© 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
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 126335
    Prev
    Next
    INDEX
    Explanations

    another

    np_max-act · gemini-2.0-flash

    This neuron flags occurrences of the custom assistant or character name placeholder (e.g. “NAME_1”) in the role-play metadata.

    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
     aff
    -0.07
    _avg
    -0.06
    -child
    -0.06
     dalla
    -0.06
    าม
    -0.06
    Trap
    -0.06
    Ship
    -0.06
    .Host
    -0.06
     Bakery
    -0.06
     第二
    -0.06
    POSITIVE LOGITS
    icensed
    0.07
     Equ
    0.06
     такого
    0.06
    hiba
    0.06
     equ
    0.06
    Lisa
    0.06
    ーク
    0.06
    ihad
    0.06
     evangelical
    0.06
    (graph
    0.06
    Activations Density 0.001%

    No Known Activations