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. 130674
    Prev
    Next
    INDEX
    Explanations

    header

    np_max-act · gemini-2.0-flash

    The neuron fires on descriptions of aerial shot events—especially headers, volleys, and similar actions in football match reports.

    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
    _rewards
    -0.07
     wik
    -0.07
     Ker
    -0.06
     зовсім
    -0.06
    tlement
    -0.06
    /login
    -0.06
     mobility
    -0.06
     Δημο
    -0.06
     FAILURE
    -0.06
    createUrl
    -0.06
    POSITIVE LOGITS
    _pm
    0.07
     나는
    0.06
    -loving
    0.06
    :eq
    0.06
    .dm
    0.06
    .tableView
    0.06
    询
    0.06
     알려
    0.06
    oli
    0.06
    dcc
    0.06
    Activations Density 0.001%

    No Known Activations