© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 44261
    Prev
    Next
    INDEX
    Explanations

    choice

    np_max-act · gemini-2.0-flash

    This neuron detects the licensing‐header “choice/option” language (words like “only,” “choice,” “either,” “elects,” etc.) that appears in software license notices.

    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
    .deck
    -0.07
     spam
    -0.07
     انج
    -0.06
     Rak
    -0.06
     acess
    -0.06
    安装
    -0.06
    ぼ
    -0.06
     دهد
    -0.06
    Trail
    -0.06
    _cipher
    -0.06
    POSITIVE LOGITS
    erged
    0.07
     acciones
    0.07
     contrast
    0.07
    cedes
    0.07
    (!$
    0.06
    canf
    0.06
     истории
    0.06
     LatLng
    0.06
    iful
    0.06
    __;
    0.06
    Activations Density 0.001%

    No Known Activations