© 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. 126773
    Prev
    Next
    INDEX
    Explanations

    injuries and medical conditions

    np_max-act · gemini-2.0-flash

    The neuron fires on mentions of a person’s medical or health status (e.g. “remains in coma,” “is critically ill,” “was operated on,” 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
    .obj
    -0.06
    عدد
    -0.06
    297
    -0.06
    laden
    -0.06
    Interrupt
    -0.06
     paramString
    -0.06
     Favor
    -0.06
    _buf
    -0.06
    hawk
    -0.06
    Safe
    -0.06
    POSITIVE LOGITS
     republican
    0.07
    +S
    0.07
    _Details
    0.07
     appreciate
    0.07
     confined
    0.06
    명의
    0.06
    .initState
    0.06
    =_('
    0.06
    idental
    0.06
    (side
    0.06
    Activations Density 0.041%

    No Known Activations