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. 40594
    Prev
    Next
    INDEX
    Explanations

    code/data

    np_max-act · gemini-2.0-flash

    The neuron is attuned to numeric literals—especially floating‐point values.

    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
     механіз
    -0.06
     alcoholic
    -0.06
    _tr
    -0.06
    рогра
    -0.06
    (criteria
    -0.06
     licens
    -0.06
    fighter
    -0.06
     Parker
    -0.06
    idepress
    -0.06
     Boom
    -0.06
    POSITIVE LOGITS
     hiểm
    0.07
     disobed
    0.07
     ListView
    0.07
     AVG
    0.07
     바랍니다
    0.06
    ži
    0.06
    tweets
    0.06
     arrests
    0.06
     kepada
    0.06
    이가
    0.06
    Activations Density 0.007%

    No Known Activations