© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 54032
    Prev
    Next
    INDEX
    Explanations

    Computer/programming context

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of URL transport protocols (e.g. “http,” “https,” “insecure content”), i.e. references to web schema or protocol.

    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
    already
    -0.06
    ists
    -0.06
    ISTS
    -0.06
    ived
    -0.06
     Herrera
    -0.06
    urgence
    -0.06
    pgsql
    -0.06
     انسانی
    -0.06
    (Page
    -0.06
     магаз
    -0.05
    POSITIVE LOGITS
     ασ
    0.07
     
    0.07
     DEV
    0.07
     '*
    0.06
    илася
    0.06
     amen
    0.06
     lit
    0.06
     Virginia
    0.06
     Голов
    0.06
     mio
    0.06
    Activations Density 0.106%

    No Known Activations