Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 346
    Prev
    Next
    INDEX
    Explanations

    Period

    np_max-act · gemini-2.0-flash

    The neuron activates on fragments of legal citation, especially abbreviated references to federal rules (e.g. “Fed. R. App. P. 34”).

    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
     emanc
    -0.08
     Flint
    -0.07
     caut
    -0.07
     ตร
    -0.06
     Ryan
    -0.06
    ้ต
    -0.06
    เจร
    -0.06
     غیر
    -0.06
    .menuStrip
    -0.06
     jeux
    -0.06
    POSITIVE LOGITS
    edn
    0.07
     หน
    0.06
     teleport
    0.06
    -selling
    0.06
    |`↵
    0.06
    ratio
    0.06
     tamam
    0.06
     soda
    0.06
     @(
    0.06
    wrapper
    0.06
    Activations Density 0.001%

    No Known Activations