© 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
    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. 92464
    Prev
    Next
    INDEX
    Explanations

    Personal experiences

    np_max-act · gemini-2.0-flash

    The neuron primarily detects first‐person, self-referential statements (e.g. “I,” “my,” “we”) indicating personal experience or testimonial voice.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    personal, conversational language—especially first- or second-person statements such as testimonials, direct address to the reader, and calls to action.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    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
     благ
    -0.07
    _color
    -0.06
    ma
    -0.06
     бума
    -0.06
    Overflow
    -0.06
    -categories
    -0.06
     Sco
    -0.06
     defect
    -0.06
    وم
    -0.06
    요
    -0.06
    POSITIVE LOGITS
    _LO
    0.07
    ("---
    0.07
     enterprises
    0.06
    (global
    0.06
     ―
    0.06
     станет
    0.06
    Practice
    0.06
    dataSource
    0.06
    采购
    0.06
    _de
    0.06
    Activations Density 0.667%

    No Known Activations