© 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. 6457
    Prev
    Next
    INDEX
    Explanations

    Skepticism/Richard Dawkins

    np_max-act · gemini-2.0-flash

    This neuron detects references to organized skepticism and skeptic figures (e.g. “James Randi,” “JREF,” “skeptic(s)”).

    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
     Alman
    -0.07
     बत
    -0.07
     자료
    -0.06
    all
    -0.06
     От
    -0.06
     essere
    -0.06
     cạnh
    -0.06
     aver
    -0.06
    (fmt
    -0.06
    рії
    -0.06
    POSITIVE LOGITS
     celebrities
    0.07
    eyes
    0.07
    山
    0.06
    虫
    0.06
    Scientists
    0.06
    izont
    0.06
     محمد
    0.06
     Geoff
    0.06
     checkBox
    0.06
     Гри
    0.06
    Activations Density 0.005%

    No Known Activations