© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 7779
    Prev
    Next
    INDEX
    Explanations

    said

    np_max-act · gemini-2.0-flash

    The neuron fires on speech-attribution words (e.g. “said,” “replied,” etc.) indicating dialogue.

    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
    测
    -0.06
     Nationals
    -0.06
    mounted
    -0.06
     трудов
    -0.06
    _invalid
    -0.06
    Solo
    -0.06
    _menu
    -0.06
     wie
    -0.06
    �
    -0.06
     stays
    -0.06
    POSITIVE LOGITS
     aplikace
    0.07
     انگ
    0.07
     setHidden
    0.07
     долго
    0.07
    ORD
    0.06
     Curry
    0.06
    useRalative
    0.06
     والع
    0.06
    ้าท
    0.06
     ανα
    0.06
    Activations Density 0.014%

    No Known Activations