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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 57282
    Prev
    Next
    INDEX
    Explanations

    with

    np_max-act · gemini-2.0-flash

    technical discussions related to digital security and cryptographic practices.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron is specialized to detect numeric tokens representing floating‐point or decimal values.

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

    sentences or phrases that give advice, recommendations, explanations or express evaluative/opinionated language.

    oai_token-act-pair · gpt-5-miniTriggered 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
    ám
    -0.06
    UTF
    -0.06
    xml
    -0.06
    pc
    -0.06
     Peb
    -0.06
    .pg
    -0.06
    ком
    -0.06
    хов
    -0.06
    =admin
    -0.06
     GC
    -0.06
    POSITIVE LOGITS
     Personen
    0.07
    中国
    0.07
    (DEFAULT
    0.06
    _makeConstraints
    0.06
    /terms
    0.06
     인구
    0.06
     Nüfus
    0.06
     ServiceException
    0.06
     علت
    0.06
     entsprech
    0.06
    Activations Density 0.079%

    No Known Activations