Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 42449
    Prev
    Next
    INDEX
    Explanations

    code quality

    np_max-act · gemini-2.0-flash

    The neuron detects Portuguese words and phrases describing code quality (e.g. “limpo,” “legível,” “robusto”).

    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
    ERROR
    -0.06
    _ACTIVITY
    -0.06
    receipt
    -0.06
    .DecimalField
    -0.06
     Lithuania
    -0.06
    bilder
    -0.06
    ande
    -0.06
    Č
    -0.06
    HeadersHeightSizeMode
    -0.06
    Worksheet
    -0.06
    POSITIVE LOGITS
     craftsmanship
    0.07
    (Py
    0.07
    ventions
    0.06
     Aut
    0.06
     injuring
    0.06
    σσ
    0.06
     aku
    0.06
     приход
    0.06
     Approx
    0.06
     nữ
    0.06
    Activations Density 0.050%

    No Known Activations