     Sensors
    -0.06
     LRV
    -0.06
     hardware
    -0.06
     surveillance
    -0.06
    ordinal
    -0.06
    /preferences
    -0.06
     cores
    -0.06
    _room
    -0.06
    967
    -0.06
    POSITIVE LOGITS
    ”—
    0.07
    .asList
    0.07
    --
    0.06
    :-
    0.06
    .setLocation
    0.06
    tingham
    0.06
     scarc
    0.06
     Charl
    0.06
     anything
    0.06
     lush
    0.06
    Activations Density 0.021%

    No Known Activations