© 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. 128850
    Prev
    Next
    INDEX
    Explanations

    headache/nausea

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of adverse events or side‐effect terms in clinical trial reports.

    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
     багать
    -0.07
     microseconds
    -0.07
    every
    -0.07
     POW
    -0.06
    _User
    -0.06
     salida
    -0.06
     Defender
    -0.06
     Uma
    -0.06
     Cole
    -0.06
     Güvenlik
    -0.06
    POSITIVE LOGITS
    Temporary
    0.07
     programu
    0.07
    perhaps
    0.06
     occasions
    0.06
    atsu
    0.06
     варто
    0.06
    	port
    0.06
    organic
    0.06
     belonged
    0.06
    (strings
    0.06
    Activations Density 0.014%

    No Known Activations