© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 12566
    Prev
    Next
    INDEX
    Explanations

    Magic 8 Ball/kaleidoscopes

    np_max-act · gemini-2.0-flash

    The neuron specializes in detecting mentions of the “Magic 8 Ball” (i.e. the digit‐“8” plus “ball” toy reference).

    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
    nowledge
    -0.07
    Lua
    -0.07
    awks
    -0.07
    騎
    -0.06
    (blank
    -0.06
    _yaw
    -0.06
    _pub
    -0.06
     Philipp
    -0.06
    ViewModel
    -0.06
    599
    -0.06
    POSITIVE LOGITS
     wisely
    0.07
     бли
    0.07
    uring
    0.07
    AutoresizingMaskIntoConstraints
    0.06
     rigid
    0.06
    	open
    0.06
    	Optional
    0.06
    	container
    0.06
     rodin
    0.06
     obchod
    0.06
    Activations Density 0.011%

    No Known Activations