© 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. 30600
    Prev
    Next
    INDEX
    Explanations

    roles being performed

    np_max-act · gemini-2.0-flash

    This neuron detects words indicating a person’s career milestones or progression (e.g. verbs like “attends,” “graduated,” “began,” “portrayed,” “continued”).

    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
     jobs
    -0.07
    umont
    -0.07
    installation
    -0.06
    _CONTACT
    -0.06
     yönet
    -0.06
     příč
    -0.06
     Bernardino
    -0.06
    имость
    -0.06
     Presidency
    -0.06
     rough
    -0.06
    POSITIVE LOGITS
    325
    0.07
    £
    0.06
    _FM
    0.06
    'na
    0.06
    .closePath
    0.06
    ódigo
    0.06
    .actor
    0.06
    .canvas
    0.06
     cest
    0.06
     entail
    0.06
    Activations Density 0.059%

    No Known Activations