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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 125839
    Prev
    Next
    INDEX
    Explanations

    court documents

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on large blocks of leading whitespace or indentation (the blank-space padding at the top of these court‐opinion texts).

    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
     Killer
    -0.07
    _Widget
    -0.07
    .attributes
    -0.06
    A
    -0.06
    σον
    -0.06
    _SWITCH
    -0.06
    slides
    -0.06
     Cipher
    -0.06
     getCode
    -0.06
    quares
    -0.06
    POSITIVE LOGITS
     dwind
    0.07
     osm
    0.07
    ạng
    0.06
     phía
    0.06
     آسی
    0.06
     ropes
    0.06
     tipped
    0.06
     lø
    0.06
     jed
    0.06
     younger
    0.06
    Activations Density 0.002%

    No Known Activations