© 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. 89047
    Prev
    Next
    INDEX
    Explanations

    Announcements and updates

    np_max-act · gemini-2.0-flash

    The neuron responds to language that creates a sense of urgency or a deadline—phrases and tokens signaling “limited time,” “only X days,” “never too late,” calls to act now.

    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
     hạ
    -0.07
     incontro
    -0.07
    еров
    -0.07
    κρα
    -0.07
     Islands
    -0.07
    _courses
    -0.06
     Raf
    -0.06
    ers
    -0.06
    /owl
    -0.06
     представ
    -0.06
    POSITIVE LOGITS
     still
    0.07
     inbox
    0.07
    ].
    0.06
     bitrate
    0.06
     숨
    0.06
     pillow
    0.06
    .Restr
    0.06
    ился
    0.06
     WAIT
    0.06
     аллерг
    0.06
    Activations Density 0.041%

    No Known Activations