© 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. 49942
    Prev
    Next
    INDEX
    Explanations

    clause

    np_max-act · gemini-2.0-flash

    This neuron responds to occurrences of the word “clause” (in any capitalization) in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    legal clauses or language related to wills and estate planning.

    oai_token-act-pair · gpt-4o-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
    telefono
    -0.07
    경기
    -0.07
     /^\
    -0.07
    ">&
    -0.07
     پیشینه
    -0.07
     이것
    -0.07
    ıyordu
    -0.07
     rendition
    -0.07
     firepower
    -0.07
     stør
    -0.07
    POSITIVE LOGITS
    clause
    0.10
     clause
    0.10
     Clause
    0.09
     clauses
    0.07
     clamp
    0.06
    草
    0.06
     clockwise
    0.06
     caul
    0.06
    Conditional
    0.06
     vested
    0.06
    Activations Density 0.003%

    No Known Activations