© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 9434
    Prev
    Next
    INDEX
    Explanations

    Advice and consultation

    np_max-act · gemini-2.0-flash

    references to the user being an AI or discussing AI capabilities.

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

    The neuron flags the assistant’s self-descriptive statements—especially the phrase “I am an AI assistant” (and closely associated “created by…” wording).

    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
    SEARCH
    -0.07
    Specification
    -0.06
    appers
    -0.06
    Document
    -0.06
    .navigator
    -0.06
     روست
    -0.06
     jobs
    -0.06
    Grab
    -0.06
    Font
    -0.06
     Job
    -0.06
    POSITIVE LOGITS
    (&___
    0.07
     hesitant
    0.07
     attn
    0.06
    �
    0.06
     heir
    0.06
     blinking
    0.06
     wirk
    0.06
    .recv
    0.06
     grosse
    0.06
    "$
    0.06
    Activations Density 0.010%

    No Known Activations