Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 130312
    Prev
    Next
    INDEX
    Explanations

    editor

    np_max-act · gemini-2.0-flash

    instructions related to creating and configuring files.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    This neuron detects imperative “open the file in a text editor” instructions (e.g. Spanish “Abre el archivo… en un editor de texto” or Portuguese “Abra o arquivo… no editor de texto”).

    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
    усти
    -0.06
    _HEAP
    -0.06
    蒙
    -0.06
    -ф
    -0.06
    щается
    -0.06
    ерб
    -0.06
     Official
    -0.06
     Tamil
    -0.06
    edBy
    -0.06
     наход
    -0.06
    POSITIVE LOGITS
    (SP
    0.07
     vowed
    0.06
    ________
    0.06
     [<
    0.06
    [opt
    0.06
     klein
    0.06
     dle
    0.06
    	throws
    0.06
    exampleInputEmail
    0.06
    endif
    0.06
    Activations Density 0.013%

    No Known Activations