© 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. 68706
    Prev
    Next
    INDEX
    Explanations

    emotions and feelings

    np_max-act · gemini-2.0-flash

    The neuron activates on words describing characters’ extraordinary powers or supernatural actions (e.g. feeding on memories, manipulating time, possessing bodies, spreading spores, casting illusions).

    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
    :not
    -0.06
    (round
    -0.06
     Jeff
    -0.06
     perse
    -0.06
     Gardner
    -0.06
     Pleasant
    -0.06
    Pal
    -0.06
     schnell
    -0.06
     flo
    -0.06
    rollo
    -0.06
    POSITIVE LOGITS
    resizing
    0.07
    _kwargs
    0.06
    +_
    0.06
     ніколи
    0.06
     vem
    0.06
    abus
    0.06
    рон
    0.06
    -inspired
    0.06
     CSRF
    0.06
     ***!↵
    0.06
    Activations Density 0.057%

    No Known Activations