© 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. 92105
    Prev
    Next
    INDEX
    Explanations

    Flat Earth

    np_max-act · gemini-2.0-flash

    The neuron activates on references to the “flat Earth” claim, i.e. tokens asserting or discussing that the Earth is flat.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    statements and arguments regarding the shape of the Earth.

    oai_token-act-pair · gpt-4o-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
     fulfilled
    -0.07
    pat
    -0.06
    uminium
    -0.06
     ماشین
    -0.06
     prerequisites
    -0.06
     seperate
    -0.06
    ursors
    -0.06
     至
    -0.06
     뜻
    -0.06
    hin
    -0.06
    POSITIVE LOGITS
    �
    0.07
     pytest
    0.07
     Arial
    0.06
    .netbeans
    0.06
    Inset
    0.06
    .Metro
    0.06
     Schools
    0.06
     easier
    0.06
    *pi
    0.06
     pozem
    0.06
    Activations Density 0.010%

    No Known Activations