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. 84466
    Prev
    Next
    INDEX
    Explanations

    Names

    np_max-act · gemini-2.0-flash

    This neuron fires on proper nouns, especially personal or place names (e.g. surnames or nisba‐style adjectives).

    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
    .ScrollBars
    -0.06
    odář
    -0.06
     Şu
    -0.06
     �
    -0.06
    >({
    -0.06
    temperature
    -0.06
     Unix
    -0.06
    -filled
    -0.06
     wildlife
    -0.06
     суп
    -0.06
    POSITIVE LOGITS
     adjustments
    0.07
    .Sequential
    0.07
    machine
    0.06
    (confirm
    0.06
     ****************************************************************
    0.06
     clashed
    0.06
    自动
    0.06
     giveaway
    0.06
     cooks
    0.06
     eliminates
    0.06
    Activations Density 0.027%

    No Known Activations