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. 15-RESID-POST-AA
    6. 74756
    Prev
    Next
    INDEX
    Explanations

    socks

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_15/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.15.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     agenda
    -0.07
    mux
    -0.07
    .Cl
    -0.07
    ,ll
    -0.07
     climate
    -0.06
    ("?
    -0.06
     inventor
    -0.06
     cure
    -0.06
     RAD
    -0.06
     gyro
    -0.06
    POSITIVE LOGITS
     socks
    0.09
     wirk
    0.07
     sock
    0.07
     vigilant
    0.07
    lico
    0.06
    ानक
    0.06
     vyk
    0.06
    lients
    0.06
    Numeric
    0.06
    Goods
    0.06
    Activations Density 0.002%

    No Known Activations