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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 130781
    Prev
    Next
    INDEX
    Explanations

    toxic exposure

    np_max-act · gemini-2.0-flash

    The neuron fires on wording that describes how a substance is taken into or affects the body (e.g. “ingested,” “inhaled,” “absorbed through the skin”).

    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
     go
    -0.06
     solids
    -0.06
    われ
    -0.06
    HK
    -0.06
     attacked
    -0.06
     biến
    -0.06
     subtitle
    -0.06
    Magn
    -0.06
    だって
    -0.06
    wort
    -0.06
    POSITIVE LOGITS
    cının
    0.08
    ری
    0.07
     Npgsql
    0.07
     meticulously
    0.06
    ALLERY
    0.06
     özgür
    0.06
    �
    0.06
    					     
    0.06
    .Embed
    0.06
     feminism
    0.06
    Activations Density 0.011%

    No Known Activations