© 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. 13369
    Prev
    Next
    INDEX
    Explanations

    health

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of health and well-being terms (e.g., “salute,” “benessere,” “salud,” “mentale,” etc.).

    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
    äß
    -0.07
    	strncpy
    -0.07
    vais
    -0.07
    peak
    -0.06
    .getSharedPreferences
    -0.06
    ngr
    -0.06
    noopener
    -0.06
    برای
    -0.06
    ่ย
    -0.06
    oltage
    -0.06
    POSITIVE LOGITS
     health
    0.10
     Health
    0.09
     salud
    0.07
     HEALTH
    0.07
     سلام
    0.06
    援
    0.06
     BRAND
    0.06
    health
    0.06
    �
    0.06
     sağ
    0.06
    Activations Density 0.021%

    No Known Activations