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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 86021
    Prev
    Next
    INDEX
    Explanations

    URLs

    np_max-act · gemini-2.0-flash

    The neuron is tuned to spot web URLs (e.g. tokens like “http” and related URL fragments).

    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
    งส
    -0.07
    media
    -0.07
    VERIFY
    -0.07
    олее
    -0.07
    arkan
    -0.06
    ámara
    -0.06
     Heritage
    -0.06
    ssql
    -0.06
     jams
    -0.06
    .jpeg
    -0.06
    POSITIVE LOGITS
     Респ
    0.06
     гол
    0.06
     нор
    0.06
     العمل
    0.06
    -cor
    0.06
    .bo
    0.06
     Αν
    0.06
     elbow
    0.06
     CHE
    0.05
     соль
    0.05
    Activations Density 0.046%

    No Known Activations