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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 44123
    Prev
    Next
    INDEX
    Explanations

    func

    np_max-act · gemini-2.0-flash

    This neuron fires on Swift declaration keywords and signatures—e.g. “class”, “func”, “override func” lines marking type or method definitions.

    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
    _question
    -0.07
    	synchronized
    -0.06
    ueling
    -0.06
     ภ
    -0.06
     Hib
    -0.06
     bankers
    -0.06
     synchronized
    -0.06
    Editor
    -0.05
    lings
    -0.05
    Ms
    -0.05
    POSITIVE LOGITS
     susceptible
    0.07
     Swift
    0.07
    _rewards
    0.07
    _sell
    0.07
     أجل
    0.07
    .OrdinalIgnoreCase
    0.07
    .linalg
    0.07
    管理员
    0.06
     tourists
    0.06
     Arrange
    0.06
    Activations Density 0.002%

    No Known Activations