© 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. 109810
    Prev
    Next
    INDEX
    Explanations

    gas exchange

    np_max-act · gemini-2.0-flash

    The neuron detects references to gas composition parameters—especially mentions of oxygen or carbon‐dioxide levels or contents.

    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
    uxe
    -0.07
     continuar
    -0.07
    _tv
    -0.06
    Soup
    -0.06
     напри
    -0.06
     processed
    -0.06
    �江
    -0.06
     побач
    -0.06
    UAGE
    -0.06
    عة
    -0.06
    POSITIVE LOGITS
     ``(
    0.06
     překlad
    0.06
     може
    0.06
     Indexed
    0.06
    /moment
    0.06
     ec
    0.06
     microscopic
    0.06
     میل
    0.06
     Sự
    0.06
     ASSIGN
    0.06
    Activations Density 0.218%

    No Known Activations