© 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. 43405
    Prev
    Next
    INDEX
    Explanations

    The main thing this neuron does is detect explicit sexual content terms (e.g. porn-related words and graphic sex act vocabulary).

    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
     gaussian
    -0.06
    _th
    -0.06
    	gr
    -0.06
     Scheduler
    -0.06
    Wars
    -0.06
     Prelude
    -0.06
     ters
    -0.06
    Ng
    -0.06
    Các
    -0.06
    Homepage
    -0.06
    POSITIVE LOGITS
     DK
    0.07
    ME
    0.07
     yapmaya
    0.06
     baise
    0.06
     itk
    0.06
    esp
    0.06
    そこ
    0.06
    ائل
    0.06
     overcoming
    0.06
    ULT
    0.06
    Activations Density 0.011%

    No Known Activations