© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 38913
    Prev
    Next
    INDEX
    Explanations

    intensity

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of high-intensity exercise jargon (e.g. “intensity,” “HI,” “IT,” “interval”).

    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
     Prot
    -0.07
     Yoshi
    -0.07
     Voldemort
    -0.06
    celik
    -0.06
     Weed
    -0.06
    807
    -0.06
    439
    -0.06
     Grat
    -0.06
    640
    -0.06
    K
    -0.06
    POSITIVE LOGITS
    (dateTime
    0.07
    YG
    0.07
     электр
    0.06
    arah
    0.06
     Assert
    0.06
    ذار
    0.06
    _inter
    0.06
    }else
    0.06
     تاث
    0.06
     система
    0.06
    Activations Density 0.004%

    No Known Activations