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. 130747
    Prev
    Next
    INDEX
    Explanations

    percent

    np_max-act · gemini-2.0-flash

    associations between health issues and risk factors.

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

    The neuron is keyed to spotting numeric expressions—percentages, decimal figures, and other numbers—in the text.

    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
     Dominican
    -0.07
    ';
    ↵
    -0.06
    sidebar
    -0.06
     существ
    -0.06
    animated
    -0.06
    _md
    -0.06
    ','"+
    -0.06
    Hay
    -0.06
    !";
    ↵
    -0.06
    .Arrays
    -0.06
    POSITIVE LOGITS
    .Pass
    0.07
    なお
    0.07
     tossed
    0.07
     omn
    0.07
     Either
    0.06
    /cat
    0.06
    .questions
    0.06
     pra
    0.06
    	reload
    0.06
     Mir
    0.06
    Activations Density 0.069%

    No Known Activations