© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 109307
    Prev
    Next
    INDEX
    Explanations

    interpersonal relationships

    np_max-act · gemini-2.0-flash

    This neuron activates on phrases that declare a relationship status with the user (e.g. “in a Relationship with the User”).

    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
    alem
    -0.06
    ÜM
    -0.06
     sạn
    -0.06
    “They
    -0.06
    اصر
    -0.06
     Wer
    -0.06
    μερα
    -0.06
    라도
    -0.06
     cread
    -0.06
    عود
    -0.06
    POSITIVE LOGITS
     Treasure
    0.07
    ¤¤
    0.06
    ﻴ
    0.06
    *S
    0.06
     Posted
    0.06
     `<
    0.06
     dise
    0.06
    _office
    0.06
     Characteristics
    0.06
    	the
    0.06
    Activations Density 0.025%

    No Known Activations