© 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. 92299
    Prev
    Next
    INDEX
    Explanations

    geographic locations

    np_max-act · gemini-2.0-flash

    The neuron activates on chunks of text that are names of geographic or administrative regions (e.g. province, region, prefecture names).

    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
     Katz
    -0.08
    SCR
    -0.07
     الض
    -0.06
    .Site
    -0.06
    Com
    -0.06
    맞
    -0.06
     Cause
    -0.06
     Addr
    -0.06
         
    -0.06
    Soup
    -0.06
    POSITIVE LOGITS
    �
    0.07
     elekt
    0.06
    existing
    0.06
     выдел
    0.06
     інтер
    0.06
    	Return
    0.06
    malı
    0.06
     AIS
    0.06
    Nov
    0.06
    urls
    0.06
    Activations Density 0.097%

    No Known Activations