© 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. 7885
    Prev
    Next
    INDEX
    Explanations

    paradigm

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the word “paradigm,” especially when it signals a fundamental change (as in “paradigm shift”).

    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
    Russ
    -0.07
    _indx
    -0.07
     Cornwall
    -0.07
    _bot
    -0.06
     fgets
    -0.06
     restless
    -0.06
     tissues
    -0.06
     intensity
    -0.06
    (getString
    -0.06
     exits
    -0.06
    POSITIVE LOGITS
     paradigm
    0.12
     Parad
    0.11
     parad
    0.11
     Pyramid
    0.08
     pyramid
    0.08
     Paige
    0.07
    ADING
    0.07
    ้าพ
    0.07
    par
    0.07
    igm
    0.07
    Activations Density 0.002%

    No Known Activations