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. 130443
    Prev
    Next
    INDEX
    Explanations

    numbers

    np_max-act · gemini-2.0-flash

    The neuron is primarily activated by numeric tokens (standalone numbers and digit sequences).

    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
     seventh
    -0.07
     Flu
    -0.06
     annual
    -0.06
     Buffer
    -0.06
     produces
    -0.06
     переш
    -0.06
     dbus
    -0.06
    .inputs
    -0.06
    IDDEN
    -0.06
     kon
    -0.06
    POSITIVE LOGITS
    meet
    0.07
     kiện
    0.07
    ampaign
    0.07
     président
    0.06
    minated
    0.06
     absent
    0.06
    ∨
    0.06
    zk
    0.06
     inst
    0.06
    кою
    0.06
    Activations Density 0.019%

    No Known Activations