© 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. 89885
    Prev
    Next
    INDEX
    Explanations

    epsom salts

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of “Epsom salts” (magnesium sulfate) in the text.

    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
    (depth
    -0.06
    frica
    -0.06
     meshes
    -0.06
     mesh
    -0.06
    Moh
    -0.06
     Birds
    -0.06
    ;;
    -0.06
    (import
    -0.05
    oulder
    -0.05
    orris
    -0.05
    POSITIVE LOGITS
    al
    0.07
     olumsuz
    0.07
     [('
    0.07
     Aston
    0.07
    roduce
    0.07
    M
    0.07
     इतन
    0.07
    Film
    0.07
    oloji
    0.07
    ات
    0.07
    Activations Density 0.000%

    No Known Activations