Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 53049
    Prev
    Next
    INDEX
    Explanations

    technology and artificial intelligence

    np_max-act · gemini-2.0-flash

    statements regarding personal beliefs and desires.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    This neuron responds to the assistant referring to itself as an AI or machine-learning language model, activating on self-descriptive statements about its identity or capabilities.

    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
     compromises
    -0.07
    Heap
    -0.06
     کاری
    -0.06
    etty
    -0.06
     рис
    -0.06
     Observation
    -0.06
    ідно
    -0.06
    esini
    -0.06
    leyen
    -0.06
     eius
    -0.06
    POSITIVE LOGITS
    hum
    0.07
                        ↵
    0.07
    /channel
    0.06
     Speaking
    0.06
     buena
    0.06
    						↵
    0.06
    					↵
    0.06
    _DIG
    0.06
    %");↵
    0.06
    essional
    0.06
    Activations Density 0.022%

    No Known Activations