© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 6997
    Prev
    Next
    INDEX
    Explanations

    Web scraping data

    np_max-act · gemini-2.0-flash

    The neuron activates on descriptive product‐attribute terms (adjectives or nouns) in e-commerce listings—e.g. words denoting scent, material, safety or other key features like “natural,” “organic,” “citrus,” “fresh,” “sensitive skin,” etc.

    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
    currentIndex
    -0.07
     recommended
    -0.06
    PRINTF
    -0.06
     ↔
    -0.06
     što
    -0.06
     volleyball
    -0.06
    NX
    -0.06
     retro
    -0.06
    jdk
    -0.06
    Hom
    -0.06
    POSITIVE LOGITS
     Mum
    0.07
     Comment
    0.07
     Funds
    0.07
    .adj
    0.06
     negotiated
    0.06
    _Size
    0.06
    ']))↵
    0.06
    =='
    0.06
    ента
    0.06
     durante
    0.06
    Activations Density 0.044%

    No Known Activations