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. 21229
    Prev
    Next
    INDEX
    Explanations

    speculative fiction excerpts

    np_max-act · gemini-2.0-flash

    sentences or phrases discussing a society-normalized weight/“weight pill” concept — i.e., talk about weight, the magic pill, and cultural acceptance of being heavy.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    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
    醫
    -0.07
    noloj
    -0.07
     एन
    -0.07
     méd
    -0.07
     mujeres
    -0.06
     Zoe
    -0.06
     joystick
    -0.06
     людини
    -0.06
     техні
    -0.06
    573
    -0.06
    POSITIVE LOGITS
    Gesture
    0.07
    aft
    0.06
    igrants
    0.06
    (import
    0.06
    igration
    0.06
    LayoutParams
    0.06
     Rim
    0.06
    bomb
    0.06
    idence
    0.06
     ग
    0.06
    Activations Density 0.179%

    No Known Activations