© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 43206
    Prev
    Next
    INDEX
    Explanations

    openly

    np_max-act · gemini-2.0-flash

    The neuron fires on words or phrases signaling personal disclosure or admission—tokens like “open(ly),” “unapologetic,” “secret,” “show,” and similar reveals of one’s own stance or identity.

    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
    strike
    -0.07
     IUser
    -0.07
    Overlap
    -0.07
     cityName
    -0.07
     demonstration
    -0.07
    (qu
    -0.07
    ALLOC
    -0.06
    flake
    -0.06
    activation
    -0.06
     skewed
    -0.06
    POSITIVE LOGITS
     نم
    0.07
    .every
    0.06
    op
    0.06
     −
    0.06
     naked
    0.06
    _outputs
    0.06
    .controls
    0.05
    .GPIO
    0.05
    implements
    0.05
     refuses
    0.05
    Activations Density 0.012%

    No Known Activations