© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 13498
    Prev
    Next
    INDEX
    Explanations

    mathematical notation

    np_max-act · gemini-2.0-flash

    This neuron fires on references to Maxwell’s equations, i.e. on tokens naming or describing those PDE “laws” (e.g. “equation(s)”, “partial”, “differential”, “law”, “Ampère’s law”, etc.).

    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
     burning
    -0.07
     합니다
    -0.07
     Scriptures
    -0.07
     Tribute
    -0.06
    album
    -0.06
     MANAGEMENT
    -0.06
    酒店
    -0.06
     gramm
    -0.06
     sunlight
    -0.06
     Menu
    -0.06
    POSITIVE LOGITS
    isSelected
    0.06
    Scalar
    0.06
    .Dense
    0.06
     Algorithms
    0.06
    .snap
    0.06
    nesc
    0.06
    (graph
    0.06
     streamline
    0.06
    賽
    0.06
    、や
    0.06
    Activations Density 0.003%

    No Known Activations