© 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. 6778
    Prev
    Next
    INDEX
    Explanations

    Korean particle 로/으로

    np_max-act · gemini-2.0-flash

    The neuron detects Korean “-로/으로” case-marking particles.

    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
    .visitMethod
    -0.07
    jan
    -0.07
     seventy
    -0.07
     same
    -0.07
    /qt
    -0.06
    IND
    -0.06
    SV
    -0.06
    Articles
    -0.06
    (event
    -0.06
    .pet
    -0.06
    POSITIVE LOGITS
    으로
    0.09
    로서
    0.09
    로
    0.08
     로
    0.08
    수로
    0.07
    ro
    0.07
    드로
    0.07
    _TextChanged
    0.07
     가
    0.07
    �로
    0.06
    Activations Density 0.013%

    No Known Activations