© 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. 130103
    Prev
    Next
    INDEX
    Explanations

    eyes

    np_max-act · gemini-2.0-flash

    The neuron activates on words referring to eyes or smiling/grinning (e.g. “eyes,” “smiling,” “grin,” “smiley”).

    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
    .RemoveAt
    -0.06
    HV
    -0.06
     picnic
    -0.06
    AW
    -0.06
     timeStamp
    -0.06
    Ice
    -0.06
    out
    -0.06
     spectators
    -0.06
     낮
    -0.06
    ")]↵↵
    -0.06
    POSITIVE LOGITS
     STDMETHODCALLTYPE
    0.07
    :image
    0.07
     تصميم
    0.07
     Vulkan
    0.07
    ouver
    0.07
    (proto
    0.07
     Brain
    0.06
     equival
    0.06
     neurological
    0.06
     Operational
    0.06
    Activations Density 0.023%

    No Known Activations