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. 86120
    Prev
    Next
    INDEX
    Explanations

    teleport

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens related to characters being moved or transformed—especially the “tele-/port(ed)” and “transport(ed)” sequences describing teleportation or sudden relocation.

    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
     Dialogue
    -0.06
    (Grid
    -0.06
    _wrapper
    -0.06
    ην
    -0.06
    ookeeper
    -0.06
     Constructor
    -0.06
     일부
    -0.06
    };↵
    -0.06
    FontSize
    -0.06
     FIXME
    -0.05
    POSITIVE LOGITS
     dieses
    0.07
     sx
    0.07
     denim
    0.06
    .defineProperty
    0.06
    (gulp
    0.06
     solutions
    0.06
    .default
    0.06
    _payment
    0.06
    .birth
    0.06
     lottery
    0.06
    Activations Density 0.014%

    No Known Activations