© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 42827
    Prev
    Next
    INDEX
    Explanations

    the

    np_max-act · gemini-2.0-flash

    This neuron activates on numeric tokens (especially date and other number strings) in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    mentions of dates, timestamps, and other numeric strings (e.g., years, IDs, IPs/ports) embedded in text or code.

    oai_token-act-pair · gpt-5Triggered by @vetterc0
    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
    ...]↵↵
    -0.08
    ToDate
    -0.07
    _hard
    -0.06
    ΡΑ
    -0.06
     tránh
    -0.06
    mensagem
    -0.06
    birth
    -0.06
    titre
    -0.06
     //////////////////////////////////////////////////////////////////////
    -0.06
    ipa
    -0.06
    POSITIVE LOGITS
    /article
    0.08
     competitive
    0.07
     GE
    0.06
    イク
    0.06
    /GL
    0.06
    ترك
    0.06
     '':↵
    0.06
     دفتر
    0.06
     Mortgage
    0.06
    /us
    0.06
    Activations Density 1.235%

    No Known Activations