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. 81400
    Prev
    Next
    INDEX
    Explanations

    is

    np_max-act · gemini-2.0-flash

    the presence of floating-point numeric literals (decimal numbers) in code.

    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
     payroll
    -0.07
    ampire
    -0.06
     leftovers
    -0.06
    _station
    -0.06
    Would
    -0.06
     busc
    -0.06
     MacOS
    -0.06
    ывать
    -0.05
    -fired
    -0.05
    Mayor
    -0.05
    POSITIVE LOGITS
    عنوان
    0.08
    /archive
    0.07
    (paths
    0.07
     Anadolu
    0.07
    げ
    0.07
    査
    0.06
    영상
    0.06
    NSNumber
    0.06
     Конститу
    0.06
    シー
    0.06
    Activations Density 0.015%

    No Known Activations