© 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. 73886
    Prev
    Next
    INDEX
    Explanations

    politics

    np_max-act · gemini-2.0-flash

    The neuron activates on words that signal an extreme or “hard‐line” ideological or militant stance.

    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
    ánu
    -0.07
    _YUV
    -0.06
    YPD
    -0.06
     Wisdom
    -0.06
    .NewReader
    -0.06
    Warn
    -0.06
    _TYPE
    -0.06
     Fund
    -0.06
    _while
    -0.06
     Ammo
    -0.06
    POSITIVE LOGITS
     environmentally
    0.07
     conqu
    0.07
    :url
    0.07
     xlink
    0.06
    各
    0.06
     traditionally
    0.06
    ysqli
    0.06
    >'
    0.06
    .Secret
    0.06
    침
    0.06
    Activations Density 0.013%

    No Known Activations