Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 91806
    Prev
    Next
    INDEX
    Explanations

    agonists

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of β₂-agonist bronchodilator medications (e.g., “short-acting beta-2 agonists,” “bronchodilators,” 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
     hodnoty
    -0.07
    =input
    -0.06
     +'
    -0.06
    Intialized
    -0.06
     territory
    -0.06
    лач
    -0.06
     multimedia
    -0.06
    Notifications
    -0.06
     elapsed
    -0.06
    xe
    -0.06
    POSITIVE LOGITS
     �
    0.07
    ouston
    0.07
     Reply
    0.07
     transformers
    0.06
    isco
    0.06
    seg
    0.06
     Libre
    0.06
    ajan
    0.06
    	delay
    0.06
     дисцип
    0.06
    Activations Density 0.001%

    No Known Activations