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. 130471
    Prev
    Next
    INDEX
    Explanations

    License

    np_max-act · gemini-2.0-flash

    Explanation of neuron 4 behavior: the main thing this neuron does is find references to the software license (e.g. the word “License” in the header comments).

    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
     moy
    -0.07
     tiện
    -0.07
    .refs
    -0.06
    BTN
    -0.06
    numerusform
    -0.06
    复
    -0.06
    _tok
    -0.06
    -ass
    -0.06
    ."),↵
    -0.06
     preorder
    -0.06
    POSITIVE LOGITS
    junction
    0.08
    -carousel
    0.07
     Facebook
    0.07
    haust
    0.07
     encompass
    0.07
    .Success
    0.06
     الطبي
    0.06
     homeowner
    0.06
    егодня
    0.06
     Fram
    0.06
    Activations Density 0.001%

    No Known Activations