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. 130673
    Prev
    Next
    INDEX
    Explanations

    medical codes

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of “ICD” (International Classification of Diseases) codes.

    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
    /link
    -0.07
    กระ
    -0.07
    (window
    -0.07
     repos
    -0.07
    گر
    -0.06
    lerle
    -0.06
     Marcel
    -0.06
     abolished
    -0.06
    isOk
    -0.06
     depuis
    -0.06
    POSITIVE LOGITS
     زیبا
    0.06
    fname
    0.06
    فة
    0.06
     Sadece
    0.06
    BarButtonItem
    0.06
     kicks
    0.06
     déf
    0.06
     lou
    0.05
    [{
    0.05
     Signs
    0.05
    Activations Density 0.001%

    No Known Activations