© 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. 54411
    Prev
    Next
    INDEX
    Explanations

    elect

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on “electro”-derived terms (words beginning with the prefix electro- or elect-) denoting electrical or electrophotographic processes.

    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
     Riy
    -0.08
     bean
    -0.08
    Nib
    -0.07
     Sin
    -0.07
     nib
    -0.07
     Husband
    -0.07
     Noah
    -0.07
     scan
    -0.07
     Rim
    -0.07
     Yu
    -0.07
    POSITIVE LOGITS
    Elect
    0.10
     elect
    0.10
     Elect
    0.09
    .alloc
    0.08
    :
    0.08
    .E
    0.08
     electorate
    0.07
     (
    0.07
     이동합니다
    0.07
    degrees
    0.07
    Activations Density 0.016%

    No Known Activations